Biological state diagnostic system

Abstract

The biological state diagnostic system of the present invention is a biological state diagnostic system having at least an acquisition unit, a receiving unit, a processing unit, a processing unit, and a diagnostic unit for biological data. In the acquisition unit, biological data is acquired by a reflective photoplethysmography (PPG) sensor. In the processing unit, the acquired biological data is processed in the most suitable manner. In the processing unit, feature points are extracted from the processed data. And in the diagnostic unit, the biological state is diagnosed using machine learning based on the extracted biological data.

Description

Technical Field

[0001] This invention relates to a biological state diagnostic system, and more specifically to a biological state diagnostic system that diagnoses based on biological data obtained using at least a reflective photoplethysmography (PPG) sensor. Background Technology

[0002] The anticipated increase in healthcare costs due to an aging population is expected to become a significant problem in the future, leading to a surge in digital transformation (DX) of healthcare. Against this backdrop, sensor technology has developed, enabling the easy acquisition of vital data in a non-invasive manner. This includes a wealth of vital data, such as electrocardiograms, electromyograms, electropulmonary electrograms, brain waves, blood oxygen concentration measured by light absorption in the blood, and pulse (see, for example, Patent Documents 1-4).

[0003] Moreover, these machines have improved accuracy and convenience, enabling even non-professionals to obtain medical-grade data.

[0004] Currently in hospitals, doctors study data obtained by nurses and other professionals to make diagnoses, but preparations are underway to create an environment that can achieve automated diagnosis through the use of sensing and AI technologies. If high-precision vital signs data can be obtained more easily, a large amount of data can be collected in time and space (all parts of the body), and by taking into account previously overlooked minute signals, it is expected that the accuracy of disease prediction and diagnosis can be improved.

[0005] The pulse contains a wealth of information about the circulatory system, and by interpreting the waveform and changes over time, diseases can be diagnosed.

[0006] However, since pulses are typically taken by wearing the fingertip, it is cumbersome and difficult to wear continuously in daily life. Furthermore, it usually uses a transmissive method, which can only measure at the end of the light path, but blood vessels are distributed throughout the body, and their location varies depending on the measurement site.

[0007] For example, in the case of a blood clot in the foot, changes in the pulse measured in the hand are difficult to detect. Therefore, there are changes that cannot be visualized by measuring only the fingertip.

[0008] Moreover, although it is possible to capture the signs of cerebral infarction and myocardial infarction caused by atrial fibrillation by simultaneously acquiring the waveform and pulse obtained from the electrocardiogram, the use of a fingertip type requires wearing separate devices for the electrocardiogram and pulse, which increases inconvenience (see, for example, Patent Document 1).

[0009] In addition to ECG, pulmonary electrocardiogram, electromyography, and brainwave also use adhesive-type sensors. Therefore, when pulse is obtained by transmissive means, there is a problem that makes it difficult to integrate the devices.

[0010] On the other hand, photoplethysmography (PPG) is a vital signs monitoring technique widely used in pulse oximeters. This technique uses highly transmissive red and infrared light from the body and calculates blood oxygen concentration by utilizing the difference in absorbance caused by the wavelengths of oxidized and non-oxidized hemoglobin.

[0011] Transmissive methods are generally used, but in living organisms, the transmissible sites are limited to extremities such as fingertips and ears, making them less desirable from a wearability perspective. In reflective photoplethysmography (PPG), which offers excellent wearability, green light has been commercialized using smartwatches, but there are fewer examples of commercialization using red or infrared light technologies.

[0012] One reason that can be cited is the low sensitivity of pulse detection. Compared to transmissive methods, in the case of reflective methods, the light is repeatedly scattered and returns to the body surface, resulting in a very small amount of light, which easily generates noise and makes it difficult to obtain a clear signal.

[0013] Furthermore, blood vessels are located deeper than the distal parts of the body, requiring the light to penetrate deeper. However, this necessitates increasing the distance between the sensor and the light source, resulting in further light attenuation. While red and infrared light are needed to measure blood oxygen concentration, green light alone can suffice for measuring pulse rate.

[0014] However, living organisms absorb a large amount of green light, which can only reach their surface. Therefore, it can only capture changes in capillaries, making it unsuitable for precisely depicting the shape of a pulse.

[0015] On the other hand, as an advantage of using green light, the pulsating SN has higher sensitivity and is less affected by external light due to its greater absorbance. In addition, capillaries are more evenly distributed, so they are also more robust to location.

[0016] When the shape of the pulse becomes important, red and infrared light are sometimes more suitable than green light. In recent years, there have been reports of preventing cerebral infarction, myocardial infarction, etc. by capturing changes in atrial fibrillation, and studies have been conducted on continuous monitoring of electrocardiograms and pulses (see, for example, Patent Document 1).

[0017] If prevention can be achieved through constant monitoring, it also relates to reducing medical costs. By designing it as a reflective device, the measurement area can be expanded, and products with less discomfort during wear can be made, increasing the freedom of choice in constant monitoring and measurement.

[0018] Therefore, there is a need for a biological state diagnostic system that uses sensors such as reflective photoplethysmography (PPG) sensors that improve the degrees of freedom of the measurement site of the subject measured by the sensor, the wearing comfort of the sensor, and the measurement accuracy, and can be applied to continuous monitoring.

[0019] Existing technical documents

[0020] Patent Document 1: Japanese Patent No. 5218139

[0021] Patent Document 2: Japanese Patent Application Publication No. 2018-130541

[0022] Patent Document 3: Japanese Patent Application Publication No. 2018-202130

[0023] Patent Document 4: Japanese Patent Application Publication No. 2019-97828 Summary of the Invention

[0024] The present invention was made in view of the above-mentioned problems and conditions, and its solution is to provide a biological state diagnostic system using a reflective photoplethysmography (PPG) sensor that improves the degree of freedom of the measurement site of the subject measured by the sensor, the wearing comfort of the sensor, and the measurement accuracy, and can be applied to continuous monitoring.

[0025] In order to solve the above-mentioned problems, the inventors studied the causes of the above-mentioned problems and found that by using a reflective photoplethysmography sensor and using machine learning to diagnose the state of a living organism, the degree of freedom of the measurement site of the subject measured by the sensor, the wearing comfort of the sensor, and the measurement accuracy are improved, so that it can be applied to constant monitoring. This invention was made based on that.

[0026] That is, the above-mentioned problems involved in this invention are solved by the following means.

[0027] 1. A biological state diagnostic system, comprising at least a biological data acquisition unit, a receiving unit, a processing unit, a data handling unit, and a diagnostic unit, wherein,

[0028] In the acquisition unit, biological data is acquired using a reflective photoplethysmography (PPG) sensor.

[0029] In the processing unit, the acquired biological data is processed in the most suitable manner.

[0030] In the processing unit, feature points are extracted from the processed data, and...

[0031] In the diagnostic unit, machine learning is used to diagnose the organism's condition based on the extracted organism data.

[0032] 2. In the biological state diagnostic system described in item 1, the light-emitting element of the reflective photoplethysmography sensor is planar and is used as a surface light source for irradiating biological tissues.

[0033] 3. In the biological state diagnostic system described in item 1 or 2, the light-emitting element of the reflective photoplethysmography pulse sensor is an organic EL element.

[0034] 4. In any of the biological state diagnostic systems described in items 1 to 3, the light-receiving element of the reflective photoplethysmography sensor is an organic photodiode.

[0035] 5. In any of the biological state diagnostic systems described in items 1 to 4, the biological data is always measured and acquired in the acquisition unit.

[0036] 6. In any of the biological state diagnostic systems described in items 1 to 5, two or more biological data are simultaneously measured and acquired in the acquisition unit.

[0037] 7. In any of the biological state diagnostic systems described in items 1 to 6, biological data from two or more biological sites are simultaneously measured and acquired in the acquisition unit.

[0038] 8. In any of the biological state diagnostic systems described in items 1 to 7, a synthesis unit is further provided to synthesize the acquired biological data and summarize it into a single data set.

[0039] 9. In the organism state diagnostic system described in any one of items 1 to 8, the diagnostic unit uses a machine learning model that uses organism data corresponding to the attribute as explanatory variables and the corresponding organism state corresponding to the attribute as the target variable.

[0040] 10. In the biological state diagnostic system described in any one of items 1 to 9, the diagnostic unit uses a deep learning model based on the obtained biological data.

[0041] 11. In any of the biological state diagnostic systems described in items 1 to 10,

[0042] The aforementioned reflective photoplethysmography (PPG) sensor is a reflective PPG sensor that includes a planar light-emitting element and a light-receiving element.

[0043] The light-emitting element and the light-receiving element are not positioned opposite each other in the vertical direction.

[0044] When the shortest distance between the light-emitting element and the light-receiving element is set as h (mm), the thickness of human skin (epidermis + dermis) is set as t (mm), and this thickness is set to be within the range of 0.1 to 4 mm,

[0045] The shortest distance h (mm) between at least one set of the light-emitting elements and the light-receiving elements satisfies the following equation (1).

[0046] Equation (1): (t×0.7) 2 ≤h≤(t×1.3) 2 .

[0047] 12. In the biological state diagnostic system described in item 11, the shortest distance between the light-emitting element and the light-receiving element is in the range of 2 to 25 mm.

[0048] 13. The biological state diagnostic system described in item 11 or 12 includes a plurality of the light-emitting elements, each light-emitting element being arranged in a substantially concentric circle with the center point of the light-receiving element as the center.

[0049] Through the above-described means of the present invention, a biological state diagnostic system can be provided, which uses a reflective photoplethysmography (PPG) sensor that improves the degree of freedom of the measurement site of the subject measured by the sensor, the wearing comfort of the sensor, and the measurement accuracy, and can be applied to continuous monitoring.

[0050] In the biological state diagnostic system of the present invention, at least a reflective photoplethysmography (PPG) sensor is used, so it is possible to measure even light-impermeable areas of the subject, thus enabling the acquisition of the pulse distribution throughout the body. That is, it expands the measurement sites and allows for continuous monitoring, increasing the degree of freedom in selecting measurement sites.

[0051] Furthermore, it is easy to integrate with other life sensors, eliminating the hassle of continuous measurement. In addition, "continuous measurement" here refers to continuous measurement over a certain period in daily life. For example, point measurements are fine for chronic diseases, but for diseases with sudden onset, regular, continuous measurement over a certain period—that is, continuous measurement—can detect early signs and issue warnings (alarms).

[0052] Furthermore, in the diagnostic unit of the organism state diagnostic system of the present invention, machine learning is used to diagnose the organism state based on the extracted organism data. Therefore, it can be considered that in this diagnostic unit, a learning model that uses the organism data corresponding to the attribute as the explanatory variable and the corresponding organism state corresponding to the attribute as the target variable is used to perform machine learning, and accurate diagnosis can be made based on a large amount of information. Attached Figure Description

[0053] Figure 1 This is a schematic diagram illustrating an example of the overall structure of a biological state diagnostic system.

[0054] Figure 2 This is a schematic diagram illustrating the relationship between the shortest distance between the light-emitting element and the light-receiving element in a reflective photoplethysmography (PPG) sensor and the thickness of the skin.

[0055] Figure 3A This is a schematic diagram illustrating the arrangement of the light-emitting element and the light-receiving element, etc., constituting the reflective photoplethysmography pulse sensor according to the first embodiment.

[0056] Figure 3B This is a schematic diagram illustrating the arrangement of the light-emitting element and the light-receiving element, etc., constituting the reflective photoplethysmography pulse sensor according to the first embodiment.

[0057] Figure 3C This is a schematic diagram illustrating the arrangement of the light-emitting element and the light-receiving element, etc., constituting the reflective photoplethysmography pulse sensor according to the first embodiment.

[0058] Figure 3D This is a schematic diagram illustrating the arrangement of the light-emitting element and the light-receiving element, etc., constituting the reflective photoplethysmography pulse sensor according to the first embodiment.

[0059] Figure 4A This is a schematic diagram illustrating the arrangement of the light-emitting element and the light-receiving element, etc., constituting the reflective photoplethysmography pulse sensor according to the second embodiment.

[0060] Figure 4B This is a schematic diagram illustrating the arrangement of the light-emitting element and the light-receiving element, etc., constituting the reflective photoplethysmography pulse sensor according to the second embodiment.

[0061] Figure 4C This is a schematic diagram illustrating the arrangement of the light-emitting element and the light-receiving element, etc., constituting the reflective photoplethysmography pulse sensor according to the second embodiment.

[0062] Figure 4DThis is a schematic diagram illustrating the arrangement of the light-emitting element and the light-receiving element, etc., constituting the reflective photoplethysmography pulse sensor according to the second embodiment. Detailed Implementation

[0063] The biological state diagnostic system of the present invention is a biological state diagnostic system having at least an acquisition unit, a receiving unit, a processing unit, a processing unit, and a diagnostic unit for biological data. In the acquisition unit, biological data is acquired by a reflective photoplethysmography (PPG) sensor. In the processing unit, the acquired biological data is processed in the most suitable manner. In the processing unit, feature points are extracted from the processed data. And in the diagnostic unit, the biological state is diagnosed using machine learning based on the extracted biological data.

[0064] This feature is a technical feature shared in the following embodiments.

[0065] As an embodiment of the present invention, from the viewpoint of presenting the effects of the present invention, it is preferable that the light-emitting element of the reflective photoplethysmography (PPG) sensor is planar, serving as a surface light source for irradiating biological tissue. Furthermore, in terms of easily obtaining a flexible sensor with good wearability, it is preferable that the light-emitting element of the reflective PPG sensor is an organic EL element. Moreover, it is preferable that the light-receiving element of the reflective PPG sensor is an organic photodiode.

[0066] As an implementation method, from the viewpoint of presenting the effects of the present invention, it is preferable to measure and acquire the biological data continuously in the acquisition section. Furthermore, it is preferable to simultaneously measure and acquire two or more types of biological data in the acquisition section. Moreover, it is preferable to simultaneously measure and acquire biological data from two or more biological sites in the acquisition section.

[0067] In terms of enabling comprehensive diagnosis, it is preferable to also include a synthesis unit that synthesizes the obtained biological data and summarizes it into a single data set.

[0068] Accurate diagnosis can be made based on a large amount of information, so it is preferable to use a machine learning model in the diagnostic unit that uses biological data corresponding to the attributes as explanatory variables and the corresponding biological state as the target variable. Alternatively, it is preferable to use a deep learning model in the diagnostic unit that uses the obtained biological data.

[0069] As an embodiment of the present invention, in terms of improving measurement accuracy, it is preferable that the reflective photoplethysmography sensor is a reflective photoplethysmography sensor having a planar light-emitting element and a light-receiving element, wherein the light-emitting element and the light-receiving element are not opposite each other in the vertical direction, and when the shortest distance between the light-emitting element and the light-receiving element is set as h (mm), the thickness of human skin (epidermis + dermis) is set as t (mm), and the thickness is set to be in the range of 0.1 to 4 mm, the shortest distance h (mm) between at least one set of the light-emitting element and the light-receiving element satisfies the formula (1).

[0070] Furthermore, from the viewpoint of measurement accuracy, it is preferable that the shortest distance between the light-emitting element and the light-receiving element is within the range of 2 to 25 mm.

[0071] Furthermore, in terms of improving robustness to positional shifts, it is preferable to have a plurality of the light-emitting elements, each of which is arranged in a substantially concentric circle with the center point of the light-receiving element as the center.

[0072] 1. Overall structure of the biological state diagnostic system of the present invention

[0073] A biological state diagnostic system is a biological state diagnostic system that has at least an acquisition unit, a receiving unit, a processing unit, a processing unit, and a diagnostic unit for biological data. In the acquisition unit, biological data is acquired by a reflective photoplethysmography (PPG) sensor. In the processing unit, the acquired biological data is processed in the most suitable manner. In the processing unit, feature points are extracted from the processed data. In the diagnostic unit, machine learning is used to diagnose the biological state based on the extracted biological data.

[0074] 1.1 Example of the overall structure of a biological state diagnostic system

[0075] Biological state diagnostic systems possess the characteristics described above, but their overall structure can take various forms. The following is an example, but not a limitation.

[0076] Figure 1 This is a schematic diagram showing an example of the overall structure of a biological state diagnostic system. The biological state diagnostic system 1 can visualize the internal state of the subject as an image such as a photoplethysmogram using received signals from a biological data acquisition unit 10 equipped with at least a reflective photoplethysmogram sensor (hereinafter also referred to as a "PPG sensor").

[0077] In this invention, "images" include waveforms, graphs, and tables of pulse.

[0078] like Figure 1As shown, the biological state diagnostic system 1 includes a biological data acquisition unit 10, a drive unit 11, a receiving unit 12, a processing unit 13, a processing unit 14S (which includes a first signal processing unit 14a, a second signal processing unit 14b, and a frame data storage unit 14c), a synthesis unit 15, a display processing unit 16, a display unit 17, a measurement location setting unit 18, a receiving data storage unit 19, a diagnostic unit 20, an operation input unit 30, and a control unit 40.

[0079] The drive unit 11, the receiving unit 12, the measurement position setting unit 18, the processing unit 14S, the synthesis unit 15, and the display processing unit 16 are composed of at least one dedicated hardware (electronic circuit) corresponding to each process, such as a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), or a PLD (Programmable Logic Device).

[0080] The control unit 40 includes a CPU (Central Processing Unit) 40a as an arithmetic / control device, a RAM (Random Access Memory) 40b as a main storage device, and a ROM (Read Only Memory) 40c. Basic programs and basic setting data are stored in the ROM 40c. The CPU 40a reads the program corresponding to the processing content from the ROM 40c, expands it in the RAM 40b, and executes the expanded program, thereby centrally controlling the operation of each functional module (drive unit 11, receiver unit 12, measurement location setting unit 18, frame data storage unit 14c, receiver data storage unit 19, display processing unit 16, display unit 17, first signal processing unit 14a, and second signal processing unit 14b) of the biological state diagnostic system 1.

[0081] In this embodiment, the functions of each functional module are realized by cooperating with the hardware components constituting the functional module and the control unit 40. Alternatively, some or all of the functions of each functional module can be realized by the control unit 40 executing a program.

[0082] The biological data acquisition unit 10 includes at least a reflective photoplethysmography (PPG) sensor 2. Furthermore, depending on the purpose, it may also include an electrocardiogram (ECG) sensor (e.g., a sensor with electrode patches worn with electrodes attached to the subject's skin), a blood pressure sensor, an ultrasound sensor, etc.

[0083] The organism data acquisition unit 10 can continuously measure and acquire organism data. Furthermore, it can simultaneously measure and acquire two or more types of organism data. Moreover, it can simultaneously measure and acquire organism data from two or more organism sites.

[0084] Furthermore, it is preferable to have an interface for transmitting biological data from the biological data acquisition unit 10 to the receiving unit 12 and transmitting drive signals from the driving unit 11 to the biological data acquisition unit 10 via wireless communication such as Bluetooth (Japanese registered trademark) and WiFi (Japanese registered trademark).

[0085] The drive unit 11 generates a drive signal according to the instructions of the control unit 40 and outputs it to a sensor such as a PPG sensor. Although not shown in the figure, the drive unit 11 includes, for example, a clock generation circuit, a pulse generation circuit, a pulse width setting unit, and a delay circuit.

[0086] The receiving unit 12 receives signals from PPG sensors, etc., according to the instructions of the control unit 40, and outputs them to the processing unit 13, the processing unit 14S (first signal processing unit 14a and second signal processing unit 14b), and the receiving data storage unit 19. Although not shown in the figure, the receiving unit 12 includes, for example, an amplifier, an A / D conversion circuit, and a phase modulation adder circuit.

[0087] The processing unit 13 processes the acquired biological data in the most suitable manner according to the purpose. For example, it processes the data for real-time display of pulse waveforms, display of acceleration pulse waveforms, display of power spectral density graphs, and display of attractor screens.

[0088] Furthermore, in the processing unit 13 or the processing unit 14S described below, in order to remove noise from the biological data output from the receiving unit 12 and improve accuracy, it is preferable to use a bandpass filter, a low-pass filter, etc., and then process / process using a moving average method or a frequency domain method.

[0089] In the processing unit 14S, based on the data processed in the above-mentioned process, feature points are extracted in the first signal processing unit 14a and the second signal processing unit 14b, and processed non-image pulse data and image data are generated.

[0090] In the first signal processing unit 14a, according to the instructions of the control unit 40, predetermined receiving data, such as pulse-related receiving data, from the receiving unit 12 or the receiving data storage unit 19 is processed in a predetermined manner to generate non-image pulse data and image data for the purpose of processing.

[0091] For example, depending on the purpose, envelope detection processing, logarithmic compression processing, and brightness transformation can be performed by adjusting the dynamic range and gain, thereby generating image data.

[0092] In addition, the first signal processing unit 14a includes a DSC (Digital Scan Converter) that performs coordinate transformation and pixel interpolation corresponding to the type of sensor such as a PPG sensor.

[0093] In the second signal processing unit 14b, according to the instructions of the control unit 40, predetermined received data, such as data related to the properties of blood, from the receiving unit 12 or the received data storage unit 19 is processed in the same way as the first signal processing unit 14a, or in other ways, to generate image data. This image data may also include brightness information and color information that are different from the image data described above.

[0094] The frame data storage unit 14c is constructed, for example, of volatile memory such as DRAM (Dynamic Random Access Memory) or non-volatile memory that can be rewritten at high speed. The frame data storage unit 14c stores frame data generated by the first signal processing unit 14a and the second signal processing unit 14b in frame units. The frame data obtained by the signal processing unit can be composed of dynamic data prior to image processing. The dynamic data includes speed data representing the speed of movement, power data representing flow rate, and variance data representing the disturbance condition. The frame data stored in the frame data storage unit 14c is read out under the control of the control unit 40 and a predetermined image processing is performed in the first signal processing unit 14a and the second signal processing unit 14b.

[0095] In the synthesis unit 15, image data obtained by the first signal processing unit 14a and the second signal processing unit 14b are synthesized according to the purpose, and data that can grasp the overall correlation is generated.

[0096] For example, when selecting the first signal, the synthesis unit 15 converts the first signal image data from the first signal processing unit 14a into a display signal and outputs it to the display processing unit 16. Furthermore, when selecting the second signal, the synthesis unit 15 synthesizes the first signal image data from the first signal processing unit 14a and the second signal image data from the second signal processing unit 14b, converts the synthesized second signal image data into a display signal, and outputs it to the display processing unit 16.

[0097] According to the instructions of the control unit 40, the display processing unit 16 transforms the image data generated in the synthesis unit 15 into a display signal corresponding to the display unit 17 and outputs it, so that the image of the display unit 17 is displayed.

[0098] In addition, the display processing unit 16 overlays the measurement part frame onto the first signal image or the second signal image based on the area of ​​interest set by the measurement part setting unit 18.

[0099] The display unit 17 is composed of, for example, a liquid crystal display, an organic EL display, a CRT display, etc. The display unit 17 displays non-image pulse data and images processed based on display signals from the display processing unit 16, according to the instructions of the control unit 40.

[0100] The measurement location setting unit 18 sets the measurement location (also called "region of interest") in the measurement object (e.g., human body) according to the instructions of the control unit 40.

[0101] The measurement location setting unit 18 sets the area set by the operation input unit 30 as the region of interest. The region of interest is set to include the diagnostic object (e.g., the blood flow section), and is displayed as a measurement location frame on the display image obtained via the first signal processing unit 14a. Within this measurement location frame, the image obtained via the second signal processing unit 14b is displayed overlaid.

[0102] Like the frame data storage unit 14c, the receive data storage unit 19 is constructed, for example, of a volatile memory such as DRAM or a non-volatile memory that can be rewritten at high speed. The receive data storage unit 19 stores received data based on the received signal from the receive unit 12. Furthermore, the frame data storage unit 14c and the receive data storage unit 19 may also be constructed of the same storage device.

[0103] The operation input unit 30 accepts input of various information, such as information related to diagnosis.

[0104] The operation input unit 30 may include, for example, an operation panel with multiple input switches, a keyboard, and a mouse. Alternatively, the operation input unit 30 may be a touch panel integrated with the display unit 17. Users can use the operation input unit 30 to set the area of ​​interest, the location to be diagnosed, the type of sensor such as the PPG sensor, and the image processing method in the processing unit.

[0105] The diagnostic unit 20 has the function of making diagnoses based on biological data displayed on the display unit.

[0106] In the diagnostic unit 20, a machine learning model can be used that uses biological data corresponding to the attributes as explanatory variables and the corresponding biological state corresponding to the attributes as the target variable.

[0107] Furthermore, in the diagnostic unit 20, a learning model using deep learning based on the obtained biological data can be used.

[0108] Here, "machine learning" refers to the mechanical learning of patterns and correlations in data using large amounts of data, and the identification, recognition, detection, and prediction of such data.

[0109] Examples of machine learning models that can be used in this invention include support vector machines (SVM), neural networks (NNET), random forests, and deep learning models. Among these models, deep learning models are preferred.

[0110] For example, in the analysis of a multiple regression model constructed using pulse data obtained from subjects such as patients with specific diseases as training data (learning data), "probability of disease %" and "blood pressure" are used as target variables, and data on characteristic quantities extracted from the shape of the pulse (such as the minimum or maximum amplitude, the time interval from the rising point to the next rising point, frequency component extraction by FFT (Fast Fourier Transform), Maharanobis-Taguchi System, dynamic time stretching method, etc.) are used as explanatory variables. Multiple regression analysis is performed to calculate the probability of disease % and blood pressure, etc., as target variables, based on the characteristic quantities of the pulse and other vital signs data obtained at the same time.

[0111] Furthermore, depending on whether the desired output is numerical or categorical, the analysis method used can be changed. It is preferable to choose a suitable regression or classification method such as multiple regression analysis, principal component analysis, or PLS regression that combines both.

[0112] 2. Reflective photoplethysmography pulse sensor

[0113] 2.1 Functions of a reflective photoplethysmography (PPG) sensor

[0114] In the reflective photoplethysmography sensor of the present invention, light is irradiated onto the measured part of a biological tissue by a light-emitting element such as a light-emitting diode as a light source, and the light reflected from the measured part of the biological tissue is detected by a light-receiving element such as a photodiode as a light detection sensor, and the changes in the volume pulse of the biological tissue are measured based on the detected signal.

[0115] For example, light emitted from a light-emitting element passes through the epidermis and reaches the blood vessels inside. The light that reaches the blood vessels is absorbed, reflected, or transmitted through the blood flowing in those vessels.

[0116] In this process, light reflected by vascular tissue and the blood flowing through the blood vessels is incident on the light-receiving element. Therefore, the light-receiving element outputs a photocurrent corresponding to the amount of incident light. Here, blood vessels repeatedly expand and contract in a cycle identical to the heartbeat. Therefore, the amount of light reflected increases or decreases in the same cycle as the expansion and contraction of the blood vessels, so the change in the photocurrent output from the light-receiving element represents the change in the volume of the blood vessels.

[0117] Furthermore, the reflective photoplethysmography (PPG) sensor according to this invention can also be used as a sensor for detecting the oxygen saturation of arterial blood. Hemoglobin in the blood absorbs red and infrared light differently depending on whether it binds to oxygen. Therefore, by preparing multiple sets of elements with different emission and reception wavelengths, such as elements that emit / receive red light and elements that emit / receive infrared light, oxygen saturation can be detected by measuring and analyzing their reflected light.

[0118] In addition, "volume pulse" refers to the waveform of the volume change caused by the pressure change within the blood vessel due to the pulsation, which can directly monitor changes in the blood vessel.

[0119] In addition, "photoplethysmography" refers to the waveform detected by using the transmission or reflection of light from blood to detect volumetric pulse.

[0120] <Methods for calculating oxygen saturation>

[0121] Generally, to measure blood oxygen saturation, a reflective photoplethysmometer (PPI) is used. The PPI measures the transmittance and reflectance of light emitted from a finger within two wavelengths in the red to near-infrared range.

[0122] Specifically, hemoglobin (Hb) in the blood exists in four states: oxidized hemoglobin (O2Hb), reduced hemoglobin (HHb), methemoglobin (MetHb), and carboxyhemoglobin (COHb). MetHb and COHb are abnormal hemoglobins that increase due to methemoglobinemia and carbon monoxide poisoning. Therefore, oxygen saturation is usually determined by the ratio of O2Hb to O2Hb + HHb.

[0123] When red light is transmitted through hemoglobin (Hb), the absorbance of red light by HHb is significantly greater than that by O2Hb, varying considerably depending on the wavelength of the red light. Conversely, when near-infrared light is transmitted through hemoglobin (Hb), the absorbance of near-infrared light by HHb is slightly lower than that by O2Hb. Therefore, the ratio R (absorbance of red light / absorbance of near-infrared light) of the absorbance of hemoglobin varies depending on the ratio of O2Hb to O2Hb + HHb in the blood, i.e., the oxygen saturation.

[0124] In addition, arterial blood, venous blood, tissue, and bone are present in measurement sites such as fingers, wrists, back of the wrist, chest, and abdomen, which affect the absorbance of red light and near-infrared light. Furthermore, the portion affected by the volume change of the pulse is arterial blood. When the absorbance of the pulse portion is set as AC, and the absorbance of the non-pulsating portion of arterial blood, venous blood, tissue, and bone is set as DC, for example, setting the AC component of the absorbance of red light at wavelength 660nm as AC660 and the DC component as DC660, and setting the AC component of the absorbance of near-infrared light at wavelength 940nm as AC940 and the DC component as DC940, the ratio R (absorbance of red light / absorbance of near-infrared light) of the absorbance of red light to that of near-infrared light is expressed by the following formula (I).

[0125] [Mathematical Expression 1]

[0126] Formula (I)

[0127]

[0128] Blood oxygen saturation can be determined from the R value calculated using equation (I) above, and from the calibration curve showing the relationship between the pre-confirmed R value and percutaneous arterial oxygen saturation (SpO2).

[0129] Furthermore, when measuring absorbance by irradiating the fingers, wrists (e.g., ulnar side, radial side), back of the wrist, chest, and abdomen with red and near-infrared light, the time-varying variation of absorbance is measured as a waveform reflecting the pulse. Therefore, the AC component of absorbance can be determined by calculating the difference between the maximum and minimum values ​​of the time-varying variation, and the DC component of absorbance can be determined by calculating the average value of the time-varying variation. In the embodiments described later, the calculated value will be expressed as "AC / DC".

[0130] 2.2 Overall Structure of Reflective Photoplethysmography (PPG) Sensor

[0131] As the overall structure of a reflective photoplethysmography pulse sensor, it can be adopted in various forms and methods.

[0132] For example, as a basic overall structure, it is preferably configured to include: a light-emitting element that is driven to flash and emit light at a frequency higher than the volumetric pulse of the organism according to a signal from an oscillator; a light-receiving element that receives light emitted from the light-emitting element and light reflected by the biological tissue of the subject being measured from external light, and generates a signal having a signal level corresponding to the amount of light received; an extraction unit that extracts the signal component generated by the external light from the signal generated by the light-receiving element; a reduction unit that reduces the noise signal caused by external light, etc., in the signal generated by the light-receiving element by subtracting the signal component extracted by the extraction unit from the signal generated by the light-receiving element; and a unit that generates a signal representing the volumetric pulse of the subject being measured from the signal after the noise signal component caused by external light, etc., has been reduced by the reduction unit.

[0133] 2.2.1 Example of the structure of the main components of a reflective photoplethysmography (PPG) sensor

[0134] The structure of the main part of the reflective photoplethysmography pulse sensor can take various forms known in the past, but in this invention, a sensor having a form that satisfies the following requirements is preferred.

[0135] <Distance between the light-emitting element and the light-receiving element>

[0136] The reflective photoplethysmography sensor is preferably a reflective photoplethysmography sensor having a planar light-emitting element and a light-receiving element. The light-emitting element and the light-receiving element are not opposite each other in the vertical direction. When the shortest distance between the light-emitting element and the light-receiving element is set as h (mm), and the thickness of human skin (epidermis + dermis) is set as t (mm), and the thickness is set to be in the range of 0.1 to 4 mm, the shortest distance h (mm) between at least one set of the light-emitting element and the light-receiving element satisfies the following formula (1).

[0137] Equation (1): (t×0.7) 2 ≤h≤(t×1.3) 2

[0138] Furthermore, the above formula (1) means that the shortest distance between the light-emitting element and the light-receiving element is a length within the range of ±30% of the square of the thickness of the skin. Additionally, in the above formula (1), h and t represent only numerical values.

[0139] "The light-emitting element and the light-receiving element (in the vertical direction) are not opposite each other" means that the light-emitting element and the light-receiving element are not arranged in a way that overlaps each other in the vertical direction, but are arranged in parallel in the planar direction.

[0140] Figure 2This is a schematic diagram used to illustrate the relationship between the shortest distance between the light-emitting element and the light-receiving element and the thickness of the skin.

[0141] In this invention, "the thickness of human skin" refers to the total thickness of the epidermis and dermis, and the thickness of the skin is set to be in the range of 0.1 to 4 mm.

[0142] Furthermore, in this invention, "the shortest distance between the light-emitting element and the light-receiving element" refers to, for example... Figure 2 As shown, the shortest distance h (mm) from the side edge of the light-emitting element 80 side of the light-receiving element 70 to the side edge of the light-receiving element 70 side of the light-emitting element 80 when viewed in a side cross-section.

[0143] exist Figure 2 In the example, a case is given in which multiple light-emitting elements 80, 90, and 90B are provided for a light-receiving element 70. The shortest distance h (mm) between the light-receiving element 70 and the light-emitting element 80 is within the range that satisfies the above formula (1).

[0144] Furthermore, the shortest distance between the light-receiving element 70 and the light-emitting element 90, and the shortest distance between the light-receiving element 70 and the light-emitting element 90B, are assumed to be within the range that does not satisfy the above formula (1).

[0145] Therefore, as described below, biological information (life data) obtained based on light emitted by light-emitting elements 90, 90B that never meet the range of the shortest distance specified in this invention contains noise, and valid life data can be calculated by subtracting the life data containing the noise from the life data obtained based on light emitted from light-emitting elements 80 that meet the range of the shortest distance.

[0146] Furthermore, regarding measurement accuracy, it is preferable that the shortest distance between the light-emitting element and the light-receiving element is within the range of 2 to 25 mm. To improve robustness against positional shifts, it is preferable to have multiple light-emitting elements arranged concentrically around the center point of the light-receiving element. In particular, considering the size of the device, the upper limit of the shortest distance is preferably 17.5 mm or less, more preferably 10 mm or less.

[0147] <Example of the configuration of light-emitting elements and light-receiving elements, etc.>

[0148] The configuration of the light-emitting element and the light-receiving element involved in this invention is not limited as long as the conditions are met. Preferred configuration examples are described below.

[0149] The first embodiment described below is a case in which multiple light-emitting elements are arranged in approximately concentric circles with the center point of the light-receiving element as the center, and the second embodiment is a case in which multiple light-receiving elements are arranged in approximately concentric circles with the center point of the light-emitting element as the center.

[0150] Here, "approximately concentric circles" means that the difference in distance (radius) from the center point of the light-receiving element to the center point of each light-emitting element is within 10%, preferably within 5%, more preferably within 3%, and also includes the same.

[0151] The "center point of the light-receiving element" refers to the geometric center of the shape of the light-receiving element when viewed from above. Similarly, the "center point of the light-emitting element" refers to the geometric center of the shape of the light-receiving element when viewed from above.

[0152] Furthermore, "top view" refers to viewing the reflective photoplethysmography sensor from the normal direction relative to the upper surface of the substrate 41.

[0153] (First Embodiment)

[0154] Figure 3 is a schematic diagram illustrating the configuration of the light-emitting element and light-receiving element, etc., constituting a reflective photoplethysmography pulse sensor. Figure 3A It is a top view. Figure 3B It is a top view that breaks down the components. Figure 3C The cross-sectional view is observed using arrow AA′, and... Figure 2 D is the cross-sectional view observed by arrow BB′.

[0155] As shown in FIG3, in the detection unit 14, a plurality of lead-out electrode portions 63a, 63b, 63c and 63d of a first electrode (anode) 50 and a second electrode (cathode) 60 with light transmittance are formed on the light-transmitting substrate 41.

[0156] As the light-transmitting substrate 41, it is preferable to use a substrate with a total light transmittance of 70% or more, more preferably 80% or more, and particularly preferably 90% or more. The total light transmittance can be measured according to JIS K7375:2008 "Plastics - Method for determining total light transmittance and total light reflectance". In addition, examples of opaque substrates (light-reflective substrates) include metal plates such as aluminum and stainless steel, films, opaque resin substrates, and ceramic substrates.

[0157] exist Figure 3B In the first electrode 50, the first electrode 50 includes: a first central electrode portion 51 formed in the center of the light-transmitting substrate 41; and first circumferential electrode portions 52 and 52, which are arranged concentrically on the left and right sides with the first central electrode portion 51 as the center.

[0158] The first circumferential electrode portions 52 and 52 appear arc-shaped when viewed from above, so that the first circumferential electrode portions 52 and 52 are formed discontinuously from each other.

[0159] Furthermore, the first central electrode portion 51 is formed extending toward the lead-out electrode portion 63b. Also, two first peripheral electrode portions 52 and 52 are formed extending toward the lead-out electrode portion 63c.

[0160] The lead-out electrode portions 63a, 63b, 63c and 63d are formed on the light-transmitting substrate 41 and serve as wiring leads to the side edge of the light-transmitting substrate 41.

[0161] A light-receiving element 70 is formed on the first central electrode portion 51 of the first electrode 50, such that it covers the first central electrode portion 51. The light-receiving element 70 has a circular shape when viewed from above.

[0162] As the light-receiving element 70, a planar organic thin-film solar cell (OPV) or an organic photodiode (OPD) is preferably used. Organic photodiodes are particularly preferred due to their flexibility, comfortable fit, and ability to reduce wavelength and brightness deviations. Details regarding these organic thin-film solar cells and organic photodiodes will be described later.

[0163] Furthermore, light-emitting elements 80 and 80 are formed on the left and right first circumferential electrode portions 52 and 52 of the first electrode 50, respectively, in a manner that covers the first circumferential electrode portions 52 and 52. Each light-emitting element 80 and 80 appears arc-shaped when viewed from above along the first circumferential electrode portions 52 and 52, so that the light-emitting elements 80 and 80 are formed discontinuously from each other.

[0164] Thus, with the center point of the light-receiving element 70 as the center, two light-emitting elements 80 and 80 formed on the first circumferential electrode portions 52 and 52 are arranged in concentric circles.

[0165] Here, the shortest distance h (mm) between at least one set of light-emitting elements 80 and light-receiving elements 70 satisfies the above equation (1).

[0166] Furthermore, the "shortest distance h" between the light-emitting element 80 and the light-receiving element 70 refers to, as described above, in Figure 1 The shortest distance h from the side edge of the light-emitting element 80 side of the light-receiving element 70 to the side edge of the light-receiving element 70 side of the light-emitting element 80 when viewed in the side cross section shown.

[0167] In Figure 3, the light-receiving element 70 is circular when viewed from above, and the light-emitting element 80 is arc-shaped when viewed from above. Therefore, the shortest distance h between the outer peripheral surface of the light-receiving element 70 and the inner peripheral surface of the light-emitting element 80 is 70. Furthermore, when the light-emitting element 80 is circular when viewed from above, the shortest distance h between the outer peripheral surface of the light-receiving element 70 and the outer peripheral surface of the light-emitting element 80 is 70.

[0168] As the light-emitting element 80, a light-emitting diode (LED) or an organic EL element (OLED) is preferably used, and an organic EL element is particularly preferred in terms of reducing wavelength deviation and brightness deviation.

[0169] In addition, the two light-emitting elements 80 and 80 are preferably elements that emit light of the same wavelength.

[0170] Furthermore, a second electrode 60 is formed on the light-receiving element 70 and the light-emitting elements 80 and 80.

[0171] The second electrode (cathode) 60 includes: a second central electrode portion 61, formed at a position corresponding to the first central electrode portion 51 of the first electrode 50; and two second circumferential electrode portions 62 and 62, formed around the second central electrode portion 61 at positions corresponding to the first circumferential electrode portions 52 and 52, which are arc-shaped and non-discontinuous when viewed from above.

[0172] A lead-out electrode portion 63a is connected to the second central electrode portion 61. The lead-out electrode portion 63a is formed on the light-transmitting substrate 41, extending towards the side edge of the light-transmitting substrate 41, between the second circumferential electrode portions 62 and 62, around one of the second circumferential electrode portions 62.

[0173] Additionally, lead-out electrode portions 63d and 63d are also connected to each end of the two second-peripheral electrode portions 62 and 62. These lead-out electrode portions 63d and 63d are formed extending toward the side edge of the light-transmitting substrate 41.

[0174] Thus, regarding the first electrode 50, the lead-out electrode portions 63a to 63d, the light-receiving element 70, the light-emitting element 80, and the second electrode 60 formed on the light-transmitting substrate 41, a light-transmitting substrate 42 is also provided on the second electrode 60, and the two light-transmitting substrates 41 and 42 are sealed between them by a sealing material 43.

[0175] Furthermore, in Figure 3, the symbol H refers to the "shortest width" of the light-emitting element 80 when viewed from above. Here, the light-emitting element 80 is arc-shaped when viewed from above, so it refers to the length of the line through which tangents are drawn to the outer and inner circumferential surfaces of the arc-shaped element and is orthogonal to the tangents. The shortest width H of the light-emitting element 80 in Figure 3 is preferably around 5 mm. Furthermore, if the light-emitting element is circular when viewed from above, the diameter of this circle is set as the shortest width H.

[0176] (Second Implementation)

[0177] The second embodiment is as described above, in which multiple light-receiving elements are arranged in concentric circles with the light-emitting element as the center, and the other structures are the same as the first embodiment.

[0178] Figure 4 is a schematic diagram illustrating the configuration of the light-emitting element and light-receiving element, etc., constituting a reflective photoplethysmography pulse sensor. Figure 4A It is a top view. Figure 4B It is a top view that breaks down the components. Figure 4C The cross-sectional view is observed using the CC' arrow. Figure 4D The DD' arrow indicates the cross-sectional view.

[0179] As shown in FIG4, in the detection unit 14, a plurality of lead-out electrode portions 63a, 63b, 63c and 63d of a first electrode (anode) 50 and a second electrode (cathode) 60 with light transmittance are formed on the light-transmitting substrate 41.

[0180] The first electrode 50 includes: a first central electrode portion 51 formed at the center of the light-transmitting substrate 41; and first circumferential electrode portions 52 and 52 arranged concentrically around the first central electrode portion 51.

[0181] The first circumferential electrode portions 52 and 52 appear arc-shaped when viewed from above, so that the first circumferential electrode portions 52 and 52 are formed discontinuously from each other.

[0182] Furthermore, the first central electrode portion 51 is formed extending toward the lead-out electrode portion 63b. Also, two first peripheral electrode portions 52 and 52 are formed extending toward the lead-out electrode portion 63c.

[0183] The lead-out electrode portions 63a, 63b, 63c and 63d are formed on the light-transmitting substrate 41 and serve as wiring leads to the side edge of the light-transmitting substrate 41.

[0184] A light-emitting element 80 is formed on the first central electrode portion 51 of the first electrode 50, such that it covers the first central electrode portion 51. The light-emitting element 80 has a circular shape when viewed from above.

[0185] Furthermore, light-receiving elements 70 and 70 are formed on the two first circumferential electrode portions 52 and 52 in the first electrode 50, respectively, so as to cover the first circumferential electrode portions 52 and 52. Each light-receiving element 70 and 70 presents an arc shape along the first circumferential electrode portions 52 and 52 when viewed from above, so that the light-receiving elements 70 and 70 are formed discontinuously from each other.

[0186] Thus, the two light-receiving elements 70, 70 formed on the first circumferential electrode portion 52 are arranged in concentric circles with the light-emitting element 80 as the center.

[0187] Furthermore, a second electrode 60 is formed on the light-receiving element 70 and the light-emitting elements 80 and 80.

[0188] The second electrode (cathode) 60 includes: a second central electrode portion 61, formed at a position corresponding to the first central electrode portion 51 of the first electrode 50; and two second circumferential electrode portions 62, 62, formed around the second central electrode portion 61 at positions corresponding to the first circumferential electrode portions 52, 52, which are arc-shaped and non-discontinuous when viewed from above.

[0189] A lead-out electrode portion 63a is connected to the second central electrode portion 61. The lead-out electrode portion 63a is formed on the light-transmitting substrate 41, extending towards the side edge of the light-transmitting substrate 41, between the second peripheral electrode portions 62, 62, and surrounding one of the second peripheral electrode portions 62.

[0190] Additionally, lead-out electrode portions 63d and 63d are also connected to each end of the two second peripheral electrode portions 62 and 62. These lead-out electrode portions 63d and 63d are formed extending toward the side edge of the light-transmitting substrate 41.

[0191] Thus, regarding the first electrode 50, the lead-out electrode portions 63a to 63d, the light-receiving element 70, the light-emitting element 80, and the second electrode 60 formed on the light-transmitting substrate 41, a light-transmitting substrate 42 is also provided on the second electrode 60, and the two light-transmitting substrates 41 and 42 are sealed between them by a sealing material 43.

[0192] Furthermore, in embodiments of the present invention, in order to improve measurement accuracy, it is preferable to have a plurality of light-emitting elements, wherein at least two of the plurality of light-emitting elements emit light of the same wavelength, thereby enabling the plurality of light-emitting elements to emit light individually.

[0193] In terms of improving measurement accuracy, it is preferable to process the obtained information and data based on light emitted from light emitted by light-emitting elements located close to the light-receiving element and light-emitting elements located farther away from the light-emitting element, which enables the calculation of effective data after noise removal.

[0194] Having multiple light-receiving elements, each light-receiving element is arranged in a roughly concentric circle with the center point of the light-emitting element as the center, which can improve robustness to positional displacement.

[0195] In terms of improving measurement accuracy, it is preferable to process the obtained information and data based on light received from light-receiving elements located close to the light-emitting element and light-receiving elements located farther from the light-emitting element, respectively, which enables the calculation of effective data after noise removal.

[0196] In terms of improving measurement accuracy, it is preferable that there are multiple shortest distances between the light-emitting element and the light-receiving element, and that the light-emitting element and the light-receiving element having the shortest distance suitable for sensing are selected.

[0197] 2.3 [Components of a Reflective Photoplethysmography Sensor]

[0198] This section provides a detailed description of the light-emitting element and the light-receiving element, which are the main components of a reflective photoplethysmography pulse sensor.

[0199] <Light-emitting element>

[0200] In this invention, the light-emitting element is used as a light source for irradiating biological tissues.

[0201] As a light-emitting element, both organic light-emitting diodes (OLEDs) and inorganic light-emitting diodes (LEDs) can be used, and there is no particular limitation. As a light-emitting element involved in this invention, in terms of flexibility, good wearing comfort, and the ability to reduce wavelength deviation and brightness deviation, it is preferable to use an organic electroluminescent element (organic electroluminescent diode: also known as "OLED", "organic EL element" and "organic photodiode") which is a light-emitting element composed of a planar organic layer.

[0202] In particular, it is preferable to configure an organic EL element that emits red light to have a wavelength conversion filter that converts the visible light of the organic EL element into near-infrared light (IR).

[0203] Alternatively, LEDs and light guide plates can be used as planar light-emitting elements. In backlights for displays, LEDs are positioned at the edge of the light guide plate and light is incident from the side edge of the light guide plate.

[0204] Furthermore, micro-LEDs can be used with a resolution that allows them to be viewed as a surface. To mitigate the localized increase in brightness directly above the LED, a scattering layer is more preferable.

[0205] (Organic EL element)

[0206] As an organic EL element suitable for the present invention, an example structure can be described as follows: an anode and a cathode are provided on a flexible resin substrate, and an organic functional layer group including a light-emitting layer is sandwiched between the anode and the cathode located at opposite positions. Moreover, depending on the purpose, various functional layers such as sealing components, gas barrier layers, and light extraction layers can be appropriately combined to form the structure.

[0207] The following are representative structural examples of organic EL elements involved in this invention, but the structures of organic EL elements applicable to this invention are not limited to these exemplified structures.

[0208] (1) Anode / Light-emitting layer / Cathode

[0209] (2) Anode / Light-emitting layer / Electron transport layer / Cathode

[0210] (3) Anode / Hole Transport Layer / Light Emitting Layer / Cathode

[0211] (4) Anode / Hole transport layer / Light emission layer / Electron transport layer / Cathode

[0212] (5) Anode / Hole transport layer / Light emission layer / Electron transport layer / Electron injection layer / Cathode

[0213] (6) Anode / Hole Injection Layer / Hole Transport Layer / Light Emitting Layer / Electron Transport Layer / Cathode

[0214] (7) Anode / Hole Injection Layer / Hole Transport Layer / (Electron Blocking Layer / )Emitting Layer / (Hole Blocking Layer / )Electron Transport Layer / Electron Injection Layer / Cathode

[0215] An organic electron laser (EL) element is subjected to an external electric field, injecting holes from the anode into the hole transport layer and electrons from the cathode into the electron transport layer. The injected charge carriers hop between molecules. In the light-emitting layer, holes and electrons recombine to generate electrically neutral excitons. These excitons emit light according to their emission quantum efficiency, thus undergoing radiative deactivation. The light emitted in the organic layer is extracted into the air through the light extraction surface.

[0216] There are no particular limitations on the details of each structural layer constituting the organic EL element applicable to the present invention, or its manufacturing method; known constituent materials and manufacturing methods can be used. For example, reference can be made to the contents described in Japanese Patent Application Publication No. 2013-089608, Japanese Patent Application Publication No. 2014-120334, Japanese Patent Application Publication No. 2015-201508, and International Publication No. 2018 / 51617.

[0217] (Wavelength conversion filter)

[0218] The organic EL element involved in this invention is preferably provided with a wavelength conversion filter that converts the visible light of the organic EL element into near-infrared light.

[0219] In the wavelength conversion filter of the present invention, it is preferable to contain a light-emitting element (such as a luminescent pigment) with wavelength conversion function. The wavelength conversion filter of the present invention is not particularly limited in form or manufacturing method, as long as it contains a luminescent pigment with wavelength conversion function, and is appropriately determined according to the intended use.

[0220] The wavelength conversion filter involved in this invention preferably has the following function: it absorbs light from an organic EL element that emits light in the visible light region (380-780 nm), preferably in the green-red region (495-750 nm), and particularly preferably in the red region (600-700 nm), which includes the near-red light region, and converts it into near-infrared light, for example, near-infrared light that emits light in the region above 700 nm and below 1500 nm and has a maximum emission near 850 nm.

[0221] Regarding wavelength conversion filters and organic EL elements, even if they are manufactured separately and then bonded together, wavelength conversion filters can be directly coated and stacked onto the organic EL material. Alternatively, a cutoff filter for removing light emitted without wavelength conversion can be stacked as needed.

[0222] From the viewpoint of miniaturization and maintaining flexibility, the thickness of the wavelength conversion filter involved in the present invention is preferably in the range of 0.01 to 1000 μm, more preferably in the range of 1 to 500 μm, and even more preferably in the range of 10 to 300 μm.

[0223] The wavelength conversion filter involved in this invention may also include, as needed, various known additives such as colorants, light stabilizers, antioxidants, surfactants, flame retardants, inorganic additives, transparentizing agents, ultraviolet absorbers, fillers, and light scattering particles, in addition to luminescent pigments.

[0224] <Light receiving element>

[0225] The light-receiving element involved in this invention functions as a sensor that detects light reflected by the biological tissue from the light-emitting element irradiating the biological tissue and converts it into an electrical signal.

[0226] As a light-receiving element, a planar organic photodiode (OPD) or an organic thin-film solar cell (OPV) is preferred. OPD is particularly preferred in terms of flexibility, good wearing comfort, and reduction of wavelength deviation and brightness deviation.

[0227] (1) Organic photodiode

[0228] As the light-receiving element involved in this invention, conventionally known organic photodiodes (OPDs) can be used.

[0229] For example, in an organic photodiode, the basic components include: an anode made of a transparent conductive film such as ITO (Indium-Tin Oxide) formed on a substrate such as a light-transmitting resin or glass by sputtering or resistance heating evaporation; a photoelectric conversion layer formed on the anode by depositing an electron donor layer and an electron acceptor layer respectively by resistance heating evaporation; and a cathode made of metal formed on top of the anode by resistance heating evaporation.

[0230] When an organic photodiode with this structure is illuminated, light absorption occurs in the photoelectric conversion region, forming excitons. Next, charge carriers are separated: electrons move to the cathode via the acceptor layer, and holes move to the anode via the donor layer. This generates an electromotive force between the two electrodes, allowing an electrical signal to be extracted by connecting an external circuit.

[0231] The photoelectric conversion layer can be composed of a single layer or multiple layers. For example, the photoelectric conversion layer can have various combinations such as intrinsic layer (I layer), p-type layer / I layer, I layer / n-type layer, p-type layer / I layer / n-type layer, p-type layer / n-type layer, etc.

[0232] For example, an example having the structure described in Figure 13D of U.S. Patent No. 2017 / 0156651 can be used. Furthermore, regarding various organic materials used in organic photodiodes, please refer to Japanese Patent Application Publication No. 2017-532546 and Japanese Patent Application Publication No. 2006-261172, etc.

[0233] (2) Organic thin-film solar cells

[0234] As the light-receiving element involved in this invention, various forms of organic thin-film solar cells (OPVs) known in the past can also be used.

[0235] For example, it is possible to use a bulk heterojunction type organic photoelectric conversion element having a basic structure in which a photoelectric conversion layer, an anode serving as a transparent electrode, a hole transport layer, a bulk heterojunction layer, an electron transport layer, and a cathode are sequentially stacked on one side of a substrate.

[0236] In addition, it may also have other layers such as a hole blocking layer, an electron blocking layer, an electron injection layer, a hole injection layer, or a smoothing layer.

[0237] Furthermore, the photoelectric conversion layer is a layer that converts light energy into electrical energy, and it is configured as a bulk heterojunction layer formed by uniformly mixing p-type semiconductor materials and n-type semiconductor materials.

[0238] p-type semiconductor materials function as electron donors, while n-type semiconductor materials function as electron acceptors.

[0239] Here, electron donors and electron acceptors are "electron donors and electron acceptors that, upon absorbing light, electrons move from electron donors to electron acceptors to form a pairing (charge separation state) of holes and electrons," not simply supplying or receiving electrons as electrodes, but supplying or receiving electrons through a photoreaction.

[0240] Alternatively, to further improve the utilization rate of sunlight (photoelectric conversion efficiency), a series structure (a structure with multiple bulk heterojunction layers) formed by stacking such photoelectric conversion elements can be constructed.

[0241] As p-type semiconductor materials, various condensed polycyclic aromatic compounds and conjugated system compounds can be listed.

[0242] Examples of n-type semiconductor materials include fullerenes, octaazaporphyrins, perfluorinated p-type semiconductors, naphthalenetetracarboxylic anhydride, naphthalenetetracarboxylic diimide, perylenetetracarboxylic anhydride, perylenetetracarboxylic diimide, and other aromatic carboxylic anhydrides, as well as polymers containing their imide compounds as a backbone.

[0243] 2.4 Example of pulse measurement using a reflective photoplethysmography pulse sensor

[0244] The following describes an embodiment of pulse measurement using the reflective photoplethysmography pulse sensor involved in this invention, as well as the measurement results.

[0245] The reflective photoplethysmography (PPG) sensors 1-11 were designed with variations in the shortest distance h (mm) between the light-emitting element and the light-receiving element. Each sensor was worn on the back of the wrist, the ulnar side of the wrist, the radial side of the wrist, the abdomen, and the chest, and the AC / DC values ​​were measured (described later).

[0246] Specifically, the light-receiving element is "KPD30S (manufactured by Kyoto Semiconductor Co., Ltd.)". As the light-emitting element, a ring-shaped light source is obtained by applying a wavelength conversion film to a 60×80mm OLED panel (peak wavelength 630nm) to obtain a peak wavelength of 790nm, and the ring-shaped light source is obtained by using a mask to cut off light outside the ring.

[0247] The light-receiving element is positioned and fixed at the center of the annular light source. The ring diameter is adjusted so that the shortest distance h between the light-receiving element and the annular light source is as shown in Tables I and II below, resulting in reflective photoplethysmography (PPG) sensors 1-11 with the structure shown in Figure 3. The width (shortest width H) of the annular light source is set to 2 mm, with an amplitude of 25-100 mA / m. 2 The light intensity is adjusted to utilize optimal conditions based on various distances. Additionally, the sensor sensitivity is adjusted according to the light intensity.

[0248] Measurements were taken for 60 seconds at a sampling frequency of 500 Hz and analyzed offline. Sampling was performed using an analog front-end (AFE) "AFE4403EVM (manufactured by Texas Instruments)". AC / DC calculations are as described below, for the dorsal wrist, ulnar wrist, radial wrist, abdominal, and thoracic sides, calculated as the average of 10 measurements for each.

[0249] Regarding the circuit side, bandpass and lowpass filters were used. For data processing, moving averages and frequency domain methods were employed.

[0250] In addition, the thickness t of the human skin (epidermis + dermis) on the back of the wrist is set to 4 (mm), the thickness t of the human skin on the ulnar side of the wrist is set to 3.5 (mm), the thickness t of the human skin on the radial side of the wrist is set to 3 (mm), the thickness t of the human skin on the abdomen is set to 2.5 mm, and the thickness t of the human skin on the chest is set to 2 (mm).

[0251] <About AC / DC>

[0252] As explained in the above-mentioned "<Method for Calculating Oxygen Saturation>", the AC component of absorbance is determined by calculating the difference between the maximum and minimum values ​​of the absorbance change over time, and the DC component of absorbance is determined by calculating the average value of the absorbance change over time. Furthermore, the value of AC / DC is calculated by (AC component of absorbance) / (DC component of absorbance), and is shown in Tables I and II below.

[0253] In this invention, the AC / DC value is preferably in the range of 0.20%, and particularly preferably 0.30% or more. Furthermore, considering the size of the device, it is preferably 0.80% or less in a practical sense.

[0254] [Table 1] Table I

[0255]

[0256] [Table 2]

[0257] Table II

[0258]

[0259] As the results above show, increasing the shortest distance h confirms an increase in the AC / DC value. However, when the shortest distance h is made too large, the attenuation of light intensity is significant, power consumption increases, and the signal strength itself decreases, making it more susceptible to noise.

[0260] Therefore, compared to the comparative example's reflective photoplethysmography (PPG) sensor, the reflective PPG sensor of the present invention, which satisfies equation (1), exhibits reduced light attenuation, no increase in power consumption, and no decrease in signal strength, making it less susceptible to noise. Thus, excellent measurement accuracy is confirmed.

[0261] Industrial availability

[0262] The present invention can be used in a biological state diagnostic system for diagnosis based on biological data obtained using at least a reflective photoplethysmography (PPG) sensor.

[0263] Explanation of symbols

[0264] 1: Organism state diagnostic system; 10: Organism data acquisition unit; 11: Drive unit; 12: Receiver unit; 13: Processing unit; 14S: Processing unit; 14a: First signal processing unit; 14b: Second signal processing unit; 14c: Frame data storage unit; 15: Synthesis unit; 16: Display processing unit; 17: Display unit; 18: Measurement location setting unit; 19: Receiver data storage unit; 20: Diagnostic unit; 30: Operation input unit; 40: Control unit; 40a: CPU; 40b: RAM; 40c: ROM; 14: Detection unit; 41, 42: Light-transmitting substrate; 43: Sealing material; 50: First electrode; 60: Second electrode; 70: Light-receiving element; 80: Light-emitting element; h: Shortest distance between the light-emitting element and the light-receiving element.

Claims

1. A biological state diagnosis system provided with at least a biological data acquisition unit, a reception unit, a processing unit, a processing unit, and a diagnosis unit, wherein, in the acquisition unit, biological data is acquired by a reflection-type photoplethysmography sensor, in the processing unit, the acquired biological data is processed in the most suitable manner, in the processing unit, feature points are extracted from the processed data, and, in the diagnosis unit, the biological state is diagnosed using machine learning based on the extracted biological data, the wearing site of the reflection-type photoplethysmography sensor is the abdomen or the chest, the reflection-type photoplethysmography sensor is provided with a light-emitting element that is an organic EL element and a light-receiving element that is an organic photodiode, and, the light-emitting element irradiates light of two wavelengths of red light and near-infrared light to the wearing site, the light-emitting element and the light-receiving element are arranged in parallel in the planar direction, when the shortest distance between the light-emitting element and the light-receiving element is set to h (mm), the thickness of the skin of the human body, that is, the total thickness of the thickness of the epidermis and the thickness of the dermis of the human body is set to t (mm), and it is set that this thickness is in the range of 0.1 to 4 mm, the shortest distance h (mm) between the light-emitting element and the light-receiving element of at least one set satisfies the following formula (1), by subtracting biological information obtained from a set of a light-emitting element and a light-receiving element whose shortest distance does not satisfy the formula (1) from biological information obtained from a set of a light-emitting element and a light-receiving element whose shortest distance satisfies the formula (1), effective biological information is calculated, in formula (1), t is in the range of 0.1 to 4 mm.

2. The biological state diagnosis system according to claim 1, wherein, in the acquisition unit, the biological data is measured and acquired at all times.

3. The biological state diagnosis system according to claim 1, wherein, in the acquisition unit, two or more kinds of biological data are measured and acquired at the same time.

4. The biological state diagnosis system according to claim 1, wherein, in the acquisition unit, biological data of two or more biological sites is measured and acquired at the same time.

5. The biological state diagnosis system according to any one of claims 1 to 4, wherein, a synthesis unit that synthesizes the acquired biological data to aggregate into one data is further provided.

6. The biological state diagnosis system according to any one of claims 1 to 4, wherein, in the diagnosis unit, a learning model that has been subjected to machine learning using biological data corresponding to attributes as explanatory variables and corresponding biological states corresponding to attributes as target variables is used.

7. The biological state diagnosis system according to any one of claims 1 to 4, wherein, in the diagnosis unit, a learning model that has been subjected to deep learning using the acquired biological data is used.

8. The biological state diagnosis system according to claim 1, wherein, ​ ​ ​ ​ Formula (1): (t x 0.7) 2 ≤ h ≤ (t x 1.3) 2 , ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ The shortest distance between the light emitting element and the light receiving element is in the range of 2 to 25 mm.

9. The biological state diagnosis system according to claim 1 or 8, wherein A plurality of the light emitting elements are provided, and each light emitting element is arranged on a substantially concentric circle with the center point of the light receiving element as the center.

Images