REAL-TIME OPTOPHYSIOLOGICAL MONITORING METHOD AND SYSTEM - Patent application

Abstract

A method for monitoring a subject using an optical physiological sensor system includes obtaining an optical property model of at least one biological tissue type to be monitored, the optical property model including definitions of static and dynamic components of transmitted optical power and a source-to-detector distance related to a normalized optical path length of an illumination source of the optical physiological sensor. The method also includes obtaining an indication of at least one physiological property of the subject from a wearable device worn by the subject, obtaining an indication of at least one physical variable from the wearable device worn by the subject, determining using the optical physiological model how the at least one physical variable affects the at least one physiological property, and determining a correction value for the physiological property based on the determination of how the at least one physical variable affects the at least one physiological property.

Description

[Technical Field]

[0001] The present invention relates to a method and system for real-time photophysiological monitoring. [Background technology]

[0002] Photoplethysmography (PPG) is a non-invasive optical technique for monitoring changes in blood volume and flow near the skin surface and measuring various physiological parameters. It uses an illumination source and a photodetector to measure the changes in intensity that occur as light passes through or reflects off biological tissue. The detected optical signal is analyzed and correlated with the blood pulsation in the body stimulated by the heartbeat.

[0003] Photophysiological monitoring, including photoplethysmography (PPG) based on the Beer-Lambert law, is applicable to the rapidly growing global mobile health market in developed countries, addressing the increasing prevalence of lifestyle-related chronic diseases and an aging population. This technology has two primary applications: i) real-time, 24 / 7 clinician-directed monitoring to understand vital health parameters in patients with chronic conditions or after surgery; and ii) self-monitoring and assessment to improve sports performance and maintain or improve general health and fitness. Additionally, monitoring and assessment may be used to identify or confirm underlying diseases, or to monitor vital parameters in at-risk subjects to track underlying conditions and provide early warning signals to prevent their deterioration. Summary of the Invention [Problem to be solved by the invention]

[0004] Of the current PPG systems, pulse oximeters (which measure blood oxygen saturation) are the most common, but these are often inaccurate because they do not take into account dynamic changes in the optical properties of living tissue, motion artifacts, or the effects of sweat, skin creams, and sprays.

[0005] In the consumer fitness market, electronic chest-worn heart rate monitoring systems are common. However, these tend to be uncomfortable and intrusive, making them unsuitable for frequent use. Other wearable monitoring systems have emerged on the market, including wristwatch-based sensors that can measure individual parameters such as heart rate, heart rate variability, oxygen saturation, and respiratory rate. However, no wearable sensor currently exists that can simultaneously measure a wider range of physiological parameters that meets the standards for clinical monitoring.

[0006] Therefore, it is an object of the present invention to provide an optophysiological (OP) sensor system including: 1) a multi-wavelength illumination light sensor (optical sensor); 2) custom electronics for driving the optical sensor and capturing signals from the optical sensor; 3) an embedded real-time artificial intelligence (AI) signal processing algorithm for generating physiological measurements; a mobile or web app as a user interface; and 4) a method for monitoring a subject using the optical sensor, which can address the aforementioned problems.

[0007] International Publication No. WO 2015 / 056007 describes an example of a method for assembling an optical physiological sensor. International Publication No. WO 2021 / 233319 describes a method, an electronic device, and a medium for detecting premature beats. International Publication No. WO 2021 / 213337 describes a wearable electronic device, a medium, and a method for monitoring the use of a wearable electronic device, related to the technical field of information processing. U.S. Patent Application Publication No. 2021 / 0393150 describes an apparatus for noninvasively measuring biological information, the apparatus including: a pulse wave sensor configured to measure multiple pulse wave signals having different wavelengths from a subject; a contact pressure sensor configured to measure the contact pressure of the subject while the multiple pulse wave signals are being measured; and a processor configured to acquire an oscillometric waveform based on the multiple pulse wave signals having wavelengths different from the contact pressure, and to acquire biological information based on the oscillometric waveform. U.S. Patent Application Publication No. 2022 / 0015716 describes an electronic device and a method for detecting sleep apnea. U.S. Patent Application Publication No. 2021 / 0330209 describes systems, devices, and methods for tracking one or more physiological indicators of a user (e.g., heart rate, blood oxygen saturation, etc.). U.S. Patent Application Publication No. 2021 / 0236014 describes a finger-worn blood pressure monitor device including a cuff, a tactile sensor array, and control circuitry. U.S. Patent Application Publication No. 2021 / 0196204 describes a method, system, and method for predicting sensor measurement quality. U.S. Patent Application Publication No. 2021 / 0375473 describes a wearable device that can be used for hypertension monitoring. U.S. Patent Application Publication No. 2021 / 0386363 describes an electronic device including a translucent layer forming part of an exterior of the electronic device, an opaque material disposed on the translucent layer and defining micropores, and a processing unit operable to determine information about the user through the translucent layer. US Patent Application Publication No. 2020 / 0297226 describes an electronic fitness device that includes a housing, a first optical transmitter array, a first optical receiver, and a second optical receiver.U.S. Patent Application Publication No. 2021 / 0212620 describes a mobile electronic device used in a vehicle that can determine a user's mental state, such as drowsiness. U.S. Patent Application Publication No. 2021 / 0353165 describes using a wearable device to measure a user's blood pressure, including using sensor data from the wearable device to extract features related to pulse waves, determine a pulse transit time (PTT), scale at least one of the features using the PTT to obtain scaled features, use the scaled features as input for a machine learning (ML) model, and use the output of the ML model to obtain the user's blood pressure. China Patent Application Publication No. 111588385 describes a method and apparatus for measuring oxyhemoglobin saturation. U.S. Patent Application Publication No. 2021 / 0121109 describes a method and system for a light-emitting diode (LED) driving circuit in an optical probe. U.S. Patent Application Publication No. 2021 / 0345899 describes a multi-sensor patch (including a multilayer flexible substrate) for simultaneous abdominal monitoring of maternal and fetal physiological data. International Publication No. WO 2021 / 250224 describes an apparatus and method for estimating one or more hemodynamic parameters, such as cardiac output or stroke volume. U.S. Patent Application Publication No. 2022 / 0000435 describes a method and apparatus for determining a subject's respiratory information. International Publication No. WO 2021 / 101705 describes an apparatus, system, and technique for monitoring a subject's condition. U.S. Patent Application Publication No. 2021 / 0290060 describes a system and method for remote subject management and monitoring. International Publication No. WO 2021 / 194622 describes a method for measuring a user's physiological characteristics. U.S. Patent Application Publication No. 2021 / 0060343 describes a system and method for managing pain in a subject. US Patent Application Publication No. 2021 / 0178164 describes a method and device for pain management with sleep detection.

[0008] Klodell CT, et al., Oximetry-derived perfusion index for intraoperative identification of successful thoracic sympathectomy. Ann Thorac Surg 2005; 80: 467-70 describes that intraoperative PI obtained from pulse oximetry is a new indicator of the success of thoracic sympathectomy in patients with upper extremity hyperhidrosis.

[0009] Park J, Seok HS, Kim SS and Shin H (2022) Photoplethysmogram Analysis and Applications: An Integrative Review. Front. Physiol. 12:808451 is a review paper that aims to examine existing research on photoplethysmograms, particularly from an engineering perspective, including the generation mechanism of photoplethysmograms, measurement principles, clinical applications, noise definition, preprocessing techniques, feature detection techniques, and postprocessing techniques for photoplethysmogram processing. [Means for solving the problem]

[0010] Aspects of the invention are set out in the independent claims, with optional features set out in the dependent claims. Aspects of the invention may be provided in combination with each other and features of one aspect may be applied to other aspects.

[0011] In a first aspect, a method for monitoring a subject using an optophysiological sensor system that takes physical factors into account is described. The optophysiological sensor system includes at least one sensor. The method includes obtaining an optical property model of at least one biological tissue type to be monitored, the optical property model including definitions of static and dynamic components of transmitted optical power and a source-detector separation related to a normalized optical path length of an illumination source of the optophysiological sensor. The method also includes obtaining an indication of at least one physiological property of the subject from a wearable device worn by the subject, the wearable device including the at least one sensor; obtaining an indication of at least one physical variable from the wearable device worn by the subject; determining, using the model, how the at least one physical variable affects the at least one physiological property; and determining a correction value for the physiological property based on the determination of how the at least one physical variable affects the at least one physiological property.

[0012] The at least one sensor may include at least one sensor configured to obtain an indication of at least one physiological characteristic and at least one sensor configured to obtain an indication of at least one physical variable or parameter (e.g., movement). The at least one sensor configured to obtain an indication of the at least one physiological characteristic may be an optical sensor. The optical sensor may be a multi-wavelength optical sensor capable of receiving different wavelength ranges of electromagnetic radiation.

[0013] JPEG2025533435000002.jpg37165

[0014] Obtaining an optical property model for at least one monitored tissue type may include obtaining an optical physiological property model for the at least one monitored tissue type. Obtaining an optical property model for the at least one monitored tissue type may include obtaining the optical property model based on physiological variables of an individual. The physiological variables of the individual may include at least one of the monitored tissue type, tissue / skin age, weight, presence or absence of tattoos, tissue location, and / or skin pigmentation. In this manner, the model may be used to accurately determine a correction value for the physiological property of an individual regardless of age, skin tone, weight, and presence or absence of tattoos.

[0015] Obtaining an indication of at least one physiological characteristic of the subject from a wearable device worn by the subject may include obtaining at least one physiological measurement of the subject from a wearable device worn by the subject, in which case determining how the at least one physical variable affects the at least one physiological characteristic using the model may include determining how the at least one physical variable affects the at least one physiological measurement using the model.

[0016] The definition of source-detector spacing relative to a normalized optical path length of the illumination source of the light sensor may be a definition of source-detector spacing relative to an optimized optical path length of the illumination source of the light sensor.

[0017] The method further includes determining accurate physiological parameter measurements using the corrected values ​​of the physiological characteristics, the physiological parameters including at least one of heart rate (HR), perfusion index (PI), oxygen saturation (SpO2%), respiratory rate (RR), blood pressure (BP), pulse wave variability (PRV), pulse transit time (PTT), and pulse wave velocity (PWV).

[0018] JPEG2025533435000003.jpg12164

[0019] JPEG2025533435000004.jpg62165

[0020] JPEG2025533435000005.jpg30165

[0021] JPEG2025533435000006.jpg28166

[0022] JPEG2025533435000007.jpg36165

[0023] JPEG2025533435000008.jpg20165

[0024] JPEG2025533435000009.jpg27165

[0025] JPEG2025533435000010.jpg22166

[0026] JPEG2025533435000011.jpg78165

[0027] In some examples, the at least one physical variable includes temperature, and using the model to determine how the at least one physical variable affects the at least one physiological characteristic includes modeling the density of a particular tissue type as a function of the wavelength and temperature of an illumination light source of the optical physiological sensor.

[0028] JPEG2025533435000012.jpg69165

[0029] JPEG2025533435000013.jpg64166

[0030] JPEG2025533435000014.jpg21165

[0031] JPEG2025533435000015.jpg32166

[0032] JPEG2025533435000016.jpg38165

[0033] JPEG2025533435000017.jpg53166

[0034] The microelectromechanical sensor may include at least two of a gyroscope, an accelerometer, and a digital compass.

[0035] The optoelectronic sensor panel may be separated from the planar pressure sensor by a flexible substrate.

[0036] The planar pressure sensor may be separated from the printed circuit board by a flexible substrate.

[0037] The printed circuit board needs to be flexible.

[0038] The optoelectronic sensor panel includes a plurality of temperature sensors configured to obtain information indicative of the condition of the wearer's skin through contact with the wearer's skin.

[0039] The photoelectric sensor panel includes a plurality of temperature sensors equally spaced on the photoelectric sensor panel.

[0040] The processor may be configured to transmit raw sensor data or real-time processed data via a wireless communication interface.

[0041] The processor is configured to apply a machine learning model to the sensor data obtained from the plurality of photodiodes, the pressure data obtained from the planar pressure sensor, and the physical variables to generate modified output data, and to transmit the modified output data via the wireless communication interface.

[0042] The processor is configured to not acquire sensor data from the plurality of photodiodes or micromechanical sensors if the temperature sensor indicates a temperature outside a selected temperature range.

[0043] JPEG2025533435000018.jpg16165

[0044] In another aspect, a method for monitoring a subject using an optical physiological sensor system is provided. The method is particularly suitable, for example, for use when the subject is engaged in vigorous exercise. The method includes acquiring, from a wearable device worn by the subject, a signal including a component indicative of at least one physiological characteristic of the subject and a component indicative of at least one physical parameter; modeling the acquired signal as a Taylor series, which is a linear combination of a pulsatile signal and a motion term, based on an indication of the at least one physical parameter; acquiring a corrected signal by projecting the modeled pulsatile signal onto an orthogonal subspace; and determining the at least one physiological characteristic of the subject using the corrected signal.

[0045] The sensor motion imparts a small time-varying modulation to the received optical signal. Although this modulation is small, it is on the same scale as the pulsation signal. Therefore, to a first approximation, the received optical signal consists of a DC component, the cardiac pulsation signal, and a motion-related signal. The product of the motion-related component and the cardiac component can be neglected to a first approximation.

[0046] JPEG2025533435000019.jpg114166

[0047] JPEG2025533435000020.jpg21165

[0048] JPEG2025533435000021.jpg16164

[0049] The method may further include filtering the pulsatile signal to remove DC components, low frequency baseline shifts, and / or high frequency noise before minimizing the integral.

[0050] The method may further include filtering the signal with a zero-phase bandpass filter before obtaining the corrected pulsatile signal.

[0051] Obtaining a signal from a wearable device worn by the subject may include obtaining a plurality of pulsating signals from the wearable device, each of the plurality of pulsating signals being an optical signal of a discrete wavelength.

[0052] JPEG2025533435000022.jpg37165

[0053] Obtaining an indication of at least one physical parameter from a wearable device worn by the subject may include receiving a plurality of signals indicative of the respective physical parameters from the wearable device, and normalizing the received plurality of signals between 0 and 1 to avoid physical unit incompatibilities.

[0054] In another aspect, a method for real-time monitoring of a subject using an optical physiological sensor system is provided. The method is useful for rapid, real-time data processing. The method includes acquiring, from a wearable device worn by the subject, a pulsating signal from an optical sensor indicative of at least one physiological characteristic of the subject; acquiring, from the wearable device worn by the subject, an indication of at least one physical variable; identifying a dominant noise source in the pulsating signal based on the acquired indication of the at least one physical variable; determining a selected noise reduction model to apply to the pulsating signal from a list of selected models to apply; and determining a correction value for the physiological characteristic using the selected noise reduction model.

[0055] The list of selected models includes orthogonal separation, acceleration and / or balance based models, temperature based models, contact pressure based models.

[0056] Identifying primary noise sources in the pulsating signal based on the acquired indication of the at least one physical variable may include determining whether the acquired indication of the at least one physical variable exceeds a selected threshold, and determining a selected noise reduction model to apply to the pulsating signal from a list of selected models to apply may include selecting a noise reduction model based on a determination of whether the acquired indication of the at least one physical variable exceeds a selected threshold.

[0057] The method may further include obtaining an optical property model for at least one biological tissue type to be monitored, the optical property model including a definition of static and dynamic components of transmitted optical power and a definition of a source-detector spacing related to a normalized path length of an illumination source of the optical physiological sensor system. Also, determining a corrected value for the physiological property using the selected noise reduction model includes determining a corrected value for the physiological property using the optical physiological property model for the at least one biological tissue type and the selected noise reduction model.

[0058] The physical variables include at least one of contact pressure, temperature, acceleration, angular velocity, and absolute orientation.

[0059] In another aspect, a method for training a machine learning model for use in monitoring a subject with a wearable device including an optophysiological sensor system is provided, the method including: acquiring a pulsatile signal indicative of at least one physiological characteristic of the subject; acquiring an indication of at least one physical variable as a function of time from the wearable device; transmitting data representing the pulsatile signal and the indication of the physical variable via a wireless interface to a remote computing platform (i.e., a workstation or a computing cloud); training a machine learning model at the remote computing platform based on the received data; and transmitting coefficients / parameters from the remote computing platform to the machine learning model embedded in the wearable device after training.

[0060] In the remote computing platform, training the machine learning model based on the received data may include training a generative adversarial network (GAN) including two generators and a discriminator, where one of the two generators is configured to learn a noise-tolerant mapping from an input noisy pulsating signal to a reference signal, and the other generator is configured to learn a noise distribution in a physical variable captured from the wearable device.

[0061] In another aspect, a method for training a machine learning model for use in monitoring a subject with a wearable device including an optical physiological sensor system is provided, the method including acquiring a pulsatile signal indicative of at least one physiological characteristic of the subject, acquiring from the wearable device an indication of at least one physical variable as a function of time, processing the pulsatile signal and the indication of the at least one physical variable with a machine learning model to provide an output indicative of the physiological characteristic, transmitting data representing the pulsatile signal and the indication of the physical variable via a wireless interface to a remote computing platform, training an improved machine learning model at the remote computing platform based on the received data, transmitting the improved trained machine learning model to the wearable device, and processing the pulsatile signal and the indication of the at least one physical variable with the machine learning model using the improved trained machine learning model to provide an output indicative of the physiological characteristic.

[0062] Obtaining a pulsatile signal providing an indication of at least one physiological characteristic of the subject and obtaining an indication of at least one physical variable as a function of time from the wearable device are performed while a user of the wearable device is performing an exercise routine selected by the user, and a machine learning model is trained on a remote computing platform based on the exercise routine.

[0063] In another aspect, a wearable optophysiological sensor system for acquiring physiological characteristics of a wearer and physical variables of the wearer's movements is provided, the optophysiological sensor comprising: an optoelectronic sensor panel including a plurality of photodiodes and a plurality of illumination sources, such as light emitting diodes, where at least selected ones of the plurality of illumination sources, i.e., the light emitting diodes, are configured to emit light at a wavelength different from that of the other illumination sources; a microelectromechanical sensor for acquiring the physical variables, a processor configured to process signals received from the plurality of microelectromechanical sensors and the plurality of photodiodes, and a printed circuit board including a wireless communication interface, where the plurality of illumination sources are arranged at intersections of a plurality of concentric circles, each of the plurality of concentric circles being located at the center of a respective one of the plurality of photodiodes.

[0064] The multiple illumination sources include at least four different illumination sources configured to emit light at different wavelengths. Because two wavelengths of illumination are required to calculate oxygen saturation (SpO2%), using illumination sources with four or more wavelengths is a better option. Furthermore, the responses of multiple photodiodes to illumination at different wavelengths can be considered as cross-correlations of the illumination at each wavelength to determine the optimal pulsation signal obtained from the device.

[0065] The plurality of illumination sources comprises at least 16 different illumination sources configured to emit light at any of four or more different wavelengths, the at least 16 different illumination sources being grouped into four groups configured to emit light at any of the four different wavelengths.

[0066] In some examples, the shortest wavelength illumination source is positioned at the intersection of an inner circle of the plurality of concentric circles, and the longer wavelength illumination source is positioned at the intersection of an outer concentric circle further from the center, with the longer wavelength illumination source being positioned farther away from each photodiode, and the wavelength of the illumination source increasing with radial distance from each photodiode.

[0067] In some examples, the photoelectric sensor panel is made up of one, two, or three photodiodes.

[0068] At least one of the wavelengths of the illumination source is 465 nm.

[0069] JPEG2025533435000023.jpg11165

[0070] JPEG2025533435000024.jpg58166

[0071] JPEG2025533435000025.jpg16166

[0072] JPEG2025533435000026.jpg32165

[0073] In another aspect, there is a computer-readable non-transitory storage medium containing a program for a computer configured to cause a processor to perform the method of any of the above aspects. [Effects of the Invention]

[0074] The present invention addresses the above-mentioned problems. [Brief explanation of the drawings]

[0075] [Figure 1] FIG. 1 shows an example of an optophysiological sensor system. [Figure 2] FIG. 2 shows another example of an optophysiological sensor system. [Figure 3] FIG. 3 shows another example of an optophysiological sensor system. [Figure 4] FIG. 4 shows a functional schematic diagram of an optoelectronic hardware system 400 . [Figure 5] FIG. 5 shows a functional schematic diagram of one example of a method for processing signals obtained from an optoelectronic sensor, such as the optoelectronic system shown in FIGS. 1-3 or the optoelectronic hardware system shown in FIG. [Figure 6] Figure 6 shows the pulse waveform of a subject with skin type I (Fitzpatrick scale) taken from the back of the left wrist using an optophysiological monitoring system. [Figure 7] Figure 7 shows the pulse waveform of a subject with skin type II (Fitzpatrick scale) taken from the back of the left wrist using an optophysiological monitoring system. [Figure 8] Figure 8 shows the pulse waveform of a subject with skin type III (Fitzpatrick scale) taken from the back of the left wrist using an optophysiological monitoring system. [Figure 9] Figure 9 shows the pulse waveform of a subject with skin type IV (Fitzpatrick scale) taken from the back of the left wrist using an optophysiological monitoring system. [Figure 10] Figure 10 shows the pulse waveform of a subject with skin type V (Fitzpatrick scale) taken from the back of the left wrist using an optophysiological monitoring system. [Figure 11] Figure 11 shows the pulse waveform of a subject with skin type VI (Fitzpatrick scale) taken from the back of the left wrist using an optophysiological monitoring system. [Figure 12] Figure 12 shows the pulse waveform of a male subject with skin type VI (Fitzpatrick scale) using four-wavelength photophysiological monitoring. [Figure 13] Figure 13 shows the pulse waveform of a male subject with skin type VI (Fitzpatrick scale) using four-wavelength photophysiological monitoring. DETAILED DESCRIPTION OF THE INVENTION

[0076] JPEG2025533435000027.jpg14978

[0077] Figure 1 illustrates an example of an optophysiological sensor system 100 used to monitor or measure an optophysiological property of a subject. The optophysiological sensor system 100 can be designed as a wearable, such as a patch, a watch, a ring, etc. Figure 2 illustrates another example of an optophysiological sensor system 200, and Figure 3 illustrates yet another example of an optophysiological sensor system 300.

[0078] 1, optophysiological sensor system 100 may include one or more sensors (e.g., including a multispectral / wavelength sensor), electronics for controlling and receiving signals from the sensors, and a computer program running on a processor and including an algorithm configured to process the signals, e.g., according to an optophysiological model described in more detail below. For example, optophysiological sensor system 100 may include a photodiode 101, a temperature sensor 103, and multiple illumination sources 105, 107, 109, and 111, each configured to emit light at a different wavelength. It may also include a MEMS motion sensor 135.

[0079] The optophysiological sensor system 100 is configured to obtain at least one physiological measurement of a subject, which measurement can provide an indication of at least one physiological characteristic of the subject. The optophysiological sensor system 100 may be configured to obtain an indication of at least one physical variable related to the wearer's movements. By accounting for these movements in an optophysiological model, more accurate optophysiological measurements can be obtained, such as heart rate (HR), perfusion index (PI), oxygen saturation (SpO2%), respiratory rate (RR), blood pressure (BP), pulse wave variability (PRV), pulse transit time (PTT), and pulse wave velocity (PWV).

[0080] In the example shown in Figure 1, the optoelectronic sensor system 100 includes an optoelectronic sensor panel 121 that includes three photodiodes 101, while in the example shown in Figure 2 it includes two photodiodes 102, and in the example shown in Figure 3 it includes only one photodiode 101. It also includes a plurality of illumination sources 105-111 (24 in the example of Figure 1, 24 in the example of Figure 2, and 16 in the example of Figure 3), which are illumination sources in this example, and optionally at least one temperature sensor 103 (4 in the examples of Figures 1 to 3).

[0081] JPEG2025533435000028.jpg64166

[0082] Thus, the microelectromechanical sensor 135,401 and microcontroller processor are located on the opposite side of the optoelectronic sensor from the photodiode, temperature sensor and illumination source.

[0083] In the illustrated example, the microelectromechanical sensor 135, 401 includes a three-axis gyroscope, a three-axis accelerometer, and a three-axis digital compass. As shown in FIG. 1, the optoelectronic sensor panel 121 is separated from the planar pressure sensor 123 by a first flexible substrate 127, and the planar pressure sensor 123 is separated from the printed circuit board 125 by a second flexible substrate 129.

[0084] The photoelectric sensor panel 121 includes a plurality of temperature sensors 103 (four in this example) configured to contact the wearer's skin and obtain information indicative of the condition of the wearer's skin. The plurality of temperature sensors 103 are evenly spaced on the photoelectric sensor panel 121.

[0085] At least some of the illumination sources 105-111 are configured to emit light at a different wavelength than the other illumination sources. The wavelength may be, for example, 525 nm, 595 nm, 650 nm, 910 nm, or other wavelengths. As described in more detail below, the use of multiple different wavelengths (four in this example) can accommodate a variety of situations, such as different tissue types, tissue locations, and skin pigmentation. In some instances, a wavelength of 475 nm may be preferred.

[0086] 1 and 2, there are six light-emitting diodes of each wavelength, and since there are four or more different wavelengths, there are a total of 24 or more light-emitting diodes. As shown in FIGS. 1 and 2, the light-emitting diodes 105-111 are located at the intersections of multiple concentric circles, and each concentric circle is located at the center of a respective one of the multiple photodiodes 101. In the case of a single photodiode, as shown in FIG. 3, the light-emitting diodes 105-111 are arranged at equal intervals on the circumference of the concentric circle centered on the photodiode 101, but the light-emitting diodes on adjacent circles can be arranged so that they are offset by 90° to increase the separation between the illumination light sources, for example, the separation between the light-emitting diodes on adjacent circles.

[0087] In the examples of FIGS. 1 and 2, the light-emitting diode with the shortest wavelength (wavelength 1) is positioned at the intersection of the inner circle of the multiple concentric circles, and the illumination sources, e.g., light-emitting diodes, with longer wavelengths (wavelengths 2-4, with 4 being the longest and 1 being the shortest) are positioned at intersections away from the center of the concentric circles. The longer wavelength illumination sources, e.g., light-emitting diodes 105-111, are positioned farther away from each photodiode, such that the wavelengths of the illumination sources, e.g., light-emitting diodes, increase with radial distance from each photodiode 101. Thus, in the examples of FIGS. 1 and 2, the light-emitting diode with wavelength 1 is positioned on the inside, followed by the light-emitting diodes with wavelengths 2 and 3, and finally the light-emitting diode with wavelength 4 on the outside. In the example of FIG. 3, the illumination sources (e.g., light-emitting diodes) are simply positioned such that the light-emitting diodes with longer wavelengths are positioned farther away from the photodiode 101, and the wavelengths of the illumination sources (e.g., light-emitting diodes) increase with radial distance from the photodiode 101. Preferably, at least one wavelength is blue light, for example 465 nm, which has surprisingly been found to be particularly effective for measuring SpO2 in certain tissue types and locations.

[0088] The microcontroller processor (MCU) 405 is configured to process the sensor data obtained from the plurality of photodiodes 101, the pressure data obtained from the planar pressure sensor 123, and the physical variables obtained from the MEMS motion sensor 135 by applying an optophysiological model to the data (described in more detail below) to obtain an indication of a physiological characteristic of the subject / wearer of the optoelectronic sensor 100.

[0089] In some examples, the microcontrol processor 405 may be configured to do this by transmitting raw sensor data or real-time processed data over a wireless communication interface. The microcontrol processor 405 may be configured to apply a machine learning model to the sensor data obtained from the plurality of photodiodes 101, the pressure data obtained from the planar pressure sensor 123, and the physical variables to generate modified output data and transmit the modified output data over the wireless communication interface 413.

[0090] In some examples, if the temperature sensor 103 indicates that the temperature (T) of the subject's contacted skin / organ tissue is outside a selected temperature range (18-370°C), the microcontrol processor 405 is configured to not acquire sensor data from the plurality of photodiodes 101 or the micromechanical sensor 135. Additionally or alternatively, in some examples, the microcontrol processor 405 is configured to not acquire sensor data from the plurality of photodiodes or the micromechanical sensor if the planar pressure sensor 123 indicates a contact pressure outside a determined pressure range.

[0091] While the LED for wavelength 1 is fully lit, the photodiode begins capturing backlight from the contact skin tissue, then turns on the LED for wavelength 2 and captures the backlight again using the same procedure as above. The LEDs for wavelengths 3 and 4 perform the same operation as above. One complete duty cycle of backlight acquisition for the four wavelength illumination sources (LEDs and background signals captured by the same photodiode) is set as one sampling point consisting of five individual numbers. This achieves the expected effect of simultaneous sampling of multiple wavelengths. There are five channel signals: four wavelength illuminations and one background, captured by the photodiode of the multi-wavelength illumination photoelectric sensor.

[0092] As explained above, an optophysiological model is applied to process the data received by the optoelectronic sensor 100 and obtain useful physiological parameters of the subject. In conventional systems, the Beer-Lambert model has been applied to correlate the absorption of light with the properties of the material through which the light passes. However, such an approach has proven to be too simplistic and not truly representative when attempting to establish the optophysiological properties of biological tissue. Therefore, an optophysiological model that takes into account the various properties of the biological tissue being monitored is needed.

[0093] In this case, the optophysiological model includes definitions of the static and dynamic components of the transmitted optical power, as well as a definition of the source-detector separation related to the normalized optical path length of the illumination source (e.g., a light-emitting diode) of the optophysiological sensor.

[0094] The at least one physical variable includes at least one of contact pressure, temperature, acceleration, angular velocity, and absolute orientation.

[0095] JPEG2025533435000029.jpg11165

[0096] JPEG2025533435000030.jpg11165

[0097] JPEG2025533435000031.jpg17166

[0098] JPEG2025533435000032.jpg11163

[0099] JPEG2025533435000033.jpg17166

[0100] JPEG2025533435000034.jpg22166

[0101] JPEG2025533435000035.jpg30166

[0102] JPEG2025533435000036.jpg30166

[0103] JPEG2025533435000037.jpg36165

[0104] JPEG2025533435000038.jpg20166

[0105] JPEG2025533435000039.jpg22166

[0106] As mentioned in the introduction, the optical properties of tissue change with acceleration (m 2 / s), angular velocity (° / s), body temperature regulation (℃), contact force (N / m 2 ) are also governed or influenced by

[0107] JPEG2025533435000040.jpg22153

[0108] JPEG2025533435000041.jpg11167

[0109] JPEG2025533435000042.jpg39165

[0110] JPEG2025533435000043.jpg16166

[0111] JPEG2025533435000044.jpg23149

[0112] JPEG2025533435000045.jpg16166

[0113] JPEG2025533435000046.jpg21167

[0114] Because capillary and peripheral blood flow is controlled by major and branch arteries, the composition and density of these capillary and peripheral blood changes in response to changes in blood flow from the local and global cardiac system. These phenomena can be accurately reflected by the multi-wavelength illumination photoelectron sensor, which captures dynamic systolic and diastolic behavior and different wavelength illumination.

[0115] Tissue temperature (°C) is a direct result of blood perfusion, defined as thermoregulation, and reflects changes in tissue composition, skin thickness, surface area, tissue volume, and ambient temperature in the presence of living tissue or the environment surrounding the living tissue.

[0116] JPEG2025533435000047.jpg21151

[0117] JPEG2025533435000048.jpg11166

[0118] JPEG2025533435000049.jpg42165

[0119] JPEG2025533435000050.jpg17165

[0120] JPEG2025533435000051.jpg27166

[0121] JPEG2025533435000052.jpg43166

[0122] JPEG2025533435000053.jpg38166

[0123] The pulse amplitude reaches its maximum depending on the contact force, indicating that the actual force acting on the arterial wall varies depending on the individual. The maximum pulse amplitude ranges, for example, from 0.15N to 1.5N. The sensor system detects the contact force signals and processes them individually based on an AI driving training strategy.

[0124] JPEG2025533435000054.jpg11166

[0125] FIG. 4 is a functional schematic diagram illustrating an example of an optophysiological sensor system, which may be an optoelectronic hardware system 400. The optophysiological sensor system 400 shown in FIG. 4 includes a sensor module 401 connected to a microcontroller unit (MCU) 405 via an analog front-end 403. The analog front-end 403 is composed of a multiplexing LED driver, an amplifier, a bandpass filter, a bandstop filter, and a demultiplexer. Features of the optophysiological sensor systems shown in FIGS. 1 to 3 are applicable to the optophysiological sensor system 400 shown in FIG. 4, and vice versa. For example, the sensor module 401 may include a photodiode 101, a temperature sensor 103, multiple illumination sources 105, 107, 109, and 111, and a MEMS motion sensor 135 shown in FIGS. 1 to 3.

[0126] The sensor module 401, the analog front end 403, and the microcontroller 405 are connected to a power management module 411, which is connected to a battery 417. The microcontroller unit 405 is also connected to an optional debugger 407 and an optional computer 409. A wireless communication interface 413 is connected in parallel to the power management module 411 and the microcontroller 405. The wireless communication interface 413 is configured to communicate with a cloud server and a mobile app 415.

[0127] The sensor module 401 comprises: 1. Multi-wavelength illumination photoelectric sensor (called photoelectric sensor) 2. MEMS motion sensor with one 3-axis gyroscope, one 3-axis accelerometer, and one 3-axis digital compass 3. One planar pressure sensor 4. Four temperature sensors

[0128] The signal captured by the sensor module 401 is split into two parts. 1. The first part is that the five-channel signals captured by the photodiodes are amplified in the first stage of the Analog Front-End Electronics (AFE) 403, filtered by a combination of bandpass and bandstop filters, and then passed to the MCU 405. 2. The second part is the reference signals acquired by the planar pressure sensor, the four temperature sensors, and the MEMS sensor unit. These reference signals are directly sent to the MCU 405 after primary amplification.

[0129] The MCU405 serves four roles. 1. Analog to Digital Conversion (ADC) 2. Control and drive of AFE403 3. Real-time signal processing incorporating AI algorithms 4. Transmitting real-time data / readings via WI-FI, Bluetooth, 4G / 5G LTE wireless communication interface 413

[0130] Thus, the MCU 405 may include computer programs and / or instructions configured to receive and process sensor signals according to an optophysiological model, as described above.

[0131] These processed or unprocessed signals are transmitted to a mobile app 405 via WI-FI, near field communication (such as Bluetooth), or 4G / 5G LTE.

[0132] The wireless communication interface 413 has two functions and is configured to communicate over the air (OTA). 1. Remote communication mode: upload the signal to the cloud server and send it to the mobile app. 2. Update control and driver systems, AI signal processing algorithms

[0133] The power management module 411 is responsible for power management of the entire system, including all voltage conversion and battery charge / discharge management.

[0134] The debugger module 407 can be connected to a PC via USB Type-C to update the control and driver system, AI signal processing algorithms, and debug system errors.

[0135] FIG. 5 shows a functional schematic diagram of an example of a method for processing signals obtained from an optoelectronic sensor via the optophysiological sensor system shown in FIGS. 1-3 or the optophysiological sensor system shown in FIG.

[0136] In step 501, raw signals are acquired by a multi-wavelength illumination optoelectronic sensor (optoelectronic sensor or photodiode). All raw signals are normalized between 0 and 1. The raw signals may then undergo pre-processing 503 (e.g., orthogonal separation processing, which will be described in more detail below).

[0137] The raw signal is input to the offline AI 507 and the real-time AI 505 in parallel.

[0138] Real-time AI505 is a generative adversarial network (GAN) framework consisting of two generators and one discriminator. Generator 1 (G1) aims to learn a noise-tolerant mapping from a noisy photoelectron signal to a reference signal. Generator 2 (G2) aims to learn the noise distribution obtained from a 3-axis accelerometer, a 3-axis gyro sensor, a temperature sensor, and a pressure sensor. These two generators G1 and G2 share a discriminator (D).

[0139] JPEG2025533435000055.jpg16167

[0140] JPEG2025533435000056.jpg70146

[0141] JPEG2025533435000057.jpg17166

[0142] JPEG2025533435000058.jpg130166

[0143] JPEG2025533435000059.jpg42167

[0144] In this step, the training strategy described above is repeated to test the model (validation) until stability is established.

[0145] Real-time datasets: Multi-wavelength illumination photoelectron sensors are attached to subjects to create real-time datasets. Data is collected at sampling rates exceeding 1 kHz on each channel of the photoelectron sensor during cycling, walking, jogging, running, and other everyday physical activities, consistent with clinical monitoring standards.

[0146] Configuration: GANs are implemented in TensorFlow and trained on an NVIDIA RTX 3080Ti or equivalent. All CNNs are trained for 200 epochs with random initialization, 0 weight decay, 0.0001 learning rate, batch size 8, and no learning rate decay strategy with Adam optimizer at 0.5 and 0.99.

[0147] Real-time AI505 works as follows: (inference / embedded on MCU) Inference Procedure: G1 can be easily ported to embedded sensor systems.

[0148] Once the offline GAN 507 (training and testing) via the cloud computing program / algorithm completes the training procedure, learning weights are generated using the real-time dataset provided by the sensor system. The learning weights are shared with the inference procedure of the GAN model. This establishes inference in the embedded MCU 405 and is used to generate the desired photoelectron signal using the input noisy photoelectron signal.

[0149] It is therefore understood that the microcontroller unit 405 acquires pulsation signals indicative of at least one physiological characteristic of the subject, acquires an indication of at least one physical variable as a function of time from the wearable device, transmits the data representing the pulsation signals and the indication of the physical variable via the wireless communication interface 413 to a remote computing platform, i.e., a workstation or a computing cloud as shown in FIG. 4, trains a machine learning model based on the received data at the remote computing platform, and after training transmits the coefficients / parameters from the remote computing platform to the embedded machine learning model in the wearable device.

[0150] It will therefore also be appreciated that the microcontroller unit 405 can be trained by obtaining a pulsation signal indicative of at least one physiological characteristic of the subject, obtaining an indicator of the at least one physical variable as a function of time from the wearable device, processing the pulsation signal and the indicator of the at least one physical variable using a machine learning model to provide an output indicative of the physiological characteristic, transmitting data representing the pulsation signal and the indicator of the physical variable via a wireless communication interface to a remote computing platform, training at the remote computing platform an improved machine learning model based on the received data, transmitting the improved trained machine learning model to the wearable device, and using the improved trained machine learning model to process the pulsation signal and the indicator of the at least one physical variable using the machine learning model to provide an output indicative of the physiological characteristic.

[0151] In the case of vigorous activity, the signals are pre-processed (503) by orthogonal separation as described below. It is understood that in some examples, the microcontroller unit 405 is configured to determine whether the wearer / subject is engaged in vigorous activity, for example, based on data from the sensor module 401. For example, the microcontroller unit 405 may determine that the wearer / subject is engaged in vigorous activity if the sensor module indicates a physical quantity (e.g., acceleration) outside a selected range or above a selected threshold level.

[0152] JPEG2025533435000060.jpg37166

[0153] JPEG2025533435000061.jpg16166

[0154] JPEG2025533435000062.jpg54166

[0155] JPEG2025533435000063.jpg54166

[0156] JPEG2025533435000064.jpg17166

[0157] JPEG2025533435000065.jpg55166

[0158] JPEG2025533435000066.jpg60166

[0159] This process can be performed very quickly (<<1 second), making it ideal for real-time physiological monitoring systems. This method is accurate when the true pulsatile and accelerometer signals occupy different regions of the frequency domain. On the other hand, even when the signals overlap in frequency space, this method is highly effective for extracting biomarkers, since the peaks of the forced frequencies are much sharper than those of the pulsatile signal (mainly due to frequency spreading caused by the presence of surrounding soft tissue).

[0160] Although orthogonal separation in this case reduces the height of the broad peaks of the pulsation frequencies, it does not completely eliminate them. Even in these rare cases, the signal quality is sufficient to obtain heart rate (HR), respiratory rate (RR), oxygen saturation (SpO2) levels, pulse transit time (PTT), pulse wave velocity (PWV), and other biomarkers.

[0161] JPEG2025533435000067.jpg38166

[0162] JPEG2025533435000068.jpg38166

[0163] JPEG2025533435000069.jpg21166

[0164] This is especially important when SpO2 measurements do not work most accurately, such as in cold conditions measured by the device's temperature sensor, or on darker skin tones.

[0165] It will be understood that such biomarkers can be obtained without the need for a wearable device, and can be measured using means such as a camera that can receive and distinguish light of specific wavelengths.

[0166] In some examples, the microcontroller unit 405 is configured to determine a dominant noise or a dominant form of noise, for example, based on a physical variable (e.g., acquired via the sensor module 401), and apply a model selected based on the dominant noise or the dominant form of noise to remove noise in the pulsating signal. For example, if the dominant noise is due to temperature, the microcontroller unit 405 may be configured to apply a temperature-based model. In some examples, the model may be applied only when the noise reaches a selected threshold level, for example, when noise from a particular source exceeds a selected threshold level. Applying a model only to the dominant noise or the dominant form of noise, rather than applying all models all the time, may advantageously improve real-time processing of the signal. This may advantageously improve real-time output of physiological parameters.

[0167] As mentioned above, in some examples, the microcontroller unit 405 is configured to not acquire data if a parameter is outside a selected threshold range. For example, if a temperature measurement is outside a selected range or if a contact pressure is outside a selected range, the microcontroller unit 405 is configured to not acquire data and / or transmit data via the wireless communication interface 413.

[0168] As described above, the optophysiological sensor and monitoring method includes an automatic attenuation feature that automatically selects the appropriate illumination intensity for optoelectronic sensor operation and is suitable for accommodating a variety of physiological variables, such as tissue type, age, weight, tissue location, presence or absence of tattoos, and skin tone / pigmentation.

[0169] The performance of the photophysiological monitoring system was tested in six healthy volunteers. System performance was evaluated across tissue types, tissue locations, and skin pigmentation. An analog front end driving multiplexed LEDs (four or more wavelengths: 525 nm (green), 595 nm (orange), 650 nm (red), and 910 nm (infrared)) provides high-resolution pulsating waveforms (<1 ms) to capture changes in light transmission through tissue. These wavelengths used are merely examples; other wavelengths, such as those in the 465-1450 nm range, can also be used. For example, the wavelengths used can be based on individual physiological variables (tissue type, age, weight, tissue location, presence or absence of tattoos, skin tone / pigmentation, etc.). For example, different wavelengths may be used for newborns with lighter skin tones and adults with darker skin tones.

[0170] Figure 6 shows the Fitzpatrick scale (Source: Hypoxia Lab, "Protocol for skin color assessment for pulse oximeter performance studies," August 2022, https: / / openoximetry.org / study-protocols). The photophysiological monitoring system was tested on skin types I to VI (skin type was determined using the Fitzpatrick scale in Figure 6). The results shown in Figures 7-12 are from six healthy volunteers: four men (skin types I, III, VI, VI) and two women (skin types II and V). Each subject completed a one-minute measurement at room temperature (20 ± 3°C) while resting.

[0171] 6-12 show pulsatile waveforms captured from illumination at four wavelengths by an optophysiological sensor of an embodiment of the present invention, such as the optophysiological sensor of FIGS. 1-3.

[0172] Figure 6 shows the pulse waveform of a subject with skin type I (Fitzpatrick scale) collected from the back of the left wrist using an optophysiological monitoring system.

[0173] Figure 7 shows the pulse waveform of a subject with skin type II (Fitzpatrick scale) taken from the back of the left wrist using an optophysiological monitoring system.

[0174] Figure 8 shows the pulse waveform of a subject with skin type III (Fitzpatrick scale) taken from the back of the left wrist using an optophysiological monitoring system.

[0175] Figure 9 shows the pulse waveform of a subject with skin type IV (Fitzpatrick scale) taken from the back of the left wrist using an optophysiological monitoring system.

[0176] Figure 10 shows the pulse waveform of a subject with skin type V (Fitzpatrick scale) taken from the back of the left wrist using an optophysiological monitoring system.

[0177] Figure 11 shows the pulse waveform of a subject with skin type VI (Fitzpatrick scale) taken from the back of the left wrist using an optophysiological monitoring system.

[0178] Figure 12 shows the pulse waveform of a subject with skin type V (Fitzpatrick scale) taken from the back of the left wrist using an optophysiological monitoring system.

[0179] The SpO2 measurement results using the photophysiological monitoring system described in Figure 1 were compared with those of the Masimo LNCS® TFA-1™ disposable transflective sensor and the NONIN 8000R reflective SpO2 sensor, as shown in Table 1 below. The measurement site in this study was the forehead. Skin type was determined using the Fitzpatrick scale. Healthy volunteers with skin types II, III, V, and VI participated in the study. Each subject completed a one-minute measurement at room temperature (20±3°C) while resting.

[0180] Table 1: Comparison of SpO2 results for an optical physiological monitoring system, a Masimo LNCS TFA-1™ disposable transmissive-reflective sensor, and a NONIN 8000R reflective SpO2 sensor.

[0181] [Table 1]

[0182] Comparison results of the perfusion index (PI) of the optical physiological monitoring system described above with respect to Figure 1, compared with the NONIN 8000R reflectance SpO2 sensor and the Masimo LNCS TFA-1™ disposable transflectance sensor, are shown in Table 2. Measurement sites for this test were the wrist, index finger, and forehead. Skin type was determined using the Fitzpatrick scale. Healthy volunteers with skin types II, III, V, and VI were tested. Each subject was measured for 1 minute at room temperature (20 ± 3°C) while at rest. The results shown in Figure 8 are from a healthy volunteer with skin type VI.

[0183] Figure 13 shows the pulse waveform of a male subject's skin type VI (Fitzpatrick scale) measured by four-wavelength photophysiological monitoring. The subject was at rest at room temperature (20±2°C) and the measurement was performed on the back of the left wrist.

[0184] Table 2: Average PI (%) results of the disclosed optophysiological sensor for four subjects (skin types II, III, V, VI) and two commercially available clinical-grade pulse oximetry probes.

[0185] [Table 2]

[0186] The embodiments shown in the drawings are merely exemplary and include features that may be generalized, omitted, or substituted as described and claimed herein. In the context of this disclosure, other examples and variations of the apparatus and methods described herein will be apparent to those skilled in the art.

Claims

1. 1. A method of monitoring a subject using an optical physiological sensor system, comprising: the optophysiological sensor system comprises at least one sensor; obtaining an optical property model of at least one biological tissue type to be monitored, the optical property model including definitions of static and dynamic components of transmitted optical power and a source-detector separation related to a normalized optical path length of an illumination source of an optical physiological sensor; obtaining an indication of at least one physiological characteristic of the subject from a wearable device worn by the subject, the wearable device including at least one sensor; obtaining an indication of at least one physical variable from a wearable device worn by the subject, the wearable device including at least one sensor; using a model to determine how said at least one physical variable affects at least one physiological characteristic; determining a correction value for the physiological characteristic based on the determination of how the at least one physical variable affects the at least one physiological characteristic; Equipped with method.

2. 10. The method of claim 1, determining an accurate physiological parameter measurement using the corrected value of the physiological characteristic; The physiological parameters include heart rate (HR), perfusion index (PI), oxygen saturation (SpO ２ %), respiratory rate (RR), blood pressure (BP), pulse wave variability (PRV), pulse transit time (PTT), pulse wave velocity (PWV), method.

3. 10. The method of claim 1, The at least one physical variable includes at least one of contact pressure, contact temperature, acceleration, angular velocity, and absolute orientation. method.

4.

5.

6.

7.

8.

9.

10. 10. The method according to any one of claims 1 to 9, the at least one physical variable includes at least one of acceleration, angular velocity, and absolute orientation; and determining how the at least one physical variable affects the at least one physiological characteristic using the model includes modeling the pumping action of the heart to determine volumetric blood flow. method.

11.

12. 12. The method according to any one of claims 1 to 11, the at least one physical variable includes temperature; and determining, using the model, how the at least one physical variable affects the at least one physiological characteristic includes modeling the density of a particular tissue type as a function of a wavelength and a temperature of an illumination source of an optical physiological sensor. method.

13.

14.

15. 15. The method of any one of claims 1 to 14, The at least one physical variable includes contact pressure between an optical physiological sensor and the subject's skin, and using the model to determine how the at least one physical variable affects at least one physiological characteristic includes modeling changes in optical density of a particular tissue type based on changes in optical path length. method.

16. 16. The method of claim 15, Using the model to determine how the at least one physical variable affects the at least one physiological property includes modeling the optical density of a particular tissue type as a function of wavelength and temperature proportional to contact force when the contact force is within a selected range, and modeling the optical density of a particular tissue type as a function of wavelength and temperature proportional to the inverse of contact force when the contact force is outside the selected range. method.

17.

18. 1. A wearable optical physiological sensor system for acquiring physiological characteristics of a wearer and physical variables of the wearer's movements, comprising: The optical physiological sensor system comprises: an optoelectronic sensor panel including a plurality of photodiodes and a plurality of illumination sources, at least selected ones of the plurality of illumination sources being configured to emit light at a wavelength different from the other illumination sources, e.g., light emitting diodes; a planar pressure sensor for obtaining an indication of contact pressure between the optophysiological sensor and the wearer; a printed circuit board comprising: a microelectromechanical sensor for acquiring the physical variable; a microcontroller processor configured to process signals received from a plurality of the microelectromechanical sensors and the plurality of photodiodes; and a wireless communication interface; Equipped with Optophysiological sensor system.

19. 20. The optophysiological sensor system of claim 18, The microelectromechanical sensor includes at least two of a gyroscope, an accelerometer, and a digital compass. Optophysiological sensor system.

20. 20. The optophysiological sensor system of claim 18 or 19, the optoelectronic sensor panel being separated from the planar pressure sensor by a flexible substrate; Optophysiological sensor system.

21. 21. The optophysiological sensor system of claim 18, wherein: the planar pressure sensor is separated from the printed circuit board by a flexible substrate; Optophysiological sensor system.

22. 22. The optophysiological sensor system of claim 18, the printed circuit board is flexible; Optophysiological sensor system.

23. 23. The optophysiological sensor system of any one of claims 18 to 22, the photoelectronic sensor panel comprising a plurality of temperature sensors configured to contact the wearer's skin and obtain information indicative of a condition of the wearer's skin; Optophysiological sensor system.

24. 24. The optophysiological sensor system of claim 23, the photoelectric sensor panel includes a plurality of the temperature sensors equally spaced on the photoelectric sensor panel; Optophysiological sensor system.

25. 25. The optophysiological sensor system of any one of claims 18 to 24, the processor is configured to transmit raw sensor data or real-time processed data via the wireless communication interface; Optophysiological sensor system.

26. 26. The optophysiological sensor system of any one of claims 18 to 25, the processor is configured to apply a machine learning model to the sensor data acquired from the plurality of photodiodes, the pressure data acquired from the planar pressure sensor, and the physical variables to generate modified output data, and to transmit the modified output data via the wireless communication interface. Optophysiological sensor system.

27. 27. The optophysiological sensor system of any one of claims 18 to 26, and wherein the processor is configured to not acquire sensor data from the plurality of photodiodes or micromechanical sensors if the temperature sensor indicates a temperature outside a selected temperature range. Optophysiological sensor system.

28. 28. The optophysiological sensor system of any one of claims 18 to 27, the microcontroller processor is configured to not acquire sensor data from the plurality of photodiodes or the micromechanical sensor if the planar pressure sensor indicates a contact pressure outside a predetermined pressure range. Optophysiological sensor system.

29. 1. A method for monitoring a subject with an optophysiological sensor system, comprising: acquiring, from a wearable device worn by the subject, a signal including a component related to at least one physiological characteristic of the subject and a component indicative of at least one physical parameter; modeling the acquired signal as a Taylor series, which is a linear combination of a pulsatile signal and a kinetic term, based on an indication of the at least one physical parameter; obtaining a correction signal as a projection of the modeled pulsation signal onto an orthogonal subspace; determining the at least one physiological characteristic of the subject using the corrected signal; A method comprising:

30.

31. 31. The method of claim 30, The integration is performed over a rolling time window, method.

32. 32. The method of any one of claims 29 to 31, the signal includes an AC component and a DC component, and obtaining the correction signal includes obtaining a projection of the measured AC signal onto an orthogonal subspace. method.

33. 30. The method of any one of claims 26 to 29, further comprising filtering the signal to remove DC components, low frequency baseline shifts, and / or high frequency noise before minimizing the integral. method.

34. 34. The method of any one of claims 29 to 33, filtering the signal with a zero-phase band-pass filter before obtaining the corrected signal. method.

35. 35. The method of any one of claims 29 to 34, acquiring a signal from the wearable device worn by the subject includes acquiring a plurality of pulsation signals from the wearable device, each of the plurality of pulsation signals being an optical signal of a discrete wavelength; method.

36. 36. The method of claim 35, Further, performing independent component analysis or principal component analysis on each of the plurality of pulsation signals. method.

37. 35. The method of any one of claims 29 to 34, obtaining an indication of the at least one physical parameter from the wearable device worn by the subject includes receiving a plurality of signals indicative of respective physical parameters from the wearable device; and normalizing the received signals between 0 and 1 to avoid physical unit incompatibilities. method.

38. 1. A method for monitoring a subject with an optophysiological sensor system, comprising: acquiring, from a wearable device worn by a subject, a pulsatile signal from an optical sensor that provides an indication of at least one physiological characteristic of the subject; obtaining an indication of at least one physical variable from the wearable device worn by the subject; identifying a dominant noise source in the pulsation signal based on the obtained indication of the at least one physical variable; determining a selected denoising model to apply to the pulsation signal from a list of selected models to apply; determining a corrected value of a physiological characteristic using the selected denoising model; A method comprising:

39. 39. The method of claim 38, The list of selected models may include: Orthogonal separation, Acceleration and / or balance based models; Temperature-based models, contact pressure-based model, Including, method.

40. 40. The method of claim 38 or 39, determining a dominant noise source in the pulsation signal based on the acquired indicator of at least one physical variable includes determining whether the acquired indicator of the at least one physical variable exceeds a selected threshold; determining a selected denoising model to apply to the pulsation signal from a list of selected models to apply includes selecting a denoising model based on a determination of whether an index of the acquired at least one physical variable exceeds a selected threshold; method.

41. 41. The method of any one of claims 38 to 40, further comprising obtaining an optical property model of the at least one biological tissue type to be monitored; the optical property model includes definitions of static and dynamic components of transmitted optical power and a definition of a source-detector spacing related to a normalized path length of an illumination source of the optical physiological sensor system; determining a correction value for the physiological property using the selected denoising model includes determining a correction value for the physiological property using the optical property model of at least one biological tissue type and the selected denoising model; method.

42.

43. 1. A method of training a machine learning model for use in monitoring a subject using a wearable device comprising an optophysiological sensor system, the method comprising: acquiring a pulsatile signal indicative of at least one physiological characteristic of the subject; obtaining an indication of at least one physical variable as a function of time from the wearable device; transmitting data representative of the pulsatile signal and an indication of the physical variable via a wireless interface to a remote computing platform; training the machine learning model at the remote computing platform based on the received data; transmitting coefficients / parameters from the remote computing platform to the machine learning model embedded in the wearable device after training; Including, method.

44. 44. The method of claim 43, and training the machine learning model based on the received data at the remote computing platform includes training a generative adversarial network (GAN) including two generators and one discriminator. method.

45. 45. The method of claim 44, one of the two generators is configured to learn a noise-tolerant mapping from an input noisy pulsatile signal to a reference signal, and the other of the two generators is configured to learn a noise distribution in a physical variable captured from the wearable device; method.

46. 1. A method of training a machine learning model for use in monitoring a subject using a wearable device comprising an optophysiological sensor system, the method comprising: acquiring a pulsatile signal indicative of at least one physiological characteristic of the subject; obtaining from the wearable device an indication of at least one physical variable as a function of time; processing the pulsatile signal and the indicator of the at least one physical variable with the machine learning model to provide an output indicative of a physiological characteristic; transmitting said pulsatile signal and data representative of said indicator of said physical variable to a remote computing platform via a wireless interface; training, at the remote computing platform, an improved machine learning model based on the received data; transmitting the improved and trained machine learning model to the wearable device; using the improved and trained machine learning model, processing the pulsatile signal and the indicator of the at least one physical variable with the machine learning model to provide an output indicative of a physiological characteristic; Including, method.

47. 47. The method of any one of claims 43 to 46, acquiring the pulsatile signal providing an indication of at least one physiological characteristic of the subject and acquiring an indication of at least one physical variable as a function of time from the wearable device are performed while a user of the wearable device is performing an exercise routine selected by the user; the machine learning model is trained on the remote computing platform based on the exercise routine. method.

48. 1. A wearable optophysiological sensor system for acquiring physiological characteristics of a wearer and physical variables of the wearer's movements, comprising: an optoelectronic sensor panel comprising a plurality of photodiodes and a plurality of illumination sources, at least selected ones of the plurality of illumination sources configured to emit light at a wavelength different from the other illumination sources; a printed circuit board including a microelectromechanical sensor for acquiring the physical variable, a processor configured to process signals received from the microelectromechanical sensor and the plurality of photodiodes, and at least one wireless communication interface; Equipped with the plurality of illumination light sources are arranged at intersections of a plurality of concentric circles, each of the plurality of concentric circles being centered on a respective one of the plurality of photodiodes; Wearable optophysiological sensor system.

49. 46. ​​The wearable optophysiological sensor system of claim 45, the plurality of illumination sources comprises at least four different illumination sources, e.g., light emitting diodes; Each of the illumination sources is configured to emit light of a different wavelength. Wearable optophysiological sensor system.

50. 50. The wearable optophysiological sensor system of claim 48 or 49, the plurality of illumination sources comprises at least 16 different illumination sources, each configured to emit light at one of four or more different wavelengths; the at least 16 different illumination sources are grouped into groups of four, each configured to emit light at one of the four or more different wavelengths; Wearable optophysiological sensor system.

51. 51. The wearable optophysiological sensor system of claim 49 or 50, the illumination source with the shortest wavelength is located at an intersection of an inner circle of the plurality of concentric circles, and the illumination source with longer wavelength is located at an intersection of the concentric circles farther from the center, the illumination source with longer wavelength is located farther from each of the photodiodes, and the wavelength of the illumination source increases with radial distance from each of the photodiodes; Wearable optophysiological sensor system.

52. 52. The wearable optophysiological sensor system of any one of claims 48 to 51, the photoelectric sensor panel comprises one, two, or three of the photodiodes; Wearable optophysiological sensor system.

53. 53. The wearable optophysiological sensor system of any one of claims 48 to 52, comprising: At least one wavelength is about 465 nm or about 1450 nm; Wearable optophysiological sensor system.

54. 54. The wearable optophysiological sensor system of any one of claims 48 to 53, and further comprising a planar pressure sensor for obtaining an indication of contact pressure between the optical physiological sensor and the wearer. Wearable optophysiological sensor system.

55.

56.

57. A non-transitory computer readable storage medium comprising a computer program configured to cause a processor to perform the method of any one of claims 1 to 17, 29 to 47, 55 and 56.

Images