Systems for and method of determining tissue chromophore content
A spectrally weighted approach addresses the bias in broadband LED-based systems by accounting for the full spectral characteristics of light sources and detectors, enhancing the accuracy of chromophore quantification and enabling a compact, wearable device for continuous monitoring.
Patent Information
- Application Number
- PCT/US2026/027177
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-05-07
- Filing Date
- 2026-05-07
- Publication Date
- 2026-11-12
AI Technical Summary
Conventional methods for determining chromophore concentrations in biological tissue using broadband light sources introduce systematic bias due to the simplification of broadband LED sources as monochromatic emitters, particularly when dealing with weak or spectrally overlapping absorbers, leading to inaccurate chromophore quantification.
A spectrally weighted approach is introduced to determine chromophore concentrations by accounting for the full spectral characteristics of both light sources and photodetectors, computing modeled detected intensities, and fitting them to measured intensities, thereby reducing systematic bias.
This method improves the accuracy of chromophore concentration determination, especially for weak or spectrally overlapping absorbers, by incorporating the spectral characteristics of both sources and detectors, and is implemented in a compact, wearable spectroscopy device for continuous monitoring.
Smart Images

Figure US2026027177_12112026_PF_FP_ABST
Abstract
Description
Atty. Dkt. No. 125141.05020. MGH2025-380SYSTEMS FOR AND METHOD OF DETERMINING TISSUE CHROMOPHORE CONTENT CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims priority to U. S. Patent Application No. 63 / 801,703 filed May 7, 2025, and entitled, “System for and Method of Determining Tissue Chromophore Content Given a Specific Hardware Configuration,” which is hereby incorporated by reference in its entirety.STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH
[0002] This invention was made with government support under award number AY2AX000007 (agreement 2024A014963) from the Advanced Research Projects Agency for Health and under grant number 1R2 INS 137436-01 awarded by the National Institutes of Health. The government has certain rights in the invention.BACKGROUND
[0003] Optical spectroscopy, such as near-infrared spectroscopy (NIRS), can be used to non-invasively determine concentrations of one or more chromophores within biological tissue.
[0004] Thus, it would be desirable to have improved systems and methods for determining chromophore concentrations in biological tissue using broadband light sources.SUMMARY OF THE DISCLOSURE
[0005] The present disclosure overcomes the aforementioned drawbacks by providing methods and systems for determining chromophore concentrations in a tissue using broadband light sources and a wavelength weighted fitting process. Unlike conventional approaches that approximate each broadband source as a monochromatic emitter at a single peak or centroid wavelength, the disclosed methods can account for the full spectral characteristics of both the light sources and the photodetectors, thereby reducing systematic bias and improving accuracy in recovering chromophore concentrations, particularly for weak or spectrally overlapping absorbers. In some configurations, a wearable spectroscopy device is provided that can integrate one or more multi -wavelength LED packages and a plurality of photodetectors ontoAtty. Dkt. No. 125141.05020. MGH2025-380a single flexible printed circuit board encapsulated in a biocompatible enclosure. Accordingly, the disclosed methods and systems provide an accurate, compact, and cost-effective platform for quantitative chromophore determination in biological tissue, with potential applications including, for example, cerebral oxygenation monitoring, metabolic assessment, muscle oximetry, and other clinical or research uses.
[0006] In accordance with one aspect of the present disclosure, a method for determining one or more chromophore concentrations in a tissue is provided. The method may include illuminating the tissue with a plurality of light sources, wherein each light source having a respective characterized emission spectrum spanning a range of wavelengths. The method may further include detecting, with at least one photodetector having a characterized spectral responsivity, light from each of the plurality of light sources after propagation through the tissue, thereby obtaining a measured detected intensity for each light source. The method may further include for each light source, determining a spectral weighting profile from the characterized emission spectrum of that light source and the characterized spectral responsivity of the at least one photodetector. The method may further include for each light source, computing a modeled detected intensity by summing, over the range of wavelengths, a product of the spectral weighting profile and a tissue attenuation profile. The method may further include determining the one or more chromophore concentrations in the tissue by fitting the modeled detected intensities to the measured detected intensities.
[0007] According to yet another aspect of the present disclosure, a system for determining one or more chromophore concentrations in tissue is provided that may include a plurality of light sources, each having a respective characterized emission spectrum spanning a range of wavelengths, configured to illuminate the tissue. The system may further include at least one photodetector having a characterized spectral responsivity, configured to detect light from each of the plurality of light sources after propagation through the tissue, thereby producing a measured detected intensity for each light source. The system may further include a processor communicatively coupled to the at least one photodetector and configured to for each light source, determine a spectral weighting profile from the characterized emission spectrum of that light source and the characterized spectral responsivity of the at least one photodetector. The processor may be furthered configured to for each light source, compute a modeled detected intensity by summing, over the range of wavelengths, a product of the spectralAtty. Dkt. No. 125141.05020. MGH2025-380weighting profile and a tissue attenuation profile. The processor may be further configured to determine the one or more chromophore concentrations by fitting the modeled detected intensities to the measured detected intensities.
[0008] According to yet another aspect of the disclosure, a wearable spectroscopy device for monitoring chromophore concentrations in a tissue is provided that may include a flexible printed circuit board. The flexible printed circuit board may include a plurality of light sources, each having a respective characterized emission spectrum spanning a range of wavelengths, configured to illuminate the tissue. The flexible printed circuit board may further include at least one photodetector having a characterized spectral responsivity, configured to detect light from each of the plurality of light sources after propagation through the tissue, thereby producing a measured detected intensity for each light source. The wearable spectroscopy device may further include a biocompatible enclosure encapsulating the flexible printed circuit board. The wearable spectroscopy device may further include a processor configured to for each light source, determine a spectral weighting profile from the characterized emission spectrum of that light source and the characterized spectral responsivity of the at least one photodetector. The processor may be further configured to for each light source, compute a modeled detected intensity by summing, over a range of wavelengths, a product of the spectral weighting profile and a tissue attenuation profile. The processor may be further configured to determine the one or more chromophore concentrations by fitting the modeled detected intensities to measured detected intensities.
[0009] The foregoing and other aspects and advantages of the present disclosure will appear from the following description. In the description, reference is made to the accompanying drawings that form a part hereof, and in which there is shown by way of illustration one or more exemplary versions. These versions do not necessarily represent the full scope of the disclosure.BRIEF DESCRIPTION OF THE DRAWINGS
[0010] The following drawings are provided to help illustrate various features of nonlimiting examples of the disclosure, and are not intended to limit the scope of the disclosure or exclude alternative implementations.Atty. Dkt. No. 125141.05020. MGH2025-380
[0011] FIG. 1 is a schematic diagram of a system of determining tissue chromophores, in accordance with some aspects of the disclosure.
[0012] FIG. 2 is a schematic drawing of the system of determining tissue chromophores, in accordance with some aspects of the disclosure.
[0013] FIG. 3 illustrates isometric views of the system of determining tissue chromophores, in accordance with some aspects of the disclosure.
[0014] FIG. 4 is a flow chart of the operation of the system of determining tissue chromophores, in accordance with some aspects of the disclosure.
[0015] FIG. 5 illustrates measured emission spectra of the system of determining tissue chromophores, in accordance with some aspects of the disclosure.
[0016] FIG. 6 illustrates wavelengths measured by the system of determining tissue chromophores, in accordance with some aspects of the disclosure.
[0017] FIG. 7 is a schematic illustration of the spectral weighting and chromophore concentration determination method, in accordance with some aspects of the disclosure.
[0018] FIG. 8 illustrates a comparison of chromophore recovery, in accordance with some aspects of the disclosure.
[0019] FIG. 9 illustrates chromophore recovery results, in accordance with some aspects of the disclosure.
[0020] FIG. 10 illustrates a simulation framework for comparing chromophore recovery, in accordance with some aspects of the disclosure.
[0021] FIG. 11 illustrates a comparison of chromophore recovery accuracy, in accordance with some aspects of the disclosure.
[0022] FIG. 12 illustrates tables of mean absolute error and percent error in chromophore recovery, in accordance with some aspects of the disclosure.
[0023] FIG. 13 illustrates absolute chromophore concentration estimation errors using the spectrally weighted multi-distance method, in accordance with some aspects of the disclosure.
[0024] FIG. 14 illustrates the wearable spectroscopy device positioned on a subject during in vivo monitoring, in accordance with some aspects of the disclosure.
[0025] FIG. 15 is a perspective view of a phantom container used for validation of the system of determining tissue chromophores, in accordance with some aspects of the disclosure.Atty. Dkt. No. 125141.05020. MGH2025-380
[0026] FIG. 16 illustrates system stability measurements of the wearable spectroscopy device over an extended operating duration, in accordance with some aspects of the disclosure.
[0027] FIG. 17 illustrates noise equivalent power measurements for each wavelength of the wearable spectroscopy device, in accordance with some aspects of the disclosure.
[0028] FIG. 18 illustrates recovered effective attenuation coefficients using different calibration approaches, in accordance with some aspects of the disclosure.
[0029] FIG. 19 illustrates correlation and agreement analysis of recovered effective attenuation coefficients, in accordance with some aspects of the disclosure.
[0030] FIG. 20 illustrates measured changes in optical density and recovered cytochrome concentration changes, in accordance with some aspects of the disclosure.
[0031] FIG. 21 is a table of chromophore concentration changes, in accordance with some aspects of the disclosure.
[0032] FIG. 22 illustrates chromophore concentration time traces, in accordance with some aspects of the disclosure.
[0033] FIG. 23 illustrates oxygenation and metabolic monitoring data acquired from a subject using the wearable spectroscopy device, in accordance with some aspects of the disclosure.DETAILED DESCRIPTION OF THE PRESENT DISCLOSURE
[0034] Determining the concentrations of chromophores within turbid media, such as biological tissue, is important across a wide range of clinical and research applications. Nearinfrared spectroscopy (NIRS) is one such optical technique that can exploit the relative transparency of biological tissue in the near-infrared wavelength range to non-invasively probe chromophore concentrations at depth. To reduce cost, size, and power consumption relative to conventional laser-based systems, there has been growing interest in LED-based NIRS systems suitable for wearable and portable use. However, unlike narrowband laser sources, LEDs emit light over a broad spectral bandwidth, and photodetectors used in such systems may exhibit wavelength-dependent responsivity. As a result, the detected signal in an LED-based system represents a spectrally integrated response jointly shaped by the emission spectrum of the source, the responsivity of the detector, and the wavelength-dependent optical properties of the biological tissue.Atty. Dkt. No. 125141.05020. MGH2025-380
[0035] Despite this inherent spectral breadth, conventional approaches to chromophore quantification in LED-based systems have generally approximated each broadband LED source as a monochromatic emitter at a single representative wavelength, such as a peak or centroid wavelength. This simplification disregards the contributions of the full LED emission bandwidth and the detector responsivity profile and can introduce systematic bias in the recovered chromophore concentrations, particularly when the chromophore of interest has a weak or spectrally overlapping absorption signature relative to other chromophores present in the biological tissue. These challenges may be further compounded in systems employing multiple broadband sources, where each source may have a distinct emission profile and bandwidth.
[0036] As LED-based systems have emerged as lower-cost and more compact alternatives to conventional laser-based systems, the need to accurately account for the broad spectral characteristics of LED sources and photodetectors has become increasingly important for reliable chromophore quantification. The present disclosure provides methods and systems that address this need by introducing a spectrally weighted approach to determining chromophore concentrations from optical measurements (e.g., broadband light sources).
[0037] In some configurations, the disclosed methods can determine a spectral weighting profile for each broadband light source from the characterized emission spectrum of that source and the characterized spectral responsivity of at least one photodetector. A modeled detected intensity for each source can be computed by summing, over a range of wavelengths, a product of the spectral weighting profile and a tissue attenuation profile that is dependent on the chromophore concentrations. The chromophore concentrations can then be determined by fitting the modeled detected intensities to measured detected intensities obtained from the at least one photodetector. By incorporating the spectral characteristics of both the sources and the detectors into the determination, the disclosed methods can reduce systematic bias that may otherwise arise from monochromatic approximations, and can improve the accuracy of recovering chromophore concentrations, including those of weak or spectrally overlapping absorbers. In some configurations, the disclosed methods may be implemented in connection with a multi -di stance measurement approach, a single-channel approach, or other measurement configurations, as described further herein.Atty. Dkt. No. 125141.05020. MGH2025-380
[0038] In some configurations, a wearable spectroscopy device is provided that may integrate one or more light source packages (e.g., multi-wavelength LEDs), a plurality of photodetectors, an analog front-end, a wireless communication module, and a processor onto a single flexible printed circuit board encapsulated in a biocompatible enclosure. The integrated design can reduce mechanical and electrical failure points, improve durability, and enable a compact and lightweight form factor suitable for continuous, prolonged monitoring in clinical and non-clinical environments. In some configurations, the device may further include one or more motion sensors, such as an accelerometer, for detecting subject motion and assessing data quality. The processor may be configured to perform the spectrally weighted chromophore determination described herein, and the wireless communication module may be configured to transmit measurement data to an external computing device for real-time display, storage, and analysis.
[0039] In some examples, the systems and methods described herein are described in connection with cerebral oxygenation and metabolic monitoring in neonates (e.g., water, cytochrome c oxidase, etc.), although the disclosed technology is not limited to this application. The disclosed methods and systems may also be applicable to, for example, cerebral monitoring in adults, muscle oxygenation and metabolic assessment, intraoperative tissue perfusion monitoring, functional brain imaging, peripheral vascular monitoring, sports and exercise physiology, monitoring during therapeutic interventions such as therapeutic hypothermia, and other applications in which quantification of chromophore concentrations in biological tissue is desired, or in other scattering media in which broadband optical spectroscopy may be employed.
[0040] While cerebral oxygenation monitoring in neonates is used throughout portions of this disclosure as a non-limiting example, the disclosed methods and systems are not limited to any particular tissue type, patient population, chromophore, or clinical indication. Accordingly, references to specific chromophores, anatomical sites, or clinical applications herein are provided for explanatory purposes and should not be construed as limiting the scope of the present disclosure.
[0041] Referring to FIG. 1, a system 100 is shown for determining one or more chromophore concentrations in a tissue. In some examples, the system 100 may include an optical probe 10, a motion sensor 12, a control module 14, a processor 16, a computing deviceAtty. Dkt. No. 125141.05020. MGH2025-38018, and a temperature sensor 20. In other configurations, one or more of these components may be combined, omitted, rearranged, or implemented in different form factors. For example, the processor 16 may be integrated into the control module 14, or the motion sensor 12 may be omitted in configurations where motion detection is not required. In the illustrated non-limiting configuration, the optical probe 10, the motion sensor 12, the control module 14, and the processor 16 may be co-located within a single device, while the computing device 18 may be a separate external component. In other configurations, some or all of these components may be distributed across separate housings or modules.
[0042] In some configurations, the optical probe 10 may include a plurality of light sources and at least one photodetector configured to illuminate the tissue and detect light after propagation through the tissue. Each light source may have a respective characterized emission spectrum spanning a range of wavelengths. In some configurations, each light source may be a broadband light source, such as a light-emitting diode (LED). In other configurations, the light sources may include, for example, organic LEDs (OLEDs), laser diodes, or other solid-state light emitters. In some examples, the plurality of light sources may be provided as one or more multi -wav elength LED packages, wherein each package may include a plurality of LED dies, and each die may have a distinct peak emission wavelength. In some configurations, a single multi -wavelength LED package may be used. In other configurations, two or more multi -wavelength LED packages may be used to provide redundant measurement channels, increased spatial coverage, and / or the ability to detect and reject measurement artifacts. In some examples, the plurality of light sources may include at least two, at least four, at least six, at least eight, or more individual emitters spanning a range of peak emission wavelengths in the near-infrared range, such as from about 650 nm to about 950 nm, from about 700 nm to about 920 nm, or other ranges as may be suitable for the chromophores of interest. In one nonlimiting example, eight LED dies may have peak emission wavelengths at approximately 735 nm, 760 nm, 780 nm, 810 nm, 830 nm, 850 nm, 870 nm, and 910 nm, although other wavelength combinations may be used.
[0043] In some configurations, the at least one photodetector may have a characterized spectral responsivity and may be configured to detect light from each of the plurality of light sources after propagation through the tissue, thereby producing a measured detected intensity for each light source. In some configurations, the at least one photodetector may include aAtty. Dkt. No. 125141.05020. MGH2025-380photodiode, a phototransistor, or other photosensitive element. In some examples, a plurality of photodetectors may be used, arranged at a plurality of source-detector separations from the light sources to enable multi-distance measurements. In one non-limiting example, four photodetectors may be arranged at source-detector separations of approximately 2.4 cm, 2.8 cm, and 3.3 cm, although other numbers of photodetectors, separation distances, and geometric arrangements may be used. In other configurations, a single photodetector may be used at a single source-detector separation for single-channel measurements. The spectral responsivity of the at least one photodetector may vary across the range of wavelengths emitted by the light sources, and this wavelength-dependent responsivity may be characterized and accounted for during chromophore concentration determination, as described further herein.
[0044] The control module 14 may be operatively coupled to the optical probe 10 and may be configured to drive the plurality of light sources and to acquire signals from the at least one photodetector. In the illustrated non-limiting configuration, the control module 14 may include an analog front-end (AFE) and a wireless communication module. The AFE may be configured to sequentially drive individual light sources and to simultaneously sample one or more photodetectors using one or more internal transimpedance amplifiers. In some examples, the AFE may be capable of driving up to 4, 8, 12, 16, or more individual light sources and simultaneously sampling up to 1, 2, 4, or more photodetectors in parallel. In one non-limiting example, the AFE may drive up to 16 individual LED dies at a drive current of up to about 250 mA and may simultaneously sample up to four photodetectors using four internal transimpedance amplifiers, although other drive current levels and sampling configurations may be used. In some configurations, the AFE may further perform automatic ambient light measurements during one or more acquisition phases in which no light source is active, and the ambient light measurement may be subtracted from active-phase measurements to reduce the influence of background illumination.
[0045] In some configurations, the control module 14 may operate the light sources and photodetectors in a time-multiplexed acquisition sequence, wherein each acquisition phase may correspond to a period during which all photodetectors are sampled in parallel while no more than one light source is active. The acquisition sequence may include a plurality of active phases corresponding to the plurality of light sources, as well as one or more ambient light phases. In one non-limiting example, the acquisition sequence may include 16 active LEDAtty. Dkt. No. 125141.05020. MGH2025-380phases and one ambient light phase, resulting in 17 total acquisition phases per cycle, although other numbers of phases may be used. Each phase may have an integration time of, for example, about 50 ps to about 500 ps, such as about 117 ps, although other integration times may be used. In some configurations, the system 100 may operate at a sampling frequency of about 1 Hz to about 100 Hz, such as about 10 Hz, about 20 Hz, or about 40 Hz, depending on the application and data throughput requirements.
[0046] In some configurations, the wireless communication module of the control module 14 may be configured to transmit measurement data from the system 100 to the computing device 18 or to another external device. In some configurations, the wireless communication module may support Bluetooth Low Energy (BLE), such as BLE 4.0, BLE 5.0, or later versions. In other configurations, the wireless communication module may support other wireless communication protocols, including, for example, Bluetooth Classic, Wi-Fi, Zigbee, near-field communication (NFC), ANT+, or other short-range or long-range wireless protocols. In still other configurations, data communication between the control module 14 and the computing device 18 may be wired, such as via a USB, UART, SPI, or I2C connection. In some examples, the wireless communication module and the processor 16 may be integrated into a single microcontroller unit (MCU).
[0047] In some configurations, the motion sensor 12 may be configured to detect motion of the tissue and / or of the system 100 during measurement. In some configurations, the motion sensor 12 may include a multi-axis accelerometer, such as a 3-axis accelerometer. In other configurations, the motion sensor 12 may include, for example, a gyroscope, a magnetometer, an inertial measurement unit (IMU), or combinations thereof. Motion data acquired by the motion sensor 12 may be used for data quality assessment, motion artifact detection, motion artifact rejection, and / or motion artifact correction. In some examples, the processor 16 or the computing device 18 may be configured to flag or exclude measurement data acquired during periods of detected motion. In other configurations, the motion sensor 12 may be omitted.
[0048] The processor 16 may be configured to receive measurement data from the control module 14 and to determine the one or more chromophore concentrations based on the measured detected intensities. In some configurations, the processor 16 may be configured to, for each light source, determine a spectral weighting profile from the characterized emission spectrum of that light source and the characterized spectral responsivity of the at least oneAtty. Dkt. No. 125141.05020. MGH2025-380photodetector; compute a modeled detected intensity by summing, over the range of wavelengths, a product of the spectral weighting profile and a tissue attenuation profile; and determine the one or more chromophore concentrations by fitting the modeled detected intensities to the measured detected intensities. In some configurations, the tissue attenuation profile may depend on wavelength-dependent optical absorption and scattering properties of the tissue, and the optical absorption may be dependent on the one or more chromophore concentrations. The chromophore concentrations determined by the processor 16 may include, for example, at least one of oxyhemoglobin, deoxyhemoglobin, total hemoglobin, blood oxygen saturation, oxidized cytochrome c oxidase, water (e.g., a water fraction, water concentration, etc.), or other chromophores of interest. In some configurations, all or a portion of the processing may be performed on the processor 16 within the system 100. In other configurations, all or a portion of the processing may be performed on the computing device 18 or on a remote server.
[0049] The processor 16 may include, for example, a microcontroller, a microprocessor, a digital signal processor (DSP), a central processing unit (CPU), a graphics processing unit (GPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or any other digital, analog, or mixed-signal processing component, as well as combinations of the foregoing. The processor 16 may include or be communicatively coupled to a memory configured to store measurement data, characterized emission spectra, characterized spectral responsivity data, operating instructions, and / or other parameters. The memory may include any suitable volatile memory, non-volatile memory, or combination thereof, such as RAM, ROM, EEPROM, flash memory, solid-state storage, or other storage media. In some configurations, the processor 16 may execute operating routines stored in the memory to perform the spectral weighting and chromophore determination described herein.
[0050] In some configurations, the computing device 18 may be configured to receive measurement data and / or chromophore concentration data from the system 100, and may provide a user interface for real-time display, storage, and / or analysis of the data. In some configurations, the computing device 18 may include a mobile device, such as a smartphone or tablet. In other configurations, the computing device 18 may include a laptop computer, a desktop computer, a bedside clinical monitor, a wearable device, a dedicated handheld reader, or a remote server or cloud platform. In some examples, the computing device 18 may executeAtty. Dkt. No. 125141.05020. MGH2025-380a software application configured to display real-time chromophore concentration trends, oxygen saturation values, and / or alerts to a user, such as a clinician, researcher, or patient. In some configurations, the computing device 18 may be configured to generate an alert when a monitored chromophore concentration or derived parameter exceeds or falls below a predetermined threshold. Alerts may include visual indicators, audible indicators, haptic feedback, push notifications, and / or transmission of information to a remote clinician or monitoring service. In some examples, the software application may be configured for a mobile operating system, such as Android or iOS, although other operating systems and platforms may be used.
[0051] In some configurations, the system 100 may support non-invasively, continuous, intermittent, or repeated monitoring over extended durations, such as from minutes to hours to days. In some examples, the system 100 may be powered by a battery, such as a lithium-ion battery, a lithium-polymer battery, a coin cell, or other portable power source. In one nonlimiting example, the system 100 may be powered by a single lithium-ion battery having a capacity of about 600 mAh, providing an operating duration of approximately 10 hours or more at maximum optical power settings, although other battery capacities and operating durations may be achieved depending on power settings and duty cycle. In some configurations, the system 100 may include a rechargeable battery and / or may be powered by an external power source during operation. In some configurations, the system 100 may include a battery connector, such as a magnetic pogo-pin connector, configured to provide a detachable and mechanically robust connection between the battery and the system 100.
[0052] Referring to FIG. 2, an example implementation of the system 100 in accordance with the present disclosure is shown. In the illustrated non-limiting configuration, the system 100 may include an optical probe 10, a control module 116, a power source 112, and a computing device 18. In other configurations, one or more of these components may be combined, separated, rearranged, or implemented in different form factors. For example, the optical probe 10 and the control module 116 may be housed on separate substrates, or the power source 112 may be external to the device.
[0053] In the illustrated non-limiting configuration, the optical probe 10 may be implemented as a flexible probe that can include a plurality of light sources 104 and a plurality of photodetectors 108 arranged in a spatial geometry on a shared substrate. In some examples,Atty. Dkt. No. 125141.05020. MGH2025-380the light sources 104 may include two multi-wavelength LED packages, each containing a plurality of LED dies having distinct peak emission wavelengths. In one non-limiting example, each LED package may contain eight LED dies, although packages with fewer or greater numbers of dies, such as two, four, six, twelve, sixteen, or other numbers, may be used in other configurations. In some examples, each multi-wavelength LED package may have a compact footprint, such as approximately 2.9 mm x 2.2 mm with a height of approximately 0.9 mm, although other package sizes and form factors may be used.
[0054] In the illustrated non-limiting configuration, the optical probe 10 may include four photodetectors 108, although other numbers of photodetectors, such as one, two, three, six, eight, or more, may be used. The photodetectors 108 may be arranged at a plurality of sourcedetector separations from the light sources 104, as indicated by the dashed lines in FIG. 2. In one non-limiting example, source-detector separations of approximately 2.4 cm, 2.8 cm, and 3.3 cm may be used, although other separations, such as from about 0.5 cm to about 5.0 cm, may be selected depending on the tissue type, the depth of interrogation, and the optical properties of the medium being measured. The arrangement of light sources 104 and photodetectors 108 on the optical probe 10 may define a multi-distance measurement geometry in which each light source 104 illuminates the tissue and each photodetector 108 detects light at its respective separation distance. In other configurations, a single photodetector 108 may be used at a single source-detector separation for single-channel measurements. In some examples, each photodetector 108 may be a fast PIN photodiode having a responsivity greater than about 57% across the wavelength range of about 700 nm to about 950 nm, although other photodetector types and responsivity characteristics may be used, as described herein.
[0055] In the illustrated non-limiting configuration, the control module 116 may be operatively coupled to the optical probe 10 and may include an analog front-end (AFE), a wireless communication module, and associated power management circuitry. In the illustrated non-limiting configuration, the AFE is shown as an AFE 4460 integrated circuit, although other analog front-end devices or custom front-end circuitry may be used. The AFE may include an LED driver configured to sequentially drive individual light sources 104, a front-end module including one or more transimpedance amplifiers (TIAs), a multiplexer (MUX), and an analog-to-digital converter (ADC) for converting detected optical signals from the photodetectors 108 into digital data, and a clock module including an internal clock and clock logic for timing theAtty. Dkt. No. 125141.05020. MGH2025-380acquisition sequence. In the illustrated non-limiting example, the front-end module may include four TIAs configured to simultaneously sample four photodetectors 108 in parallel, although other numbers of TIAs and parallel sampling channels may be used. The AFE may further include a serial peripheral interface (SPI) for communicating digital data to a microcontroller unit (MCU).
[0056] In the illustrated non-limiting configuration, the control module 116 may further include a BLE 5 module, as shown in the blue-outlined region of FIG. 2. In the illustrated nonlimiting configuration, the BLE 5 module may include an MCU and a BLE module configured to wirelessly transmit measurement data to the computing device 18. The MCU may be communicatively coupled to the AFE via the SPI interface and may be configured to read out digitized measurement data from the AFE upon completion of each acquisition cycle. In some examples, the MCU may also be communicatively coupled to a motion sensor, such as a 3-axis accelerometer, via an inter-integrated circuit (I²C) bus, as shown in FIG. 2. In other configurations, the MCU may communicate with the motion sensor and / or the AFE via other communication interfaces, such as SPI or other serial or parallel protocols. In some examples, the MCU and the BLE module may be integrated into a single microcontroller unit, such as a system-on-chip (SoC) device. In one non-limiting example, the MCU may support Bluetooth Low Energy 5 communication, enabling higher data transfer rates relative to earlier BLE versions, although other wireless protocols may be used as described herein. Once an acquisition cycle is complete, the MCU may read out the digitized data from the AFE and the motion sensor and transmit the data to the computing device 18 for real-time display, storage, and / or analysis.
[0057] In the illustrated non-limiting configuration, the power source 112 may be configured to supply electrical power to the optical probe 10 and the control module 116. In the illustrated non-limiting configuration, the power source 112 may be a lithium-ion battery, such as a single-cell lithium-ion battery having a nominal voltage of about 3.7 V and a capacity of about 600 mAh, although other battery chemistries, voltages, and capacities may be used. The control module 116 may include power management circuitry, such as a DC / DC converter and a low-dropout regulator (LDO), as shown in FIG. 2, configured to regulate the battery voltage and supply appropriate voltage levels to the AFE, the MCU, and the optical components. In some configurations, the power source 112 may be connected to the systemAtty. Dkt. No. 125141.05020. MGH2025-380100 via a detachable connector, such as a magnetic pogo-pin connector, to minimize the risk of mechanical breakage during setup and handling.
[0058] The computing device 18 may be wirelessly coupled to the control module 116 via the BLE module and may be configured to receive measurement data, display real-time chromophore concentration information, store acquired data, and control operation of the flexible probe 10 (e.g., transmitting instructions thereto, such as, for example, transmitting an characterized emission spectrum for each light source, transmitting a characterized spectral responsivity for each photodetector, etc.).. In the illustrated non-limiting configuration, the computing device 18 may execute a software application, such as the FlexNIRS Application shown in FIG. 2, configured to operate on a mobile operating system. In some examples, the software application may be configured to display one or more of real-time intensity data, chromophore concentration time traces, oxygen saturation trends, motion-flagged data segments, battery status, and / or device connection status. In some configurations, the processor configured to perform the spectral weighting profile determination and chromophore concentration fitting described herein may be implemented within the MCU of the control module 116. In other configurations, the processor may be implemented within the computing device 18, or the processing may be distributed between the control module 116 and the computing device 18.
[0059] In the illustrated non-limiting configuration, the optical probe 10 and the control module 116 may be integrated onto a single flexible printed circuit board (PCB). The single flexible PCB may carry the light sources 104, the photodetectors 108, the AFE, the MCU, the BLE module, the motion sensor, and associated passive components and interconnects. By integrating all optical and electronic components onto a single flexible substrate, the system 100 may reduce the overall number of connectors, cables, and mechanical joints, thereby reducing potential points of mechanical and electrical failure and improving durability during prolonged use. In other configurations, the optical probe 10 and the control module 116 may be implemented on separate substrates, such as separate rigid PCBs, separate flexible PCBs, or a combination of rigid and flexible substrates, connected by cables, flex connectors, or other interconnects.
[0060] In some configurations, the system 100 may further include a temperature sensor 20 configured to measure a temperature of one or more components of the system 100 and / orAtty. Dkt. No. 125141.05020. MGH2025-380of the tissue. The temperature sensor 20 may be positioned proximate to the light sources 104, proximate to the photodetectors 108, on or near the flexible PCB, on or near the tissue-facing surface of the enclosure 120, or at other locations within or on the system 100. In some configurations, the temperature sensor 20 may be a thermistor, a thermocouple, a fiber optic temperature sensor, a digital temperature sensor integrated circuit, an infrared temperature sensor, or other temperature-sensing element. In some examples, multiple temperature sensors 20 may be used at different locations, such as one proximate to the light sources 104 and one proximate to the tissue surface, to capture both device temperature and tissue surface temperature. In other configurations, the temperature sensor 20 may be omitted.
[0061] Because the emission spectra of the light sources 104, the spectral responsivity of the photodetectors 108, and the optical properties of the tissue may each vary with temperature, the temperature sensor 20 may provide temperature data to the processor 16 to improve the accuracy of the chromophore concentration determination. In some configurations, the processor 16 may use the measured temperature to select, from a plurality of pre-characterized emission spectra stored in memory, a characterized emission spectrum for each light source 104 corresponding to the measured temperature, and may use the selected emission spectrum to determine the spectral weighting profile in block 412 of FIG. 4. For example, characterized emission spectra may be pre-measured and stored for a range of operating temperatures, such as from about 15 °C to about 45 °C at intervals of about 1 °C, about 2 °C, about 5 °C, or other intervals, and the processor 16 may select the stored spectrum nearest to the measured temperature, or may interpolate between two or more stored spectra to estimate the emission spectrum at the measured temperature. In some configurations, the processor 16 may similarly use the measured temperature to select or adjust the characterized spectral responsivity of the at least one photodetector 108, because the responsivity of photodetectors such as silicon PIN photodiodes may shift with temperature. In some configurations, the processor 16 may use the measured temperature to adjust the tissue attenuation profile, for example by adjusting the absorption coefficient or the reduced scattering coefficient to account for temperaturedependent changes in tissue optical properties. In some configurations, the processor 16 may use the measured temperature to adjust one or more parameters of the chromophore determination model, such as extinction coefficients, differential pathlength factors, boundary conditions, or other model parameters that may be temperature-dependent. In this manner, theAtty. Dkt. No. 125141.05020. MGH2025-380temperature sensor 20 may enable the system 100 to dynamically adapt the spectral weighting profiles and the tissue attenuation profiles to the current operating conditions, rather than relying on spectral data characterized at a single fixed temperature.
[0062] Referring to FIG. 3, a system 100 in accordance with the present disclosure is shown illustrating the physical construction of the device. An upper view shows the flexible printed circuit board (PCB) prior to encapsulation, and a lower view shows the fully assembled device after encapsulation within a biocompatible enclosure 120. In the illustrated non-limiting configuration, the flexible PCB can form a single, continuous substrate that carries both the optical probe 10 and the control module 116 as interconnected regions of the same board. The optical probe 10 may occupy a first region of the flexible PCB, and the control module 116 may occupy a second region of the flexible PCB, with the two regions joined by one or more narrow flexible bridge sections of the PCB. This integrated single-substrate construction can eliminate cables, connectors, and solder joints between the optical and electronic portions of the device, reducing potential points of mechanical and electrical failure. In some configurations, the flexible PCB may be encapsulated within a biocompatible enclosure 120 to protect the electronic and optical components and to provide a safe, skin-compatible interface for contact with the tissue. In one non-limiting example, the biocompatible enclosure 120 may be formed of a medical-grade silicone elastomer incorporating a light-blocking dye to prevent ambient light from reaching the photodetectors 108. In other examples, the biocompatible enclosure 120 may be formed of other biocompatible materials, such as other medical -grade silicones, polyurethane, thermoplastic elastomers, epoxy, or other encapsulant materials. In some configurations, the flexible PCB may be coated with a conformal coating, such as an acrylic conformal coating, prior to encapsulation to reduce the impact of moisture and perspiration on the electronic components. The resulting device may have a compact form factor, such as approximately 8.2 cm x 3.2 cm x 0.6 cm and a weight of approximately 9.4 g excluding the power source 112, although other dimensions and weights may be achieved depending on the configuration, component selection, and encapsulation materials In some examples, the device may be secured to the tissue surface using a headband, an adhesive patch, a strap, a clip, a hat, headwear, or other wearable attachment mechanism. In one non-limiting example, the device may be placed inside a fabric headband that wraps around the subject's head, and a soft skin-safe silicone tape may be placed over the device to provide additionalAtty. Dkt. No. 125141.05020. MGH2025-380support and reduce motion during monitoring. In some examples, a hydrogel tape or other moisture-barrier layer may be applied between the tissue-facing surface of the enclosure 120 and the skin to prevent moisture from interfering with optical coupling and to reduce the risk of skin irritation during prolonged use. In other configurations, the enclosure 120 itself may include an integrated adhesive layer on the tissue-facing surface for direct attachment to the skin without an external fixture.
[0063] As shown in the upper view of FIG. 3, the optical probe 10 may carry two light source groups 104 and four photodetectors 108 mounted on the flexible PCB. In the illustrated non-limiting configuration, each light source group 104 can be a multi -wavelength LED package containing a plurality of light sources arranged within a single compact housing. Each light source within a light source group 104 may be an LED die having a distinct peak emission wavelength, although other types of light emitters may be used as described herein. Each LED package may have a rectangular footprint, and the individual LED source within each package may be arranged in a two-dimensional array, such as a 2×4 array, although other array configurations, such as 1*8, 2><2, 4x4, or linear arrangements, may be used. In other configurations, each light source group 104 may include a plurality of discrete, individually packaged light sources mounted at separate locations on the optical probe 10 rather than integrated within a single package. In still other configurations, a single light source group 104 containing all individual light sources may be used. The two light source group 104 may be positioned on opposite sides of the optical probe 10, with the four photodetectors 108 arranged between and around the light source group 104. In the illustrated configuration, the colored arrows indicate different source-detector separations between each light source group 104 and respective photodetectors 108. The spatial arrangement of the light source group 104 and photodetectors 108 on the optical probe 10 can define a multi-distance measurement geometry in which each light source 104 can illuminate the tissue and each photodetector 108 can detect light that has propagated through the tissue at a respective source-detector separation. In other configurations, the optical probe 10 may carry one, three, four, or more light source group 104 and one, two, three, six, eight, or more photodetectors 108 in various geometric arrangements, such as linear, cross-symmetric, rectangular, circular, or other layouts.
[0064] In the illustrated non-limiting configuration, each light source group 104 may be surface-mounted to the flexible PCB using standard surface-mount technology (SMT)Atty. Dkt. No. 125141.05020. MGH2025-380processes, such as reflow soldering, although other mounting techniques, such as wire bonding, flip-chip bonding, conductive adhesive attachment, or press-fit mounting, may be used. Similarly, each photodetector 108 may be surface-mounted to the flexible PCB. Electrical traces on the flexible PCB may connect the light source group 104 and the photodetectors 108 to the control module 116, providing drive current to the light source group 104 and carrying detected signals from the photodetectors 108 to the analog front-end within the control module 116. The flexible PCB may be fabricated from a polyimide substrate, such as Kapton, although other flexible substrate materials, such as polyester (PET), polyethylene naphthalate (PEN), liquid crystal polymer (LCP), or other flexible dielectric materials, may be used. The conductive traces may be formed of copper, gold, silver, or other conductive materials, and may include one or more conductive layers separated by dielectric layers to form a multi-layer flexible circuit.
[0065] In the illustrated non-limiting configuration, the control module 116, visible in the upper view of FIG. 3 as a second region of the flexible PCB, may carry the analog front-end, the microcontroller unit, the wireless communication module, the motion sensor, power management circuitry, and associated passive components such as resistors, capacitors, and inductors. In the illustrated non-limiting configuration, the components of the control module 116 can be surface-mounted on the flexible PCB, and the various integrated circuits and discrete components can be interconnected by conductive traces and vias within the flexible PCB. In other configurations, some or all of the control module components may be mounted on a rigid portion or a rigid-flex portion of the PCB to provide mechanical stability for fine-pitch integrated circuit packages, while the optical probe region remains flexible to conform to curved tissue surfaces.
[0066] As shown in the lower view of FIG. 3, the fully assembled device includes the flexible PCB encapsulated within the biocompatible enclosure 120. The enclosure 120 may surround and protect the flexible PCB, the light source group 104, the photodetectors 108, and the control module 116, while leaving the optical surfaces of the light source group 104 and the photodetectors 108 exposed or flush with the tissue-facing surface of the enclosure 120 to permit optical coupling with the tissue. In the illustrated non-limiting configuration, the enclosure 120 may be molded directly around the coated flexible PCB. In one non-limiting example, the enclosure 120 may be silicone elastomer such as Elkem RTV-4420 or anAtty. Dkt. No. 125141.05020. MGH2025-380equivalent medical-grade formulation, although other silicone elastomers, biocompatible polymers, or encapsulant materials may be used. The enclosure 120 may incorporate a lightblocking dye, such as a medical-grade carbon black dye, dispersed throughout the silicone to render the enclosure 120 opaque to ambient light and prevent stray light from reaching the photodetectors 108 through the enclosure walls. In one non-limiting example, the dye may be FI-205 or an equivalent medical-grade pigment, although other light-blocking additives, such as other carbon-based pigments, iron oxide pigments, or opaque fillers, may be used. In other configurations, ambient light rejection may be achieved by other means, such as opaque coatings, light-blocking films, metallic shielding layers, or mechanical baffles integrated into the enclosure 120.
[0067] In the illustrated non-limiting configuration, prior to encapsulation within the enclosure 120, the flexible PCB may be coated with a conformal coating to provide moisture and chemical resistance. In one non-limiting example, the conformal coating may be an acrylic-based conformal coating, although other conformal coatings, such as silicone-based, urethane-based, epoxy-based, or Parylene coatings, may be used. The conformal coating may protect the conductive traces, solder joints, and electronic components from moisture, perspiration, and bodily fluids that may be present during use on or near biological tissue.
[0068] In the illustrated non-limiting configuration, the enclosure 120 may define the overall external form factor of the device. In the illustrated non-limiting configuration, the enclosure 120 may have a generally elongated shape with a narrowed central bridge region corresponding to the flexible bridge section of the underlying PCB, and wider end regions corresponding to the optical probe 10 and the control module 116. This shape may allow the device to flex and conform to curved tissue surfaces, such as the forehead of a neonate or adult. The enclosure 120 may have a tissue-facing surface that is substantially smooth and planar, or may have a contoured tissue-facing surface shaped to conform to a particular anatomical site. In some configurations, the tissue-facing surface may include one or more recessed regions or standoff features to maintain a controlled spacing between the optical components and the tissue surface.
[0069] Referring to FIG. 4, a method 400 is shown for determining one or more chromophore concentrations in a tissue. In some examples, all or part of the method 400 may be performed by the system 100 of FIGS. 1-3, including the optical probe 10, the controlAtty. Dkt. No. 125141.05020. MGH2025-380module 116, the processor 16, and / or the computing device 18. In other configurations, the method 400 may be performed by other systems or devices that include a plurality of light sources, at least one photodetector, and a processor configured to perform the spectral weighting and fitting operations described herein.
[0070] In block 404, the tissue may be illuminated with a plurality of light sources 104, each light source having a respective characterized emission spectrum spanning a range of wavelengths. In some configurations, each light source 104 may be a broadband light source, such as a light-emitting diode (LED), that emits light over a spectral bandwidth having a full width at half maximum (FWHM) of, for example, about 20 nm to about 60 nm or more. In some cases, each respective characterized emission spectrum of each light source can have a first intensity less than the peak wavelength intensity of characterized emission spectrum and a second intensity greater than the peak wavelength intensity of the characterized emission spectrum.
[0071] In some configurations, the characterized emission spectrum for each light source 104 may be obtained by measuring the spectral output of that light source under controlled conditions using a spectrometer or other spectral measurement instrument. Because LED emission spectra can shift with temperature, the characterized emission spectrum may be measured at a predetermined operating temperature that approximates real-use conditions. In one non-limiting example, the emission spectra may be measured at approximately 38 °C to approximate conditions when the device is in contact with skin, and / or at approximately 19 °C or other temperatures for laboratory or phantom validation purposes. In some cases, therefore, a system (e.g., the system 100, an NIRS device, the flexible probe 10, etc.) can include a temperature sensor to, for example, rej ect data outside a desired temperature range, or to obtain a temperate specific emission spectra. In some configurations, the emission spectra may be measured across a range of drive currents to verify spectral stability with current. In some examples, each measured emission spectrum may be normalized to its peak intensity to facilitate comparison across light sources. In other cases, each characterized emission spectrum can be obtained from a data sheet (e.g., from the manufacturer of the light source, such as light source package, etc.). Accordingly, a computing device can transmit each characterized emission spectrum for each light source to a system (e.g., the system 100, an NIRS device, etc.).Atty. Dkt. No. 125141.05020. MGH2025-380
[0072] In block 408, light from each of the plurality of light sources 104 may be detected by at least one photodetector 108 after propagation through the tissue, thereby obtaining a measured detected intensity for each light source 104. The at least one photodetector 108 may have a characterized spectral responsivity that describes how the sensitivity of the photodetector 108 varies as a wavelength changes across the emission range of the light sources 104. Because the photodetector 108 may not have uniform sensitivity across all wavelengths, the characterized spectral responsivity may weight the detected signal differently at different wavelengths. For example, a silicon PIN photodiode may have a responsivity that peaks at a wavelength of about 800 nm to about 850 nm and decreases toward both shorter and longer wavelengths within the near-infrared range.
[0073] In some configurations, the characterized spectral responsivity may be obtained by measuring the photodetector response at a plurality of known wavelengths, such as using one or more narrowband laser sources at known wavelengths, and combining the measured data with manufacturer-reported responsivity values at additional wavelengths. In some configurations, the spectral responsivity may be measured at a predetermined operating temperature, such as approximately 38 °C, because photodetector responsivity can also vary with temperature. In one non-limiting example, the responsivity may be measured over a temperature range of about 20 °C to about 38 °C and interpolated to estimate the responsivity at the target operating temperature across all relevant wavelengths.
[0074] In some configurations, the at least one photodetector 108 may include a plurality of photodetectors positioned at different source-detector separations from each light source 104. In such configurations, the measured detected intensity for each light source 104 may include intensities detected at the different separations. This multi-distance arrangement can enable estimation of an effective attenuation coefficient for each light source, as described further below (e.g., and can provide a self-calibrating configuration). In other configurations, a single photodetector 108 may be used at a single source-detector separation, and the measured detected intensity for each light source 104 may be obtained at a plurality of time points to enable a single-channel change-in-optical-density approach.
[0075] In some configurations, the light sources 104 may be driven sequentially in a time-multiplexed manner, such that during each acquisition phase no more than one light source may be active while all photodetectors 108 can be sampled in parallel. In this manner, theAtty. Dkt. No. 125141.05020. MGH2025-380measured detected intensity for each light source 104 at each photodetector 108 may be acquired without optical crosstalk between light sources. In some examples, one or more ambient light acquisition phases may be interleaved with the active-source phases, and the ambient light signal may be subtracted from the active-phase signals to reduce the influence of background illumination.
[0076] In block 412, for each light source 104, a spectral weighting profile may be determined from the characterized emission spectrum of that light source, the characterized spectral responsivity of the at least one photodetector 108, etc. The spectral weighting profile may represent the combined spectral characteristics of the source-detector pair and describes how the optical measurement channel can weigh the contribution of each wavelength within the emission range. In some configurations, the spectral weighting profile for a given light source-photodetector pair may be determined as a product of the characterized emission spectrum of that light source and the characterized spectral responsivity of the photodetector at each wavelength across the emission range.
[0077] Denoting the characterized emission spectrum of a given light source as ILED(A)and the characterized spectral responsivity of the photodetector as( ), the spectral weighting profile W7(A)for that source-detector pair may be expressed as:
[0078] VF(A) = ILED(X) • RPDX)
[0079] The spectral weighting profile W (A) may be a wavelength-dependent quantity that is non-zero over the emission bandwidth of the light source and that accounts for the combined effect of the source emission shape and the detector sensitivity shape. In some examples, the spectral weighting profile may be precomputed and stored in a memory accessible to the processor 16 prior to the measurement, or may be computed in real time during the measurement. In some configurations, the spectral weighting profile may be updated if the operating temperature or drive current changes, to account for any resulting spectral shifts.
[0080] In block 416, for each light source 104, a modeled detected intensity may be computed, based on, the characterized emission spectrum of the light source 104, the characterized spectral responsivity of the photodetector, and the tissue attenuation profile (e.g., the spectral tissue attenuation profile), etc. For example, for each light source 104, the modeled detected intensity may be computed by summing, over the range of wavelengths, a product of the spectral weighting profile and a tissue attenuation profile. The tissue attenuation profileAtty. Dkt. No. 125141.05020. MGH2025-380may describe the wavelength-dependent attenuation of light as it propagates through the tissue, and may depend on the optical absorption and scattering properties of the tissue. The optical absorption may be dependent on the one or more chromophore concentrations that are to be determined.
[0081] In some configurations, the optical absorption coefficient pa( ) of the tissue at a given wavelength A may be expressed as a sum of contributions from each chromophore present in the tissue:
[0083] where e(-(A)can be the wavelength-dependent extinction coefficient of the i -th chromophore and can be the concentration of the i -th chromophore. In some configurations, the extinction coefficients f;-(A) may be obtained from from independent calibration measurements.
[0084] In some configurations, the reduced scattering coefficient ^s( )of the tissue may be modeled using an empirical power law:
[0086] where a can be a scattering amplitude, b can be a scattering power, and Aocan be a reference wavelength. In other configurations, the reduced scattering coefficient may be modeled using other parametric or non-parametric representations, may be measured directly using a frequency-domain or time-domain optical measurement, or may be estimated from literature values appropriate for the tissue type and patient population of interest.
[0087] In some configurations, using the optical absorption coefficient / ia(A)and the reduced scattering coefficient ( ), a tissue attenuation profile A(A,p) may be computed at each wavelength A and for each source-detector separation p. In some configurations, the tissue attenuation profile may be derived from a solution to the photon diffusion equation for a semi-infinite homogeneous geometry. In one non-limiting example, the tissue attenuation profile may be expressed as:100881 [expJL
[0089] where:Atty. Dkt. No. 125141.05020. MGH2025-380
[0095] In the above expressions, ryand r2may represent distances from a real source and an image source, respectively, to the detection point; z0may be the transport mean free path; zbcan be the extrapolated boundary distance, which may depend on the refractive index n of the tissue (e.g., n~ 1.37 for biological tissue, although other values may be used); and K may be an attenuation term. In other configurations, the tissue attenuation profile may be derived from other light transport models, such as a multi-layered tissue model, a finite-element solution, or other analytical or numerical approaches. The tissue attenuation profile is not limited to the model described above.
[0096] The modeled detected intensity ILED P) foragiven light source at a given sourcedetector separation may then be computed by summing, over the range of wavelengths, the product of the spectral weighting profile V7 ( ) and the tissue attenuation profile A (A, p):
[0098] or equivalently, in a continuous formulation:
[0100] where Aminand Amaxdefine the wavelength range over which the spectral weighting profile can be non-negligible, and AA can be the wavelength step size used in the discrete summation. In this manner, the modeled detected intensity can account for the full spectral breadth of the light source emission, the wavelength-dependent sensitivity of the photodetector, and the wavelength-dependent optical properties of the tissue, rather than approximating the light source as a monochromatic emitter at a single peak or centroid wavelength. The modeled detected intensity can be a predicted value that depends on the assumed chromophore concentrations through the absorption coefficient pa(A) within the tissue attenuation profile A(A, p).
[0101] In some configurations, the wavelength range may extend from about 650 nm to about 950 nm, from about 700 nm to about 920 nm, or over other ranges appropriate for theAtty. Dkt. No. 125141.05020. MGH2025-380light sources and chromophores of interest. The wavelength step size AT may be, for example, about 1 nm, about 2 nm, about 5 nm, or other values, depending on the spectral resolution of the characterized emission spectra and the desired computational accuracy.
[0102] In block 420, the one or more chromophore concentrations may be determined by fitting the modeled detected intensities to the measured detected intensities. In some configurations, a linear optimization model can be used to determine the chromophore concentrations. The fitting may be performed across the plurality of light sources 104 such that the chromophore concentrations that best account for the measured data across all source wavelengths are determined simultaneously. Because the modeled detected intensities can depend on the chromophore concentrations through the tissue attenuation profile, the fitting may involve iteratively adjusting the chromophore concentration values and recomputing the modeled detected intensities until the modeled values sufficiently approximate the measured values.
[0103] The fitting in block 420 may be implemented according to at least two approaches: a multi-distance approach and a single-channel change-in-optical-density approach, as described in the following paragraphs. In some configurations, other fitting approaches may additionally or alternatively be used.
[0104] In some configurations, the at least one photodetector 108 may include a plurality of photodetectors positioned at different source-detector separations from each light source 104, and the fitting may be performed using a multi-distance approach. In the multi -di stance approach, for each light source 104, a measured effective attenuation coefficient may be derived from the measured detected intensities at the different separations. The effective attenuation coefficienteff may characterize how rapidly light intensity decays with increasing source-detector separation in the tissue, and can be related to both the absorption and scattering properties of the tissue.
[0105] In some configurations, the measured effective attenuation coefficient for a given light source may be derived from a linear relationship between the log-scaled measured detected intensities and the source-detector separations. In one non-limiting example, the measured effective attenuation coefficient may be obtained from:Atty. Dkt. No. 125141.05020. MGH2025-380
[0107] where p can be the source-detector separation, I (p) can be the measured detected intensity at separation p, peff can be the effective attenuation coefficient, and / 0(p = 0) can be an intercept term. The measured effective attenuation coefficient may be determined as the negative slope of a line fitted to the data points In [p2■ / (p)] plotted against p across the plurality of source-detector separations. This procedure may be repeated for each light source 104, yielding a set of N measured effective attenuation coefficients, where N can be the total number of light sources.
[0108] Correspondingly, for each light source 104, a modeled effective attenuation coefficient pey may be derived from the modeled detected intensities at the same sourcedetector separations using the same linear relationship:
[0110] where ILED P) can be the modeled detected intensity computed in block 416 at separation p, and peyy can be the modeled effective attenuation coefficient. The modeled effective attenuation coefficient can depend on the assumed chromophore concentrations through the tissue attenuation profile.
[0111] The chromophore concentrations may then be determined by fitting the modeled effective attenuation coefficients to the measured effective attenuation coefficients across all light sources. In some configurations, the fitting may involve minimizing a loss defined as a sum of squared differences between the modeled effective attenuation coefficient and the measured effective attenuation coefficient for each light source:
[0113] where n can be the index of each light source, N can be the total number of light sources,eff(n) can be the modeled effective attenuation coefficient for the n-th light source, and Pe / / -(n)canbe the corresponding measured effective attenuation coefficient.
[0114] In some configurations, the minimization may be performed using a non-linear least-squares optimization process, such as a gradient descent algorithm or other numerical optimization methods. In other configurations, such as when the relationship between the modeled quantities and the chromophore concentrations is substantially linear, the fitting may be performed using a linear optimization. In some configurations, the optimization may includeAtty. Dkt. No. 125141.05020. MGH2025-380bounds or constraints on the chromophore concentration values, such as non-negativity constraints or physiologically plausible ranges.
[0115] In some configurations, the multi-distance approach may further include a calibration procedure in which a reference measurement is performed on a medium with known optical properties, such as a tissue-simulating phantom, to establish a relationship between measured intensities and absolute optical properties. The calibration may correct for systemspecific factors, such as differences in LED optical power, photodetector gain, and optical coupling efficiency, that may affect the measured intensities but are not related to the tissue optical properties.
[0116] In some configurations, the measured detected intensity for each light source 104 may be obtained at a single source-detector separation at a plurality of time points, and the fitting may be performed using a single-channel change-in-optical-density approach. This approach may be used, for example, when a multi-distance measurement may not be feasible due to geometric constraints, such as when monitoring on small anatomical sites, or when the device geometry may not support multiple source-detector separations for a given light source.
[0117] In the single-channel approach, for each light source 104, a measured change in optical density AO may be derived from the measured detected intensity at a current time point relative to a baseline measured detected intensity:
[0119] where ILEDcan be the measured detected intensity at a current time point t,can be the baseline measured detected intensity at a reference time t0, p can be the sourcedetector separation, DPF(,p) can be a wavelength-dependent differential pathlength factor, and A^a(A) can be the change in the absorption coefficient relative to baseline. The change in absorption coefficient may in turn be expressed as:
[0120] A / za(A) = Si£i ) - AQ
[0121] where A is the change in concentration of the i-th chromophore relative to the baseline.
[0122] The differential pathlength factor DPF(, p~) may account for the increased optical path traveled by photons due to scattering in the tissue and may be expressed as:Atty. Dkt. No. 125141.05020. MGH2025-380
[0124] In some configurations, the DPF may be computed using baseline optical properties of the tissue, which may be obtained from reference measurements, from prior multi -di stance measurements, or from literature values appropriate for the tissue type, anatomical site, and patient population. In one non-limiting example, DPF values for neonatal cerebral tissue may be interpolated from published optical property data for healthy term neonates. In other configurations, the DPF may be estimated using other models, empirical formulas, or lookup tables.
[0125] In the single-channel approach, a modeled change in optical density may be derived from the modeled detected intensities incorporating the differential pathlength factor, and the chromophore concentrations (or changes in chromophore concentrations) may be determined by fitting the modeled changes in optical density to the measured changes in optical density across the plurality of light sources. The spectral weighting profile determined in block 412 may be incorporated into the modeled change in optical density to account for the broadband nature of the light sources, in the same manner as described for the multi-distance approach.
[0126] In some configurations, the baseline measured detected intensity lE°Dmay be periodically updated during monitoring. For example, the baseline may be redefined at intervals of about 10 minutes, about 30 minutes, about 60 minutes, or other intervals, to reduce the influence of slow instrumental drifts, changes in optical coupling, and / or thermal variations over long recording durations. In other configurations, the baseline may be updated adaptively based on signal stability criteria, or may be defined as a running average over a preceding time window.
[0127] In some configurations, the fitting in block 420 may be performed subject to one or more constraints on the chromophore concentrations. For example, the fitting may be performed subject to a constraint that a total concentration of a subset of the chromophores remains constant, such as a constraint that the change in total hemoglobin concentration ( HbT = HbO + HbR) is equal to zero. Such constraints may reduce parameter crosstalk between chromophores having spectrally overlapping absorption features, such as betweenAtty. Dkt. No. 125141.05020. MGH2025-380hemoglobin and oxidized cytochrome c oxidase, and may improve the stability and reliability of the recovered concentrations. In other configurations, the fitting may be unconstrained.
[0128] In some configurations, the fitting may be performed using a subset of the plurality of light sources. For example, the fitting may be restricted to light sources having peak emission wavelengths above a threshold wavelength, such as above about 780 nm, to enhance sensitivity to particular chromophores such as oxidized cytochrome c oxidase. In other configurations, the fitting may be performed using all available light sources across the full wavelength range to maximize spectral information and reduce the risk of model degeneracy.
[0129] In some other configurations, the method 400 may be implemented using an approach in which the spectral weighting profile determined in block 412 can be used to adjust or correct the chromophore concentrations. In such configurations, the detected intensities obtained in block 408 may be provided as inputs to a spectral correction model that accounts for the characterized emission spectra of the light sources and the characterized spectral responsivity of the at least one photodetector, and the spectral correction model may output adjusted chromophore concentrations. The spectral correction model may take various forms. In some examples, the spectral correction model may be a pre-computed lookup table that maps detected intensities, or quantities derived therefrom such as intensity ratios, changes in optical density, or attenuation values, to chromophore concentrations that have been corrected for the spectral characteristics of the source-detector pairs. In some examples, the spectral correction model may be a set of wavelength-dependent correction factors, derived from the spectral weighting profiles, that are applied to chromophore concentrations to yield spectrally adjusted chromophore concentrations. In some examples, the spectral correction model may be a regression model, a polynomial mapping, a neural network, or other trained model that has been calibrated using data generated from the spectral weighting profiles and known or simulated tissue optical properties, and that receives detected intensities or derived optical quantities as input and outputs adjusted chromophore concentrations. In some examples, the spectral correction model may be a concentration curve or calibration curve that relates detected intensities at each light source to expected chromophore concentrations, where the concentration curve has been generated using the spectral weighting profiles to account for the broadband nature of the light sources and the wavelength-dependent responsivity of the photodetectors. In this manner, the spectral weighting profiles may be incorporated into theAtty. Dkt. No. 125141.05020. MGH2025-380chromophore determination without requiring iterative computation of modeled intensities and without requiring real-time evaluation of a tissue attenuation profde at each wavelength, which may reduce computational complexity and enable implementation on resource-constrained processors.
[0130] After block 420, the method 400 may return to block 404 such that additional illumination, detection, and concentration determination cycles may be performed. This iterative process may support continuous, intermittent, or repeated monitoring over extended durations, such as from minutes to hours to days. In some configurations, the determined chromophore concentrations and / or derived parameters may be transmitted to the computing device 18 for real-time display, storage, trend analysis, and / or alert generation.
[0131] Referring to FIG. 5, the characterized emission spectra and the characterized spectral responsivity used to determine the spectral weighting profiles described in connection with block 412 of FIG. 4 are illustrated for one non-limiting example implementation of the system 100. FIG. 5 shows the normalized emission spectra (solid lines) for each of eight LEDs in a multi -wavelength LED package, with manufacturer-reported peak emission wavelengths of approximately 735 nm, 760 nm, 780 nm, 810 nm, 830 nm, 850 nm, 870 nm, and 910 nm, measured at an operating temperature of approximately 38 °C using a spectrometer. FIG. 5 also shows the normalized spectral responsivity of the photodetector (dotted line, right axis), which in this example is a silicon PIN photodiode. As shown in FIG. 5, the photodetector responsivity may not be uniform across the near-infrared range; rather, it may peak at approximately 800 nm to 850 nm and decreases toward both shorter and longer wavelengths. As a result, the photodetector may weight the detected contribution from wavelengths near the center of the near-infrared range more heavily than contributions from the spectral extremes.
[0132] FIG. 5 further shows the effective emission spectra (dashed lines) obtained when the characterized emission spectrum of each LED is combined with the characterized spectral responsivity of the photodetector. These effective emission spectra represent the spectral weighting profiles described in connection with block 412 of FIG. 4, for example, the product IV (A) = ILED( • RPD(A) for each light source-photodetector pair. As can be observed from FIG. 5, the spectral weighting profiles can differ from the LED-only emission spectra in several respects. For LEDs at the shorter- wavelength end of the range (e.g., 735 nm, 760 nm), the effective emission may be shifted toward longer wavelengths relative to the LED-onlyAtty. Dkt. No. 125141.05020. MGH2025-380emission, because the photodetector responsivity may increase toward longer wavelengths in this region. Conversely, for LEDs at the longer-wavelength end of the range (e.g., 870 nm, 910 nm), the effective emission may be shifted toward shorter wavelengths, because the photodetector responsivity may decrease at longer wavelengths. These spectral shifts illustrate why a monochromatic approximation based solely on the LED peak or centroid wavelength may introduce systematic bias in the recovered chromophore concentrations: the wavelength at which the measurement channel is effectively most sensitive may differ from the manufacturer-reported peak wavelength by several nanometers or more, and the shape of the effective spectral response may be asymmetric or broadened relative to the LED-only emission. By explicitly characterizing and incorporating these spectral weighting profiles into the modeled detected intensity, the method 400 can account for such effects and reduce the resulting bias.
[0133] Referring to FIG. 6, a table summarizes the measured spectral parameters for each of the eight LEDs at two operating temperatures: approximately 19 °C and approximately 38 °C. For each LED, the table reports the manufacturer-reported peak wavelength, the measured peak wavelength, the centroid wavelength, and the full width at half maximum (FWHM) of the emission spectrum. Values in parentheses indicate the shift in each parameter when the photodetector responsivity is accounted for, for example, the difference between the parameter computed from the LED-only emission spectrum and the parameter computed from the spectral weighting profile (e g., LED emission multiplied by photodetector responsivity).
[0134] As shown in FIG. 6, the measured peak wavelengths may differ from the manufacturer-reported values by several nanometers, and these differences may vary with operating temperature. For example, when the operating temperature increases from approximately 19 °C to approximately 38 °C, peak wavelength may shift of up to approximately 16 nm and centroid wavelength may shift of up to approximately 11 nm may be observed. In addition, the FWHM of the emission spectra may generally increase at higher temperatures, with values ranging from approximately 20 nm to approximately 59 nm at 19 °C and from approximately 25 nm to approximately 56 nm at 38 °C. These temperature-dependent spectral shifts and bandwidth variations further illustrate why the characterized emission spectra used to determine the spectral weighting profiles may be measured at a predeterminedAtty. Dkt. No. 125141.05020. MGH2025-380operating temperature that approximates real-use conditions, as described in connection with block 412 of FIG. 4.
[0135] As further shown in FIG. 6, when the photodetector responsivity is accounted for, additional shifts in peak wavelength, centroid wavelength, and FWHM may be observed, as indicated by the parenthetical values. For some LEDs, the centroid wavelength may shift by up to approximately 3 nm and the FWHM may change by up to approximately 9 nm when the photodetector responsivity is included. These shifts may generally be larger for LEDs at the spectral extremes (e.g., 735 nm and 910 nm), where the photodetector responsivity varies most steeply across the emission bandwidth, and may be smaller for LEDs near the center of the responsivity curve (e.g., 810 nm, 830 nm). These observations confirm that both the light source emission characteristics and the photodetector responsivity characteristics can contribute meaningful spectral shifts and bandwidth changes to the effective measurement channel, and that accounting for both in the spectral weighting profile, as performed in block 412 of FIG. 4, may improve the accuracy of the chromophore concentration determination relative to approaches that rely on manufacturer-reported peak wavelengths or centroid approximations alone.
[0136] Referring to FIG. 7, a schematic illustration of the spectral weighting and chromophore concentration determination method described in connection with blocks 412-420 of FIG. 4 is shown. On the left side, FIG. 7 illustrates a set of model parameters that may be used to compute the tissue attenuation profile, including the optical absorption coefficient / / a(A), the reduced scattering coefficient ^((2), the index of refraction n, the free mean transport path ltr, and the source-detector separation p. The absorption coefficient may be dependent on the concentrations of one or more chromophores, such as oxyhemoglobin (HbO), deoxyhemoglobin (HbR), and water (H₂O), as shown in the lower-left panel. On the right side, FIG. 7 illustrates the calibration measurement of the LED emission spectra and the photodetector responsivity used to determine the spectral weighting profiles, as described in connection with block 412 of FIG. 4. The center equation illustrates the computation of the modeled intensity i^ED^y summing, over the range of wavelengths, the product of the LED emission spectrum 7( ), the photodetector responsivity 7?(A), and the tissue attenuation profile derived from a solution to the photon diffusion equation. The right-side panels illustrate the resulting effective attenuation coefficient and the modeled intensity estimate for each lightAtty. Dkt. No. 125141.05020. MGH2025-380source. In this manner, FIG. 7 illustrates the data flow from characterized spectral profdes and assumed chromophore concentrations through the forward model to the modeled intensity, which may then be compared to the detected intensity to determine the chromophore concentrations as described in connection with block 420 of FIG. 4.
[0137] Referring to FIG. 8, validation results are shown comparing chromophore recovery during an in vitro blood deoxygenation experiment. In this experiment, a solution containing bovine blood was deoxygenated by bubbling nitrogen gas, reducing the oxygen content and converting oxyhemoglobin to deoxyhemoglobin. FIG. 8 (left panel) shows the recovered changes in oxyhemoglobin (AHbO) and deoxyhemoglobin (AHb) over time using a conventional nominal wavelength approach, in which the extinction coefficients of oxy- and deoxyhemoglobin are assumed to be at the manufacturer-reported peak wavelengths of the LEDs. FIG. 8 (right panel) shows the recovered changes using the spectrally weighted approach described in connection with FIG. 4, in which the spectral weighting profiles can account for the full emission bandwidth of each LED and the photodetector responsivity. Dotted lines represent reference measurements obtained using a frequency-domain NIRS (FD-NIRS) system, and solid lines represent measurements obtained using the LED-based system. As shown, the spectrally weighted approach of FIG. 8 (right panel) may yield a AHbO error of approximately 0.53 pM (approximately 8.0% error), compared to an error of approximately 2.25 pM (approximately 33.5% error) using the nominal wavelength approach of FIG. 8 (left panel), representing an approximately four-fold reduction in recovery error.
[0138] Referring to FIG. 9, additional validation results are shown demonstrating the performance of the spectrally weighted approach across multiple device configurations and experimental conditions. The upper-left panel of FIG. 9 shows a hemoglobin deoxygenation experiment with minimal water crosstalk, in which the recovered changes in oxyhemoglobin, deoxyhemoglobin, and water are plotted over time for an LED-based device (FlexNIRS) and a reference FD-NIRS system. The upper-right panels show bar charts comparing the recovered changes in oxyhemoglobin (AHbO), deoxyhemoglobin (AHbR), and water fraction (Awater) across a reference FD-NIRS system, a continuous-wave laser-based NIRS system (CW-NIRS), and the LED-based system using the spectrally weighted approach (GS-FlexNIRS), demonstrating water content recovery with minimal deoxyhemoglobin crosstalk. The lower-left panel shows a further deoxygenation experiment comparing the LED-based system usingAtty. Dkt. No. 125141.05020. MGH2025-380the spectrally weighted approach to both the FD-NIRS reference and a broadband NIRS (bbNIRS) reference. These results illustrate that the spectrally weighted approach may enable accurate chromophore recovery across different device configurations, including LED-based and laser-based systems, and across different chromophores including hemoglobin species and water.
[0139] Referring to FIGS. 10-13, a simulation-based comparison is shown between the spectrally weighted method described herein and a prior art method for chromophore concentration recovery. FIG. 10 illustrates the simulation framework, in which: (1) a set of tissue chromophore concentrations and scattering properties are selected, as shown in the table on the left side of FIG. 10, including seven experimental sets with varying concentrations of oxyhemoglobin (e.g., about 45 to about 55 pM), deoxyhemoglobin (e.g., about 25 to about 35 pM), and water fraction (e.g., about 70% to about 80%); (2) Monte Carlo simulations are performed for each pair of absorption and scattering properties in the spectrum to generate simulated detected intensities at multiple source-detector separations (e.g., about 0.77 cm, about 2.49 cm, about 2.99 cm, and about 3.47 cm), as illustrated on the right side of FIG. 10; and (3) the simulated data are fitted using both the prior art method and the spectrally weighted method. The prior art method generates one recovered value per source-detector separation, whereas the spectrally weighted method uses the intensities across multiple source-detector separations together to generate a single recovered value. The reduced scattering coefficient inthis simulation was modeled as i / ((A) = 9.3 ( - ) cm1.\750 nmz
[0140] Referring to FIG. 11, the recovered changes in oxyhemoglobin (AHbO), deoxyhemoglobin (AHbR), and water fraction (AH2O) are shown for six perturbation conditions: positive and negative changes in each of HbO, HbR, and H₂O. Blue bars represent the prior art method applied at individual source-detector separations, orange bars represent the spectrally weighted method described herein, and yellow bars represent the ground truth values. As shown, the spectrally weighted method may produce recovered values that are closer to the ground truth and may exhibit reduced crosstalk between chromophores compared to the prior art method. For example, when the ground truth perturbation is an increase in oxyhemoglobin (+HbO), the prior art method may show variable recovery across different source-detector separations and spurious changes in deoxyhemoglobin and water, whereas theAtty. Dkt. No. 125141.05020. MGH2025-380spectrally weighted method may recover the expected change with reduced spurious variation in the other chromophores.
[0141] Referring to FIG. 12, tables of mean absolute error (MAE) and percent error are shown comparing the prior art method (left values) to the spectrally weighted method (right values) for each perturbation condition. As shown, the spectrally weighted method may achieve an approximately four-fold reduction in percent error for both real changes (e.g., detecting an increase in HbO when HbO increases) and crosstalk changes (e.g., spurious change in HbO when HbR increases) across all three chromophores. These results further illustrate the reduced systematic bias and improved accuracy that may be achieved by incorporating the spectral weighting profiles into the chromophore concentration determination, as described in connection with blocks 412-420 of FIG. 4.
[0142] Referring to FIG. 13, a table of absolute chromophore concentration estimation errors is shown for each of the seven simulated experimental sets using the spectrally weighted multi-distance approach. The table reports the mean absolute error in oxyhemoglobin (MAE in AHbO, in pM), deoxyhemoglobin (MAE in AHbR, in pM), and water fraction (MAE in AH2O, in volume %). As noted in FIG. 13, while the error in absolute estimation may be higher than in relative estimation, the ability to obtain absolute estimates of tissue chromophore concentrations from the spectrally weighted multi -di stance approach may be a significant advantage for applications where absolute concentration values are desired, such as initial assessment of tissue oxygenation or comparison across subjects or time points.EXAMPLES
[0143] The following examples have been presented in order to further illustrate aspects of the disclosure, and are not meant to limit the scope of the disclosure in any way. The examples below are intended to be examples of the present disclosure and these (and other aspects of the disclosure) are not to be bounded by theory.
[0144] To assess system stability during prolonged operation, effects of self-heating and battery voltage on LED intensity were investigated. The MW-FlexNIRS device was placed on a solid phantom with known optical properties and powered using a fully charged battery. Light intensity data were continuously acquired until the battery was fully depleted. In addition to light intensity, battery voltage, LED temperature, and ambient temperature at a samplingAtty. Dkt. No. 125141.05020. MGH2025-380rate of 1 Hz were monitored. Temperature was measured using a fiber optic temperature sensor (Reflex, Neoptix) with one probe positioned directly on the backside of the flexible PCB, immediately beneath the 8-wavelength LED package, and a second probe placed near the phantom to record room temperature. To monitor battery voltage, a modified lithium-ion battery with a secondary BNC-connectorized output was measured by a digital acquisition device (USB-6003, National Instruments). All LEDs were driven at the maximum current of 250 mA. Prior to turning on the LEDs, the system was allowed to reach equilibrium with room temperature.
[0145] To evaluate the sensitivity of the MW-FlexNIRS system, the noise equivalent power (NEP) was measured. Light from each LED was coupled into a fiber bundle and passed through two sets of neutral density filters mounted on a filter wheel. All LEDs were driven at their maximum current (250 mA), and PD gain was set to its maximum to replicate real-use conditions. To estimate the optical power incident on the PD for each LED / filter combination, the wavelength-dependent attenuation of the filters using a power meter (843 -R power with 818-SL detector, Newport Corporation) was characterized. From these measurements, an incident power matrix was generated based on the known filter combinations and the unattenuated optical power for each LEDs. For each attenuation level, light intensity was measured using MW-FlexNIRS for 10 seconds, and the signal was averaged in 500 ms intervals. NEP was determined by plotting the detector signal as a function of incident power and identifying the point where the extrapolated sensitivity curve intersected the noise floor, measured with the LED off. The noise floor was defined as the standard deviation of the dark measurement. The intersection of the two lines represents the minimum detectable optical power above the system’s intrinsic noise.
[0146] Two sets of phantom validation experiments were conducted to evaluate key performance metrics of the MW-FlexNIRS system. First, solid silicone tissue-like phantoms with known optical properties were used to assess the accuracy of effective attenuation coefficient ( / .w) recovery using the multi-distance approach. Second, blood-Intralipid-yeast liquid phantoms were used to test the system’s ability to resolve dynamic changes in chromophore concentrations under physiologically relevant conditions using the singlechannel AOD approach.Atty. Dkt. No. 125141.05020. MGH2025-380
[0147] To evaluate the accuracy of recovery with the MW-FlexNIRS device, measurements were performed on four silicone phantoms with known optical properties. One phantom was used to calibrate the device, while the remaining three were reserved for validation. For the low attenuation phantom, neutral density filter films (Wratten 2, Kodak) were used to prevent detector saturation. Since the manufacturer provided optical property values only at the wavelengths of the reference FD-NIRS system, absorption was interpolated and scattering values to the MW-FlexNIRS wavelengths using a linear and power-law approach, respectively.
[0148] .eff was estimated using two methods: (1) a per-source multi-distance WWF approach, applied independently to each source (SI and S2) using detectors DI, D2, and D4 (at 3.3, 2.8 and 2.4 cm source-detector separation); and (2) a self-calibrating, cross-symmetric source-detector geometry using SI, S2 and detectors DI and D2. In the latter case, centroid wavelengths derived from the combined LED emission spectra and PD responsivity were used due to the method relying on discrete wavelengths. Since the WWF algorithm calculates the weighted effective attenuation coefficient across the entire emission range for a given LED, recovered values were compared to the / le interpolated from manufacturer-reported values at the centroid wavelength of each LED.
[0149] To validate the MW-FlexNIRS system’s ability to recover changes in HbO, HbR, and oxCCO concentrations, a study was conducted on liquid phantoms containing baker’s yeast (Saccha-romyces cerevisiae, YSCII, Sigma-Aldrich). Yeast was chosen as a biologically relevant oxygen consumer via mitochondrial metabolism and would provide redox changes in CCO during aerobic deoxygenation.
[0150] To characterize yeast-specific cytochrome absorption features, a bloodless liquid phantom was prepared, similar to blood-Intralipid-phantoms, consisting of 750 mL deionized water, 750 mL phosphate-buffered saline (PBS), 84 mL of 20% Intralipid solution, and 12 g of baker’s yeast. The yeast concentration was increased relative to typical values to enhance the CCO signal. This phantom was measured using a broadband imaging spectrometer (bbNIRS) and a reference FD-NIRS system.
[0151] The measurement followed three stages: 1) Baseline - optical properties measured before adding yeast. 2) Aerobic metabolism - Yeast was added and mixed for approximately 2 min until homogenous. Data were acquired for 3 minutes as oxygen was consumed. 3)Atty. Dkt. No. 125141.05020. MGH2025-380Metabolic inhibition - 1 g sodium dithionite (Na2S2O4) was added to suppress aerobic metabolism of yeast by removing residual oxygen. Data were acquired for an additional 10 minutes.
[0152] Measurements were acquired sequentially at 60-second intervals using:
[0153] Broadband imaging spectrometer (bbNIRS): consisting of a ProEM HS camera (Princeton Instruments), a SpectraPro 2150i spectrograph (Acton Research Corporation), and a halogen light source (HL-2000-FHSA, Ocean Optics). Spectral range 650-920 nm, 2 nm spectral resolution, 10 s integration time.
[0154] Reference FD-NIRS system: a multi-distance system operating at 10 Hz with 30 second averaging, used to estimate scattering and interpolate DPF for bbNIRS fitting at 8 discrete wavelengths between 670-830 nm.
[0155] The phantom was housed in a 3D-printed resin container (e.g., 12.8 x 12.8 x 11 cm3), with optical windows sealed using transparent film (e.g., Suresite 123, Medline) to prevent leakage and ensure good optical coupling. A magnetic stirrer maintained mixing and prevented yeast sedimentation.
[0156] Single channel bbNIRS measurements were used at 20 mm to capture changes in optical density (AOD) across the 650-920 nm range. The absorption spectra were fitted using the oxidized-minus-reduced difference spectra of cytochromes b (Cyt b), c (Cyt c), and CCO. Changes in chromophore concentrations were recovered using least squares fitting, with DPF values interpolated from the FD-NIRS measurements.
[0157] A separate control experiment confirmed that sodium dithionite did not significantly alter absorption or scattering when added to a liquid phantom containing only water and Intralipid.
[0158] A phantom identical in com-position to the previous experiment was used, with the addition of 10 mL of bovine blood (#7200806, Lampire Biological Labs) with an estimated HbT of 13.2 pM once diluted in the phantom. The phantom was maintained at 20 °C in the same 3D-printed resin container. bbNIRS and FD-NIRS probes were mounted on opposite walls, and MW-FlexNIRS was placed orthogonally on the third window (see Fig. 3). Absence of optical crosstalk between the MW-FlexNIRS and FD-NIRS probes was confirmed, and their light sources were turned on 30 minutes before acquisition to reach thermal equilibrium.Atty. Dkt. No. 125141.05020. MGH2025-380
[0159] A multi-distance FD-NIRS measurement was performed before yeast addition using a transla-tional stage setup on top of the container, with source and detector fibers immersed in the liquid. A single detector was incrementally moved from 2.0 to 2.8 cm away from a fixed source, in 2 mm steps. The resulting optical property estimates were used to calibrate both MW-FlexNIRS and FD-NIRS probes.
[0160] After a short 10-second baseline with the 3 devices, 12 g of yeast was added to the phantom and stirred until the solution appeared visually homogenous. The container was sealed with plastic film to limit oxygen exchange with ambient air. MW-FlexNIRS and FD-NIRS collected data continuously throughout the deoxygenation process. Once a desaturation plateau was reached, 1 g sodium dithionite was added to remove any residual dissolved oxygen and halt yeast metabolism, preventing reoxygenation
[0035] , bbNIRS measurements were acquired at three timepoints: baseline, plateau, and post-dithionite. During each bbNIRS scan, light sources from the other devices were temporarily turned off to avoid interference. The pH of the phantom was measured at the start and at the end of the deoxygenation, prior to the addition of sodium dithionite.
[0161] MW-FlexNIRS data were analyzed using several approaches: i) the WWF model across the full LED range (700-920 nm); ii) the WWF model restricted to LEDs with peak emission above 780 nm
[0032] ; iii) the full WWF model fitting only for hemoglobin; and iv) a conventional peak-based discrete-wavelength approach in which each LED was modeled as monochromatic at its measured peak wavelength (Table 1), without spectral weighting. For comparison, the reference FD-NIRS device was limited to fitting only HbO and HbR, as it lacked wavelengths above 830 nm necessary for reliable sensitivity to CCO redox changes. All data from the 3 devices were evaluated under two model assumption: unconstrained total hemoglobin (HbT) changes, and constrained AHbT = 0. Moreover, to evaluate potential crosstalk, we calculated the Pearson’s correlation coefficient and the linear relationship between AHbT and AoxCCO. Changes in absorption for the bbNIRS system were computed using the modified Beer-Lambert law. DPF values were interpolated from FD-NIRS optical properties during the whole experiment, and nonlinear fitting was performed using the Levenberg-Marquardt algorithm.
[0162] A series of pilot feasibility measurements was performed on term infants (e.g., older than 36 weeks gestational age) diagnosed with transient respiratory distress requiring4±0Atty. Dkt. No. 125141.05020. MGH2025-380respiratory support within the first 28 days of life. Infants with known congenital or genetic anomalies were excluded. Clinical staff approached eligible families and obtained written informed consent prior to enrollment. Enrolled infants were continuously monitored with the MW-FlexNIRS device starting at the time of enrollment. Monitoring was conducted for 6 (+ 4) hours per day, for up to 3 days.
[0163] To secure the MW-FlexNIRS device to the infant’ s forehead, a modified headband was used. A soft skin-safe silicone tape (e.g., Mepitac, Molnlycke) was placed over the top of the device to provide additional support and prevent the device from moving during monitoring. The optical probe of the device was wrapped using a thin hydrogel tape (BTHG-250X, Breachers Tape) to prevent moisture from interfering with the device, as well as avoid skin irritation. The device can be placed inside a headband during a measurement on an RDS infant. In addition to MW-FlexNIRS, peripheral oxygen saturation (SpO₂) was recorded using a commercially available pulse oximeter (Rad-87, Masimo). Pulse oximeter sensor placement, either on the infant’s hand or foot, followed routine clinical practice and was determined by clinical staff based on the need for periodic rotation to maintain skin integrity. For the feasibility measurement reported here, the sensor was placed on the infant’s foot.
[0164] The infant’s data were analyzed using a single-channel AOD approach. In this analysis, the reference LED intensity FFDin Eq. (6) was redefined every 30 minutes as a local baseline. This periodic baseline update reduces slow drifts from instrumental and coupling changes over long recordings. Because absolute optical properties (.a and / is') could not be measured directly, to estimate the differential pathlength factor (DPF) and perform concentration fitting, average values of HbT, SO2, and ps’ were used.
[0165] The normalized emission spectra of each source in the 8-wavelength LED package was measured at 38 °C using the STS-NIR spectrometer. The detected spectral shift due to the wavelength-dependent detector responsivity. When measured at a lower temperature (19 °C), peak wavelength shifts of up to 16 nm and centroid shifts up to 11 nm. Due to the short duty cycle (<1.2%), spectral variations with driving current were negligible.
[0166] System stability was shown during an overnight measurement. Optical intensity data were downsampled to one datapoint every 4 minutes. The recording lasted 10 hours and 11 minutes until battery depletion. The temperature measured on the back of the LED package increased from 21.3 °C to a stable 25.4 °C within the first 30 minutes and then remained almostAtty. Dkt. No. 125141.05020. MGH2025-380constant. LED intensity changed to a maximum of 2.3% relative to its final value, primarily due to the change in temperature. The signal showed a standard deviation of only 0.003% of the full-scale range (FSR) across all channels. After calibration, the per-source multi -di stance calculated peff had a coefficient of variation <0.9% across all wavelengths.
[0167] The noise equivalent power was measured at each wavelength, obtaining an average NEP of 123.2 ( ±25.0) fW / √Hz. The measured intensity can be a function of incident power for each wavelength, with the NEP determined by the intersection of the extrapolated sensitivity curve and the noise floor.
[0168] An example comparison ofvalues vs wavelength obtained in a block from individual LED sources using the multi-distance WWF algorithm (D 1, D2 and D4) and using the cross-symmetric geometry.
[0169] The recovered [.leg values for three silicone phantoms plotted against the manufacturer reported values (interpolated at the wavelengths of interest) using the multidistance WWF algorithm applied to individual sources and the corresponding Bland- Altman plot.
[0170] Changes in light intensity were characterized, absorption and scattering due to yeast metabolism in a bloodless phantom during aerobic metabolism using bbNIRS and FD-NIRS. Between the yeast-only and the yeast + Na2S2CU stages, / changed by 1% 0.4% across all wavelengths, slightly larger than in the sodium-only control, possibly due to minor yeast sedimentation despite stirring.
[0171] To quantify relative contributions of individual cytochromes to the measured absorption changes, the AOD spectra obtained with bbNIRS was analyzed using least-squares fitting against the oxidized-reduced spectra of Cyt b, Cyt c, and CCO. The DPF values used for the fitting were interpolated from FD-NIRS measurements.
[0172] The results: (a) Measured AOD spectra at 1-minute intervals (b) and recovered chromophore concentrations changes over time. Following sodium dithionite addition (minute 3), Cyt b showed the most pronounced reduction. Final concentration changes were -4.69 pM for Cyt b, -0.49 pMforCytC, and -0.28 pMforoxCCO. The residual fitting error had an average value of 17% across the spectral range, with the largest deviations observed near 730 nm (52%).Atty. Dkt. No. 125141.05020. MGH2025-380The recovered concentrations reflect clear metabolic effects, with Cyt b exhibiting a ~5 pM reduction.
[0173] The translational stage FD-NIRS measurement of the baseline phantom estimated a HbT concentration of 13.6 pM and an SO2 of 96.7%. Both the probe-based reference FD-NIRS system AC intensity and phase slopes, as well as the MW-FlexNIRS intensity showed a high linearity (R2> 0.995). Table 2 reports the measured changes in chromophore concentrations during the desaturation plateau (700 to 1050 seconds after the addition of yeast), for each device and fitting procedure.
[0174] When analyzed without any constraints, changes in HbT during the desaturation plateau were -0.83 pM, 0.71 pM, and 2.19 pM for MW-FlexNIRS (WWF Full), FD-NIRS, and bbNIRS, respectively, despite the fact that total hemoglobin concentration was expected to remain constant in this phantom configuration. Following common practice in previous oxCCO studies, which restrict analysis to longer wavelengths to enhance oxCCO sensitivity [17,32,33], the WWF algorithm was evaluated using only wavelengths above 780 nm. In this case the MW-FlexNIRS data showed a larger estimated HbT change of 2.71 pM. To assess crosstalk between hemoglobin and CCO in the unconstrained MW-FlexNIRS fits, Pearson’s correlation coefficients were calculated between the AHbT and AoxCCO time traces prior to sodium dithionite addition. Using the WWF algorithm across the full wavelength range and using wavelengths above 780 nm, the correlations yielded r values of -0.971 and -0.999 (p < 0.001 for both), respectively, with corresponding regression slopes of -0.27 and -0.53.
[0175] For comparison, results obtained using a conventional manufacturer-reported peak discrete- wavelength approach. In this case, HbO and HbR changes were sub -stanti ally underestimated and oxCCO estimates differed markedly from both FD-NIRS and bbNIRS. The full unconstrained chromophore time traces for all fitting approaches, as well as the corresponding HbT-oxCCO regression plots.
[0176] Notably, differences between constrained and unconstrained fits were not limited to magnitude but also affected the sign of the recovered oxCCO changes. Taken together, the polarity reversals between constrained and unconstrained fits, and between WWFs and bbNIRS, indicate that oxCCO changes in this phantom are highly model-dependent and should not be treated as a quantitative ground truth for cross-device validation.Atty. Dkt. No. 125141.05020. MGH2025-380
[0177] Chromophore time traces obtained under the constraint that AHbT = 0 pM. Under this constraint, the average errors in HbO and HbR concentration estimates for the MW-FlexNIRS WWF Full algorithm, relative to the probe-based reference FD-NIRS system, were 0.22 pM across the full measurement and 0.37 pM at the desaturation plateau.
[0178] Time traces of cerebral SO2, HbT, and AoxCCO acquired can be shown using the MW-FlexNIRS device. The preliminary data were collected from an infant diagnosed with transient respiratory distress. The MW-FlexNIRS data was obtained by averaging the recovered parameters for the longer SD separations, 2.8 and 3.3 cm, across each source, SI and S2, with the single-channel
[0179] AOD WWF algorithm. Peripheral SpO₂ and heart rate data were simultaneously recorded using a commercial pulse oximeter and are show for comparison. Periods where motion was detected are highlighted in gray. To evaluate the consistency between the two instruments, Pearson correlation coefficients were computed between the SO2 time traces from S I and S2 and the SpO₂ signal from the pulse oximeter. The correlation values for SI and S2 were r = 0.61 and r = 0.63, respectively (p-value <0.001 for both).
[0180] The MW-FlexNIRS device can incorporate improvements in both internal architecture and form factor to enable investigation of oxCCO variations. Integration of a new AFE supports simultaneous operation of two compact 8-wavelength LEDs, while the updated MCU and BLE 5.0 interface allow higher data transfer rates with minimal data loss during wireless transmission.
[0181] By integrating all electrical and optical components onto a single flexible PCB, the MW-FlexNIRS improves robustness over earlier designs that relied on external cables and discrete modules prone to mechanical failure. The integrated design eliminated the risk of disconnections and connector breakage, making the device more suitable for NICU use, where infant movement can lead to equipment displacement. The compact form factor and soft medical-grade silicone enclosure enable safe and comfortable placement alongside other clinical monitors. In addition, the encapsulation process streamlines manufacturing by embedding the coated flexible PCB directly into a silicone mold, eliminating the need for 3D-printed housing and manual device assembly.
[0182] The use of compact multi -wavelength LED packages (Marubeni America Corporation) enabled stable and uniform spectral output within a small footprint. EmissionAtty. Dkt. No. 125141.05020. MGH2025-380profiles were consistent across devices and driving currents. Operating well below ANSI maximum permissible limits provided sufficient optical power at the longest source-detector separations (up to 3.3 cm) to support low-noise measurements. MW-FlexNIRS demonstrated excellent noise performance, with an average NEP of 123.2 fW / √Hz, comparable to the two-wavelength FlexNIRS despite the expanded wavelength range, enabling detection of subtle optical intensity changes associated with weak absorbers such as CCO. Both emission and detection remained stable over prolonged operation and were unaffected by battery voltage decay. Intensity drift during overnight testing was minimal (<2.5%) and, after initial LED warm-up, was driven primarily by ambient temperature rather than device self-heating. Operation in a temperature-controlled clinical environment, such as the NICU, is therefore expected to further improve stability. For single-channel AOD measurements, a 1 h warm-up period was observed, after which the intensity stabilized to within 1% of its steady-state value. In contrast, multi-distance measurements were highly stable from the start of operation.
[0183] In silicone phantom validation experiments, MW-FlexNIRS accurately recovered peff using the calibrated per-source multi-distance approach, with an average error of 2.1%, even during the LED warm-up period. In contrast, error increased substantially to 14.4% when using the self-calibrating, cross-symmetric geometry. This discrepancy arises from a geometrical asymmetry in the MW -FlexNIRS probe: the two sources use identical 24 multiwavelength LED arrays, with individual wavelengths originating from different physical locations within the array. Consequently, a given wavelength does not correspond to the same effective source-detector separation in mirrored configurations, violating the symmetry assumption underlying cross-symmetric calibration. This limitation was not present in the original two- wavelength FlexNIRS design, where each wavelength originated from a symmetrically placed emitter.
[0184] To address limitations inherent to LED-based systems, a wavelength-weighted fitting (WWF) algorithm was developed for quantitative chromophore estimation. Unlike conventional discrete- wavelength approaches that approximate each source by a single representative wavelength, the WWF algorithm accounts for the full LED emission spectra, the wavelength-dependent responsivity of photodiodes and tissue optical properties. This distinction is critical for LED-based NIRS, where the detected signal represents an integrated spectral response. Without spectral weighting, peak- or centroid-based models can introduceAtty. Dkt. No. 125141.05020. MGH2025-380systematic bias, particularly when estimating weak or spectrally overlapping absorbers such as oxCCO.
[0185] To evaluate WWF in a biologically relevant setting, blood-Intralipid-yeast phantom experiments was conducted. Using WWF with the constraint that total hemoglobin remained constant (AHbT = 0), MW-FlexNIRS recovered HbO and HbR changes in close agreement with FD-NIRS and bbNIRS references. MW-FlexNIRS estimates differed by only 0.37 pM from FD-NIRS and 0.22 pM from bbNIRS, confirming accurate hemoglobin recovery. In contrast, peak-based discrete-wavelength fitting substantially underestimated hemoglobin changes and produced inconsistent oxCCO estimates.
[0186] During yeast-driven aerobic metabolism, a decrease in oxCCO is expected as oxygen levels are depleted, reflecting reduced mitochondrial electron transport activity. These experiments were not intended to validate absolute oxCCO magnitude, but rather to stress-test spectral separability in a biologically active, non-mammalian, metabolic system. When applying the full-spectrum WWF model with AHbT = 0, MW-FlexNIRS recovered a decrease in oxCCO (-0.15 pM) consistent with prior yeast phantom studies even if the magnitude was lower than expected given the high yeast concentration used [32-34] This differed from the bbNIRS measurements, which showed a small oxCCO increase (+0.25 pM). Following sodium dithionite addition, a larger apparent oxCCO decrease (-0.74 pM) was observed, accompanied by a reduction in HbO / HbR (0.94 pM), suggesting increased parameter coupling under extreme deoxygenation.
[0187] When the HbT constraint was removed, hemoglobin estimates remained similar to the constrained case with an apparent HbT change (-0.83 pM), while oxCCO increased slightly (+0.20 pM). This dependence of oxCCO sign on hemoglobin constraint indicates residual crosstalk between oxCCO and hemoglobin. Restricting the fit to wavelengths above 780 nm, an approach commonly adopted to enhance oxCCO sensitivity, amplified oxCCO changes but introduced imbalanced HbO / HbR recovery and a larger apparent HbT increase (+2.71 pM), indicative of model degeneracy when spectral information is reduced. Strong negative correlations between oxCCO and HbT further support the interpretation. Similar behavior was observed in the bbNIRS when shorter wavelengths were excluded. The ability to invert oxCCO polarity by changing either model constraints (AHbT = 0 vs. unconstrained) or wavelength rangeAtty. Dkt. No. 125141.05020. MGH2025-380(full vs. >780 nm) suggests that the phantom oxCCO signal is below the robust detection limit and dominated by fitting degeneracy rather than a stable underlying physiological change.
[0188] These observations suggest that oxCCO ambiguity may reflect limitations of the yeast phantom model rather than shortcomings of the MW-FlexNIRS device. Unlike some previous yeast phantom studies that added glucose to stimulate metabolism, sugar was intentionally excluded to promote aerobic respiration and maximize mitochondrial activity. In glucose-rich environments, yeast preferentially undergo fermentation rather than oxidative phosphorylation, reducing CCO engagement and potentially confounding oxCCO measurements.
[0189] To further isolate mitochondrial contributions, measurements were performed in a bloodless yeast-Intralipid phantom. Sodium dithionite produced no measurable optical changes in an Intralipid-water phantom, whereas clear spectral changes were observed in the presence of yeast. Spectral fitting suggested a dominant reduction in cytochrome b ( 4.7 pM), with smaller contributions from cytochrome c ( 0.5 pM) and CCO ( 0.28 pM).
[0190] To assess the feasibility of MW-FlexNIRS for in vivo neonatal monitoring, preliminary data was acquired in a term infant undergoing management of transient respiratory distress. Cerebral SO2 changes measured with MW-FlexNIRS showed moderate correlation with peripheral SpO₂ obtained from a commercial pulse oximeter, suggesting that the device can capture broad oxygenation trends at the bedside, although larger studies are needed to confirm this.
[0191] For oxCCO, channel-dependent behavior was observed. Given the clinical stability of the infant and absence of brain injury, large oxCCO fluctuations were not expected. In the SI channels, a gradual increase in oxCCO began shortly after a motion event detected at approximately 9:00. This change was accompanied by a concurrent downward trend in the HbT. In a clinically stable, resting infant with no diagnosed brain injury, this combination is not physiologically expected and suggests a non-physiological origin. In contrast, the S2 channels showed stable oxCCO and HbT signals throughout the recordings, with no significant drift. The divergence between S 1 and S2 channels, together with the temporal association with motion and the opposing behavior of oxCCO and HbT in SI, indicates that the observed SI trend is more likely attributable to changes in optode coupling or motion-related artifacts rather than true metabolic variation. This interpretation is consistent with the known sensitivity of oxCCO estimates to coupling changes and with the weak optical signature of oxCCO relativeAtty. Dkt. No. 125141.05020. MGH2025-380to hemoglobin. Given that the infant remained clinically stable throughout the measurement and minimal metabolic fluctuations were expected, the absence of significant oxCCO variation in the S2 channels is consistent with physiological expectations.
[0192] Taken together, these phantom and initial human results highlight both the strengths and current limitations of oxCCO quantification using LED-based NIRS systems. The yeast phantom experiments suggest that oxCCO may represent a weak optical absorber in yeast, with near-infrared spectral changes likely dominated by other mitochondrial chromophores such as cytochrome b. Spectral overlap and incomplete chromophore modeling further contribute to parameter crosstalk, particularly between HbT and oxCCO, complicating interpretation in both blood and bloodless phantom configurations. Even with eight wavelengths, these results show that extracting weak oxCCO signals without physiological constraints remains challenging, and current estimates should be regarded as feasibility-level rather than ready for clinical decision-making.
[0193] At the same time, the in vivo neonatal measurements demonstrate that MW-FlexNIRS provides stable, reliable cerebral oxygenation estimates and physiologically plausible oxCCO behavior when signal coupling is preserved. The combination of hardware stability, wavelength-weighted fitting, and multi-channel consistency supports the feasibility of MW-FlexNIRS for neonatal brain monitoring.
[0194] A MW-FlexNIRS system was developed for noninvasive monitoring of cerebral oxygenation and metabolism in neonates. The design can integrate custom eight- wavelength LED sources, a new analog front-end and wireless BLE 5 data transmission on a single flexible PCB encapsulated in medical-grade silicone, resulting in a compact, robust, and clinically compatible device suitable for continuous monitoring in the neonatal intensive care unit.
[0195] To support quantitative analysis with LED-based sources, a wavelength-weighted fitting algorithm was introduced that explicitly accounts for LED emission spectra and photodiode responsivity. Phantom experiments demonstrated that MW-FlexNIRS combined with WWF provides accurate and stable hemoglobin recovery, closely matching reference FD-NIRS and broadband NIRS systems and outperforming conventional discrete- wavelength approaches. These results highlight the importance of spectral weighting for quantitative chromophore estimation in compact multi-wavelength LED-PD systems.Atty. Dkt. No. 125141.05020. MGH2025-380
[0196] Initial neonatal feasibility measurements showed that MW-FlexNIRS can track cerebral SO2 with moderate correspondence to peripheral pulse oximetry, while oxCCO remained stable in a clinically normal infant, consistent with physiological expectations. Although we present data from a single representative term infant, the measurements demonstrate the feasibility of MW-FlexNIRS for bedside neonatal monitoring and highlight areas for continued refinement.
[0197] Referring to FIGS. 14-19, validation results for the system 100 are shown. FIG. 14 shows the wearable spectroscopy device secured to a subject's forehead using a headband and skin-safe tape during in vivo monitoring in a clinical environment. FIG. 15 shows a three-dimensional printed phantom container with optical windows on multiple walls, configured to allow simultaneous measurements by the wearable spectroscopy device, a reference frequencydomain NIRS system, and a broadband imaging spectrometer for cross-validation. FIG. 16 shows system stability over an approximately 10-hour operating period, illustrating that optical intensity, recovered effective attenuation coefficients, battery voltage, and LED temperature remained stable throughout the measurement. FIG. 17 shows noise equivalent power measurements for each wavelength. FIG. 18 shows recovered effective attenuation coefficients on a solid phantom comparing a per-source multi -di stance approach. FIG. 19 shows correlation and Bland-Altman agreement analyses across three solid phantoms using the spectrally weighted multi-distance method.
[0198] Referring to FIGS. 20-23, chromophore recovery and in vivo monitoring results are shown. FIG. 20 shows measured changes in optical density and recovered cytochrome concentration changes in a bloodless yeast phantom, illustrating the spectral characteristics of mitochondrial chromophores in the near-infrared range. FIG. 21 is a table comparing chromophore concentration changes recovered during a blood-yeast phantom desaturation experiment across multiple fitting approaches, including the spectrally weighted method, a conventional discrete peak-wavelength approach, and reference systems, under both unconstrained and constrained (e.g., AHbT = 0) conditions. FIG. 22 shows chromophore concentration time traces during the same experiment, demonstrating close agreement between the spectrally weighted method and the reference systems. FIG. 23 shows in vivo cerebral monitoring data acquired from a subject with transient respiratory distress, including total hemoglobin, cerebral oxygen saturation, changes in oxidized cytochrome c oxidase, heart rate,Atty. Dkt. No. 125141.05020. MGH2025-380and peripheral oxygen saturation, illustrating the feasibility of the system 100 for non-invasive cerebral oxygenation and metabolic monitoring using the spectrally weighted method described herein.
[0199] The present disclosure has described one or more preferred examples, and it should be appreciated that many equivalents, alternatives, variations, and modifications, aside from those expressly stated, are possible and within the scope of the invention.
[0200] It is to be understood that the disclosure is not limited in its application to the details of construction and the arrangement of components set forth in the accompanying description or illustrated in the accompanying drawings. The disclosure is capable of other examples and of being practiced or of being carried out in various ways. Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of “including,” “comprising,” or “having” and variations thereof herein is meant to encompass the items listed thereafter and equivalents thereof as well as additional items. Unless specified or limited otherwise, the terms “mounted,” “connected,” “supported,” and “coupled” and variations thereof are used broadly and encompass both direct and indirect mountings, connections, supports, and couplings. Further, “connected” and “coupled” are not restricted to physical or mechanical connections or couplings.
[0201] As used herein, unless otherwise limited or defined, discussion of particular directions is provided by example only, with regard to particular examples or relevant illustrations. For example, discussion of “top,” “front,” or “back” features is generally intended as a description only of the orientation of such features relative to a reference frame of a particular example or illustration. Correspondingly, for example, a “top” feature may sometimes be disposed below a “bottom” feature (and so on), in some arrangements or examples. Further, references to particular rotational or other movements (e.g., counterclockwise rotation) is generally intended as a description only of movement relative a reference frame of a particular example of illustration.
[0202] In some examples, aspects of the disclosure, including computerized implementations of methods according to the disclosure, can be implemented as a system, method, apparatus, or article of manufacture using standard programming or engineering techniques to produce software, firmware, hardware, or any combination thereof to control a processor device (e.g., a serial or parallel general purpose or specialized processor chip, aAtty. Dkt. No. 125141.05020. MGH2025-380single- or multi-core chip, a microprocessor, a field programmable gate array, any variety of combinations of a control unit, arithmetic logic unit, and processor register, and so on), a computer (e.g., a processor device operatively coupled to a memory), or another electronically operated controller to implement aspects detailed herein. Accordingly, for example, examples of the disclosure can be implemented as a set of instructions, tangibly embodied on a non-transitory computer-readable media, such that a processor device can implement the instructions based upon reading the instructions from the computer-readable media. Some examples of the disclosure can include (or utilize) a control device such as an automation device, a special purpose or general purpose computer including various computer hardware, software, firmware, and so on, consistent with the discussion below. As specific examples, a control device can include a processor, a microcontroller, a field-programmable gate array, a programmable logic controller, logic gates etc., and other typical components that are known in the art for implementation of appropriate functionality (e g., memory, communication systems, power sources, user interfaces and other inputs, etc.).
[0203] The term “article of manufacture” as used herein is intended to encompass a computer program accessible from any computer-readable device, carrier (e.g., non-transitory signals), or media (e.g., non-transitory media). For example, computer-readable media can include but are not limited to magnetic storage devices (e.g., hard disk, floppy disk, magnetic strips, and so on), optical disks (e.g., compact disk (CD), digital versatile disk (DVD), and so on), smart cards, and flash memory devices (e.g., card, stick, and so on). Additionally it should be appreciated that a carrier wave can be employed to carry computer-readable electronic data such as those used in transmitting and receiving electronic mail or in accessing a network such as the Internet or a local area network (LAN). Those skilled in the art will recognize that many modifications may be made to these configurations without departing from the scope or spirit of the claimed subject matter.
[0204] Certain operations of methods according to the disclosure, or of systems executing those methods, may be represented schematically in the FIGS, or otherwise discussed herein. Unless otherwise specified or limited, representation in the FIGS, of particular operations in particular spatial order may not necessarily require those operations to be executed in a particular sequence corresponding to the particular spatial order. Correspondingly, certain operations represented in the FIGS., or otherwise disclosed herein, can be executed in differentAtty. Dkt. No. 125141.05020. MGH2025-380orders than are expressly illustrated or described, as appropriate for particular examples of the disclosure. Further, in some examples, certain operations can be executed in parallel, including by dedicated parallel processing devices, or separate computing devices configured to interoperate as part of a large system.
[0205] As used herein in the context of computer implementation, unless otherwise specified or limited, the terms “component,” “system,” “module,” and the like are intended to encompass part or all of computer-related systems that include hardware, software, a combination of hardware and software, or software in execution. For example, a component may be, but is not limited to being, a processor device, a process being executed (or executable) by a processor device, an object, an executable, a thread of execution, a computer program, or a computer. By way of illustration, both an application running on a computer and the computer can be a component. One or more components (or system, module, and so on) may reside within a process or thread of execution, may be localized on one computer, may be distributed between two or more computers or other processor devices, or may be included within another component (or system, module, and so on).
[0206] In some implementations, devices or systems disclosed herein can be utilized or installed using methods embodying aspects of the disclosure. Correspondingly, description herein of particular features, capabilities, or intended purposes of a device or system is generally intended to inherently include disclosure of a method of using such features for the intended purposes, a method of implementing such capabilities, and a method of installing disclosed (or otherwise known) components to support these purposes or capabilities. Similarly, unless otherwise indicated or limited, discussion herein of any method of manufacturing or using a particular device or system, including installing the device or system, is intended to inherently include disclosure, as examples of the disclosure, of the utilized features and implemented capabilities of such device or system.
[0207] As used herein, unless otherwise defined or limited, ordinal numbers are used herein for convenience of reference based generally on the order in which particular components are presented for the relevant part of the disclosure. In this regard, for example, designations such as “first,” “second,” etc., generally indicate only the order in which the relevant component is introduced for discussion and generally do not indicate or require a particular spatial arrangement, functional or structural primacy or order.Atty. Dkt. No. 125141.05020. MGH2025-380
[0208] As used herein, unless otherwise defined or limited, directional terms are used for convenience of reference for discussion of particular figures or examples. For example, references to downward (or other) directions or top (or other) positions may be used to discuss aspects of a particular example or figure, but do not necessarily require similar orientation or geometry in all installations or configurations.
[0209] This discussion is presented to enable a person skilled in the art to make and use examples of the disclosure. Various modifications to the illustrated examples will be readily apparent to those skilled in the art, and the generic principles herein can be applied to other examples and applications without departing from the principles disclosed herein. Thus, examples of the disclosure are not intended to be limited to examples shown, but are to be accorded the widest scope consistent with the principles and features disclosed herein and the claims below. The accompanying detailed description is to be read with reference to the figures, in which like elements in different figures have like reference numerals. The figures, which are not necessarily to scale, depict selected examples and are not intended to limit the scope of the disclosure. Skilled artisans will recognize the examples provided herein have many useful alternatives and fall within the scope of the disclosure.
[0210] Also as used herein, unless otherwise limited or defined, “or” indicates a nonexclusive list of components or operations that can be present in any variety of combinations, rather than an exclusive list of components that can be present only as alternatives to each other. For example, a list of “A, B, or C” indicates options of A; B; C; A and B; A and C; B and C; and A, B, and C. Correspondingly, the term “or” as used herein is intended to indicate exclusive alternatives only when preceded by terms of exclusivity, such as “either,” “one of,” “only one of,” or “exactly one of.” Further, a list preceded by “one or more” (and variations thereon) and including “or” to separate listed elements indicates options of one or more of any or all of the listed elements. For example, the phrases “one or more of A, B, or C” and “at least one of A, B, or C” indicate options of one or more A; one or more B; one or more C; one or more A and one or more B; one or more B and one or more C; one or more A and one or more C; and one or more of each of A, B, and C. Similarly, a list preceded by “a plurality of’ (and variations thereon) and including “or” to separate listed elements indicates options of multiple instances of any or all of the listed elements. For example, the phrases “a plurality of A, B, or C” and “two or more of A, B, or C” indicate options of A and B; B and C; A and C; and A,Atty. Dkt. No. 125141.05020. MGH2025-380B, and C. In general, the term “or” as used herein only indicates exclusive alternatives (e g. “one or the other but not both”) when preceded by terms of exclusivity, such as “either,” “one of,” “only one of,” or “exactly one of.”
[0211] Also as used herein, unless otherwise specified or limited, the terms “about” and “approximately,” as used herein with respect to a reference value, refer to variations from the reference value of ± 15% or less (e.g., ± 10%, ± 5%, etc.), inclusive of the endpoints of the range. Similarly, the term “substantially equal” (and the like) as used herein with respect to a reference value refers to variations from the reference value of less than ± 30% (e.g., ± 20%, ± 10%, ± 5%) inclusive. Where specified, “substantially” can indicate in particular a variation in one numerical direction relative to a reference value. For example, “substantially less” than a reference value (and the like) indicates a value that is reduced from the reference value by 30% or more, and “substantially more” than a reference value (and the like) indicates a value that is increased from the reference value by 30% or more.
[0212] Various features and advantages of the disclosure are set forth in the following claims.
Claims
Atty. Dkt. No. 125141.05020. MGH2025-380CLAIMSWhat is claimed is:
1. A method for determining one or more chromophore concentrations in a tissue, the method comprising:(a) illuminating the tissue with a plurality of light sources, wherein each light source has a respective characterized emission spectrum spanning a range of wavelengths;(b) detecting, with at least one photodetector having a characterized spectral responsivity, light from each of the plurality of light sources after propagation through the tissue, thereby obtaining a detected intensity for each light source; (c) for each light source, determining a spectral weighting profile from the characterized emission spectrum of that light source and the characterized spectral responsivity of the at least one photodetector;(d) for each light source, computing a modeled intensity by summing, over the range of wavelengths, a product of the spectral weighting profile and a tissue attenuation profile; and(e) determining the one or more chromophore concentrations in the tissue by adjusting the one or more chromophore concentrations to reduce a difference between the modeled intensities and the detected intensities..
2. The method of claim 1, wherein each light source is a broadband light source and comprises a light-emitting diode (LED).Atty. Dkt. No. 125141.05020. MGH2025-380 3. The method of claim 1, wherein the tissue attenuation profile depends on wavelength-dependent optical absorption and scattering properties of the tissue and wherein the optical absorption is dependent on the one or more chromophore concentrations.
4. The method of claim 1, wherein the at least one photodetector comprises a plurality of photodetectors positioned at different distances from each light source, and wherein the measured detected intensity for each light source comprises intensities detected at the different distances.
5. The method of claim 4, wherein step (e) comprises:deriving, for each light source, a measured effective attenuation coefficient from the measured detected intensities at the different separations;deriving, for each light source, a modeled effective attenuation coefficient from the modeled detected intensities at the different separations; andfitting the modeled effective attenuation coefficients to the measured effective attenuation coefficients.
6. The method of claim 5, wherein step (e) further comprises minimizing a loss function defined as a sum of squared differences between the modeled effective attenuation coefficient and the measured effective attenuation coefficient for each light source.
7. The method of claim 1, wherein the measured detected intensity for each light source is obtained at a single source-detector separation at a plurality of time points.Atty. Dkt. No. 125141.05020. MGH2025-380 8. The method of claim 7, wherein step (e) comprises:deriving, for each light source, a measured change in optical density from the measured detected intensity at a current time point to a baseline measured detected intensity;deriving, for each light source, a modeled change in optical density from the modeled detected intensities incorporating a differential pathlength factor; and fitting the modeled changes in optical density to the measured changes in optical density.
9. The method of claim 8, wherein the baseline measured detected intensity is periodically updated during monitoring.
10. The method of claim 1, wherein the one or more chromophore concentrations comprise at least one of oxyhemoglobin, deoxyhemoglobin, or oxidized cytochrome c oxidase.
11. The method of claim 1, wherein the tissue comprises biological tissue, and the method is performed non-invasively.
12. The method of claim 11, wherein the biological tissue is cerebral tissue.
13. The method of claim 1, wherein the characterized emission spectrum and the characterized spectral responsivity are measured at a predetermined operating temperature.Atty. Dkt. No. 125141.05020. MGH2025-38014. The method of claim 1, further comprising detecting motion of the tissue using a motion sensor.
15. A system for determining one or more chromophore concentrations in tissue, the system comprising:a plurality of light sources, each having a respective characterized emission spectrum spanning a range of wavelengths, configured to illuminate the tissue;at least one photodetector having a characterized spectral responsivity, configured to detect light from each of the plurality of light sources after propagation through the tissue, thereby producing a detected intensity for each light source; anda processor communicatively coupled to the at least one photodetector and configured to:for each light source, determine a spectral weighting profile from the characterized emission spectrum of that light source and the characterized spectral responsivity of the at least one photodetector;for each light source, compute a modeled intensity by summing, over the range of wavelengths, a product of the spectral weighting profile and a tissue attenuation profile; anddetermine the one or more chromophore concentrations by adjusting the one or more chromophore concentrations to reduce a difference between the modeled intensities and the detected intensities..Atty. Dkt. No. 125141.05020. MGH2025-380 16. The system of claim 15, wherein the tissue attenuation profile depends on wavelength-dependent optical absorption and scattering properties of the tissue and wherein the optical absorption is dependent on the one or more chromophore concentrations.
17. The system of claim 15, further comprising a plurality of photodetectors arranged at a plurality of distances from the plurality of light sources to enable multi-distance measurements.
18. A wearable spectroscopy device for monitoring chromophore concentrations in a tissue, the wearable spectroscopy device comprising:a flexible printed circuit board comprising:a plurality of light sources, each having a respective characterized emission spectrum spanning a range of wavelengths, configured to illuminate the tissue; andat least one photodetector having a characterized spectral responsivity, configured to detect light from each of the plurality of light sources after propagation through the tissue, thereby producing a detected intensity for each light source;a biocompatible enclosure encapsulating the flexible printed circuit board; and a processor configured to:for each light source, determine a spectral weighting profile from the characterized emission spectrum of that light source and the characterized spectral responsivity of the at least one photodetector;Atty. Dkt. No. 125141.05020. MGH2025-380 for each light source, compute a modeled intensity by summing, over a range of wavelengths, a product of the spectral weighting profile and a tissue attenuation profile; anddetermine the one or more chromophore concentrations by adjusting the one or more chromophore concentrations to reduce a difference between the modeled intensities and the measured intensities.
19. The wearable spectroscopy device of claim 18, wherein the tissue attenuation profile depends on wavelength-dependent optical absorption and scattering properties of the tissue and wherein the optical absorption is dependent on one or more chromophore concentrations.
20. The wearable spectroscopy device of claim 18, wherein the biocompatible enclosure comprises a medical-grade silicone elastomer incorporating a light-blocking dye.Atty. Dkt. No. 125141.05020. MGH2025-380 21. A system for determining a chromophore concentration in tissue, the system comprising:a light source having a characterized emission spectrum spanning a range of wavelengths, the light source being configured to illuminate the tissue;a photodetector having a characterized spectral responsivity, the photodetector being configured to detect light from the light source after propagation through the tissue, thereby producing a measured detected intensity for each light source; anda processor communicatively coupled to the light source and the photodetector, the processor being configured to:cause the light source to emit light towards the tissue thereby illuminating the tissue;receive, from the photodetector, a detected intensity in response to the light illuminating the tissue;adjust the detected intensity to produce an adjusted intensity, based on at least one of:the characterized emission spectrum of the light source;the characterized spectral responsivity of the photodetector; or the tissue attenuation profile of the tissue; anddetermine the chromophore concentrations, based on the adjusted intensity.
22. The system of claim 21, wherein adjusting the detected intensity to produce the adjusted intensity includes the processor being further configured to:Atty. Dkt. No. 125141.05020. MGH2025-380 combine the characterized emission spectrum of the light source with the detected intensity.
23. The system of claim 22, wherein the characterized emission spectrum of the light source has a peak wavelength at an arbitrary intensity unit of 1 and wherein the characterized spectral responsivity of the photodetector has a peak wavelength at an arbitrary intensity unit of 1.
24. The system of claim 22, wherein combining the characterized emission spectrum of the light source with the detected intensity includes the processor being further configured to:determine, for each wavelength of a plurality of wavelengths within the characterized emission spectrum, a modified intensity by multiplying the decimal intensity value at the wavelength by the detected intensity by the decimal value; andadd each modified intensity to determine the adjusted intensity.
25. The system of claim 21, wherein adjusting the detected intensity to produce the adjusted intensity is based on the characterized emission spectrum of the light source, the characterized spectral responsivity of the photodetector, and the tissue attenuation profile of the tissue.
