Estimation device, learning model, estimation method, learning model generation method, biological data processing system, biological data processing method, and computer program

Abstract

The present disclosure provides an estimation device, a learning model, an estimation method, a learning model generation method, a biological data processing system, a biological data processing method, and a computer program. The estimation device estimates biologically relevant state information of a subject. The estimation device is provided with: an estimation unit that inputs input information to a learning model and acquires output information from the learning model; and a storage unit that stores the output information. The output information includes the biologically relevant state information. The biologically relevant state information includes information representing the mental and physical states of the subject or information representing an environmental state that may affect the mental and physical states of the subject. The input information includes at least cumulative data calculated by accumulation processing performed on a plurality of pieces of biological data of the subject which are arranged in time series. The accumulation processing indicates processing of accumulating the biological data under the condition in that a value does not diverge due to the accumulation.

Description

Estimation device, learning model, estimation method, learning model generation method, biological data processing system, biological data processing method, and computer program

[0001] The present disclosure relates to an estimation device, a learning model, an estimation method, a learning model generation method, a biometric data processing system, a biometric data processing method, and a computer program.

[0002] The information processing system described in Patent Document 1 detects signs of drowsiness based on a cumulative pulse rate (cumulative value). Specifically, a calculation unit calculates the cumulative pulse rate by accumulating trend values ​​similar to the RRI value. The cumulative pulse rate can detect an accumulative situation rather than an instantaneous change.

[0003] Japanese Patent Application Laid-Open No. 2021-145851

[0004] However, the information processing system described in Patent Document 1 does not mention the possibility of the cumulative value diverging. In particular, a diverging cumulative value may have difficulty representing the homeostatic mechanism of a living organism. In other words, a diverging cumulative value may have difficulty reflecting the homeostasis of a living organism. Homeostasis is the tendency to maintain a constant mental and physical state in the face of changes within the living organism and changes in the living organism's external environment. Furthermore, when using a cumulative value that does not easily reflect the homeostasis of a living organism, it is not necessarily likely that reliable information related to the living organism can be obtained.

[0005] The object of the present disclosure is to provide an estimation device, a learning model, an estimation method, a learning model generation method, and a computer program that can obtain reliable information related to a living organism by utilizing cumulative data that reflects the homeostasis of the living organism.

[0006] Another object of the present disclosure is to provide a biological data processing system, a biological data processing method, and a computer program that can derive cumulative data that reflects the homeostasis of a living organism.

[0007] According to the present disclosure, an estimation device can be provided that estimates bio-related state information of a subject, the estimation device comprising: an estimation unit that inputs input information to a learning model and acquires output information from the learning model; and a memory unit that stores the output information, wherein the output information includes the bio-related state information, and the bio-related state information includes information representing the mental and physical state of the subject, or information representing the state of the environment that affects the mental and physical state of the subject, and the input information includes at least cumulative data calculated by cumulative processing of multiple bio-data of the subject arranged in time series, and the cumulative processing indicates a process of accumulating the bio-data under the condition that values ​​do not diverge due to accumulation.

[0008] According to the present disclosure, a learning model can be provided that causes a computer to function to estimate bio-related state information of a subject, the learning model causing the computer to function to input input information and output output information, the output information including the bio-related state information, the bio-related state information including information representing the mental and physical state of the subject, or information representing the state of the environment that affects the mental and physical state of the subject, the input information including at least cumulative data calculated by cumulative processing of multiple bio-data of the subject arranged in time series, and the cumulative processing indicating a process of accumulating the bio-data under the condition that values ​​do not diverge due to accumulation.

[0009] According to the present disclosure, an estimation method can be provided for estimating bio-related state information of a subject, the estimation method comprising the steps of inputting input information into a learning model and acquiring output information from the learning model to which the input information has been input, wherein the output information includes the bio-related state information, and the bio-related state information includes information representing the mental and physical state of the subject, or information representing the state of the environment that affects the mental and physical state of the subject, and the input information includes at least cumulative data calculated by cumulative processing of multiple bio-data of the subject arranged in time series, and the cumulative processing indicates a process of accumulating the bio-data under conditions where values ​​do not diverge due to accumulation.

[0010] According to the present disclosure, it is possible to provide a computer program that causes a computer to execute the above estimation method.

[0011] According to the present disclosure, a learning model generation method can be provided, comprising the steps of: acquiring a learning dataset; and performing learning using the learning dataset to generate a learning model that outputs output information when input information is input, wherein the learning dataset includes at least cumulative data calculated by cumulative processing of multiple biological data of a learning subject arranged in chronological order; and biological-related state information of the learning subject, wherein the biological-related state information of the learning subject includes information representing the mental and physical state of the learning subject, or information representing an environmental state that affects the mental and physical state of the learning subject; the input information includes at least cumulative data calculated by cumulative processing of multiple biological data of the subject arranged in chronological order; the output information includes the biological-related state information of the subject, and the biological-related state information of the subject includes information representing the mental and physical state of the subject, or information representing an environmental state that affects the mental and physical state of the subject; and the cumulative processing represents a process of accumulating the biological data under the condition that values ​​do not diverge due to the accumulation.

[0012] According to the present disclosure, a computer program can be provided that causes a computer to execute the above learning model generation method.

[0013] According to the present disclosure, a biometric data processing system can be provided that includes a processing unit that acquires multiple pieces of biometric data arranged in chronological order, and that performs accumulation processing on the multiple pieces of biometric data, where the accumulation processing indicates a process of accumulating the biometric data under the condition that the values ​​do not diverge due to accumulation.

[0014] According to the present disclosure, a biometric data processing method can be provided, which includes the steps of acquiring multiple pieces of biometric data arranged in chronological order and performing accumulation processing on the multiple pieces of biometric data, wherein the accumulation processing indicates a process of accumulating the biometric data under the condition that the values ​​do not diverge due to the accumulation.

[0015] According to the present disclosure, a computer program can be provided that causes a computer to execute steps of acquiring multiple pieces of biometric data arranged in chronological order and performing accumulation processing on the multiple pieces of biometric data, wherein the accumulation processing indicates a process of accumulating the biometric data under the condition that the values ​​do not diverge due to the accumulation.

[0016] According to the present disclosure, an estimation device, a learning model, an estimation method, a learning model generation method, and a computer program are provided that can obtain highly reliable information related to a living organism by utilizing cumulative data that reflects the homeostasis of the living organism.

[0017] According to the present disclosure, it is possible to provide a biological data processing system, a biological data processing method, and a computer program capable of deriving cumulative data that reflects the homeostasis of a living organism.

[0018] 1 is a block diagram showing an example of a configuration of a biometric data processing system according to a first embodiment of the present disclosure. FIG. 2 is a flowchart showing an example of a biometric data processing method according to the first embodiment. FIG. 3 is a block diagram showing an example of a configuration of a biometric data processing device according to the first embodiment. FIG. 4 is a graph showing an example of biometric data calculated by a preprocessing unit according to the first embodiment. FIG. 5 is a graph for explaining a classifying unit according to the first embodiment. FIG. 6 is a graph showing an example of first biometric data and second biometric data calculated by the classifying unit according to the first embodiment. FIG. 7 is a graph showing an example of corrected data calculated by a correcting unit according to the first embodiment. FIG. 8 is a graph showing an example of accumulated data calculated by an accumulating unit according to the first embodiment. FIG. 9 is a graph showing an example of accumulated data after statistical processing calculated by a statistical processing unit according to the first embodiment. FIG. 10 is a flowchart showing an example of accumulation processing of biometric data by the biometric data processing device according to the first embodiment. FIG. 11 is a diagram showing an example of a configuration of a biometric data analysis device according to the first embodiment. FIG. 12 is a graph showing another example of accumulated data after statistical processing calculated by a statistical processing unit according to the first embodiment. FIG. 13 is a graph showing an example of a result of frequency analysis by a first analyzing unit according to the first embodiment. FIG. 14 is a flowchart showing an example of analysis processing of accumulated data by the biometric data analysis device according to the first embodiment. FIG. 15 is a graph showing an example of biometric data according to a second modified example of the first embodiment. 1 is a graph showing an example of first biometric data and second biometric data calculated by a classifying unit according to a second modified example of embodiment 1. FIG. 2 is a graph showing an example of corrected data calculated by a correcting unit according to the second modified example of embodiment 1. FIG. 3 is a graph showing an example of accumulated data calculated by an accumulating unit according to the second modified example of embodiment 1. FIG. 4 is a graph showing an example of accumulated data after statistical processing calculated by a statistical processing unit according to the second modified example of embodiment 1. FIG. 5 is a graph showing an example of biometric data according to a fourth modified example of embodiment 1. FIG. 6 is a graph showing an example of accumulated data according to the fourth modified example of embodiment 1. FIG. 7 is a block diagram showing an example of a configuration of an information processing system according to a second modified example of embodiment of the present disclosure. FIG. 8 is a diagram showing an example of a training dataset according to the second modified example. FIG. 9 is a diagram showing an example of input information and output information according to the second modified example. FIG. 10 is a block diagram showing an example of a configuration of an estimation system according to the second modified example.1 is a flowchart showing an example of an estimation method executed by an estimation system according to embodiment 2. FIG. 2 is a block diagram showing an example of the configuration of an estimation device according to embodiment 2. FIG. 3 is a diagram showing typically an example of a deep neural network according to embodiment 2. FIG. 4 is a diagram showing typically another example of a deep neural network according to embodiment 2. FIG. 5 is a diagram showing typically yet another example of a deep neural network according to embodiment 2. FIG. 6 is a flowchart showing an example of estimation processing according to embodiment 2. FIG. 7 is a block diagram showing an example of the configuration of a learning device according to embodiment 2. FIG. 8 is a flowchart showing an example of a learning method by the learning device according to embodiment 2. FIG. 9 is a block diagram showing an example of the configuration of a training data creation device according to embodiment 2. FIG. 10 is a flowchart showing an example of a training data creation method by the training data creation device according to embodiment 2.

[0019] Preferred embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configurations are designated by the same reference numerals, and redundant description will be omitted.

[0020] (First Embodiment) FIG. 1 is a diagram illustrating an example configuration of a biometric data processing system 1 according to a first embodiment of the present disclosure. As illustrated in FIG. 1, the biometric data processing system 1 receives biometric data from a client system 100. The biometric data processing system 1 then accumulates the biometric data to calculate cumulative data. In this case, the biometric data processing system 1 executes a process for suppressing post-accumulation divergence of the biometric data before accumulation. As a result, the biometric data processing system 1 can derive cumulative data that reflects the homeostasis of a living organism, which attempts to maintain a constant mental and physical state. This point will be described in detail later. In this specification, the term "mind and body" refers to the mind and / or body.

[0021] The biometric data processing system 1 is connected to a network NW. The network NW includes, for example, the Internet, a closed network, a public telephone network, a local area network (LAN), and a short-distance wireless network.

[0022] At least one client system 100 is connected to the network NW. The client system 100 includes a cloud server 101, a plurality of first terminals 102, and a plurality of biometric information detection devices 103. The cloud server 101, the first terminals 102, and the biometric information detection devices 103 are connected to the network NW. A plurality of second terminals 200 are also connected to the network NW. In the following description, when it is not necessary to distinguish between the first terminal 102 and the second terminal 200, the first terminal 102 and the second terminal 200 may be collectively referred to as "terminal TM."

[0023] The biometric information detection device 103 detects biometric information (the subject's biological condition) from the subject. The biometric information detection device 103 includes, for example, a sensor that detects biometric information (biological condition) contactlessly or non-contactly. The biometric information detection device 103 includes, for example, an optical sensor (a light-emitting element and a light-receiving element) that performs photoplethysmography (PPG). In this case, for example, the subject's pulse waveform is detected. Therefore, the biometric information detection device 103 calculates the pulse rate (bpm) based on the pulse waveform. The optical sensor may measure arterial oxygen saturation. The biometric information detection device 103 may include, for example, electrodes that measure an electrocardiogram waveform. The biometric information detection device 103 may include, for example, a sensor that detects the subject's body temperature or skin temperature. The biometric information detection device 103 includes, for example, an acceleration sensor and / or a gyro sensor. In this case, for example, the subject's number of steps and / or exercise intensity (METs) are detected. That is, the biological information detection device 103 calculates the number of steps and / or exercise intensity (METs) based on the output of the acceleration sensor and / or gyro sensor. The biological information detection device 103 may include, for example, an acceleration sensor and / or a gyro sensor and a microphone. In this case, for example, the sleep state of the subject is detected.

[0024] The biometric information detection device 103 generates multiple pieces of time-series biometric data indicating biometric information. The biometric data will be described later. The biometric information detection device 103 is, for example, a wearable device. The wearable device is, for example, a wristwatch type, a ring type, or a sticker type. The biometric information detection device 103 is synchronized with the first terminal 102 and transmits the biometric data to the first terminal 102. The first terminal 102 is, for example, a mobile terminal such as a smartphone. The biometric information detection device 103 may transmit the biometric data to the cloud server 101 or the biometric data processing system 1 via the network NW.

[0025] The biological information detection device 103 may be a wearable type, a portable type, a tabletop type, or a dedicated detector (measuring device) for detecting biological information. The first terminal 102 may have some or all of the functions of the biological information detection device 103. For example, the first terminal 102 may be a personal computer (PC).

[0026] The first terminal 102 transmits the biometric data to the cloud server 101 via the network NW. The cloud server 101 transmits the biometric data to the biometric data processing system 1 via the network NW. The biometric data processing system 1 processes the biometric data.

[0027] It is also possible to have the first terminal 102 transmit the biometric data to the biometric data processing system 1 via the network NW without providing the cloud server 101 .

[0028] More specifically, the biometric data processing system 1 includes a relay server 2, a first database 3, a biometric data processing device 4, a second database 5, a biometric data analysis device 6, and an information providing server 7. The relay server 2, the first database 3, the biometric data processing device 4, the second database 5, the biometric data analysis device 6, and the information providing server 7 are connected to a network NW. The biometric data processing device 4 and the biometric data analysis device 6 are, for example, servers.

[0029] Each of the relay server 2 and the information providing server 7 includes a control unit, a communication unit, and a memory unit, and may also include an input unit and a display unit. The hardware configurations of the control unit, communication unit, memory unit, input unit, and display unit are similar to the hardware configurations of the control unit 41, communication unit 44, memory unit 45, input unit 42, and display unit 43 of the biometric data processing device 4 shown in FIG. 3 , which will be described later. The first database 3 and the second database 5 include at least a storage device such as a hard disk drive. The first database 3 and the second database 5 may have the same hardware configuration as the relay server 2 or the information providing server 7. The biometric data processing system 1 does not necessarily include all or some of the relay server 2, the first database 3, the second database 5, the biometric data analysis device 6, and the information providing server 7.

[0030] 2 is a flowchart showing an example of a data processing method executed by the biometric data processing system 1. The data processing method includes steps S1 to S7.

[0031] 1 and 2 , first, in step S1, the relay server 2 receives multiple pieces of time-series biometric data for each subject from the cloud server 101. The relay server 2 is equipped with, for example, an API (Application Programming Interface).

[0032] Next, in step S2, the first database 3 stores the plurality of time-series biometric data received by the relay server 2 for each subject.

[0033] Next, in step S3, the biometric data processing device 4 (controller 41, described below) acquires multiple time-series biometric data stored in the first database 3 for each subject. The biometric data processing device 4 (controller 41, described below) then performs an accumulation process on the multiple time-series biometric data stored in the first database 3 for each subject. The accumulation process refers to a process of accumulating biometric data under the condition that values ​​do not diverge due to accumulation. By performing such an accumulation process, cumulative data reflecting the homeostasis of the subject can be derived. Specifically, the accumulation process refers to a process of classifying the biometric data according to the subject's physical and mental state, correcting the classification results to adjust the balance of values, and accumulating the corrected results along the time axis. In this way, by performing a series of processes including classification, correction, and accumulation, cumulative data reflecting the homeostasis of the subject can be derived. More specifically, the biometric data processing device 4 performs a classification process and a correction calculation on the multiple time-series biometric data stored in the first database 3 for each subject. The biometric data processing device 4 then accumulates the multiple time-series biometric data that have been classified and corrected, and calculates multiple cumulative time-series data. By performing the classification and correction, it is possible to derive cumulative data that reflects the homeostasis of the living body. This will be described in detail later.

[0034] Next, in step S4, the second database 5 stores, for each subject, a plurality of time-series cumulative data calculated by the biometric data processing device 4.

[0035] Next, in step S5, the biological data analysis device 6 analyzes the plurality of time-series cumulative data stored in the second database 5 for each subject.

[0036] Next, in step S6, the second database 5 stores, for each subject, information indicating the analysis results of the plurality of time-series cumulative data by the biological data analysis device 6.

[0037] Next, in step S7, the information providing server 7 transmits information indicating the cumulative data and / or analysis results stored in the second database 5 to the second terminal 200 or the first terminal 102 via the network NW. The second terminal 200 is, for example, a terminal of a medical institution, a health management institution, or a research institution. The first terminal 102 is, for example, a terminal of the subject. When step S7 is completed, the data processing method ends.

[0038] 1, the information indicating the accumulated data and / or the analysis results is transmitted to the first terminal 102 via the cloud server 101, but may also be transmitted directly to the first terminal 102. The information providing server 7 corresponds to an example of an "output control unit" of the present disclosure.

[0039] Fig. 3 is a block diagram showing an example configuration of the biometric data processing device 4 of Fig. 1. As shown in Fig. 3, the biometric data processing device 4 includes a control unit 41, an input unit 42, a display unit 43, a communication unit 44, and a storage unit 45. The control unit 41 corresponds to an example of a "processing unit" in the present disclosure.

[0040] The input unit 42 is an input device for inputting various information to the control unit 41. For example, the input unit 42 is a keyboard and pointing device, or a touch panel.

[0041] The display unit 43 displays various types of information and is, for example, a liquid crystal display or an organic electroluminescence display.

[0042] The communication unit 44 is connected to the network NW. The communication unit 44 communicates with external devices connected to the network NW. The communication unit 44 is a communication device that performs communication in accordance with a predetermined communication protocol, and includes, for example, a network interface controller. The predetermined communication protocol is, for example, a protocol conforming to Ethernet (registered trademark), the Internet Protocol Suite, or a protocol conforming to a short-range wireless communication standard.

[0043] The storage unit 45 includes one or more storage devices and stores data and computer programs. The storage unit 45 includes a main storage device such as a semiconductor memory and an auxiliary storage device such as a semiconductor memory and a hard disk drive. The storage unit 45 may also include removable media such as an optical disk. The storage unit 45 may also be, for example, a non-transitory computer-readable storage medium.

[0044] The control unit 41 executes various calculations. The control unit 41 includes one or more processors. The processors are a central processing unit (CPU), a graphics processing unit (GPU), a field programmable gate array (FPGA), a digital signal processor (DSP), or an application specific integrated circuit (ASIC). The processors may be operated by a computer program or by hardwired logic.

[0045] Specifically, the control unit 41 includes a classification unit 411, a correction unit 412, and an accumulation unit 413. The control unit 41 may include at least one of a pre-processing unit 410 and a statistical processing unit 414. For example, the control unit 41 functions as the pre-processing unit 410, the classification unit 411, the correction unit 412, the accumulation unit 413, and the statistical processing unit 414 by executing a computer program stored in the storage unit 45.

[0046] Next, the biometric data processing device 4 will be described in detail with reference to Figures 1 to 10. Hereinafter, the biometric data that the biometric data processing system 1 receives from the client system 100 will be referred to as "original biometric data D0," and the biometric data that the classification unit 411 processes will be referred to as "biometric data D1." The "original biometric data D0" is data output by the biometric information detection device 103 in Figure 1.

[0047] The original biometric data D0 and the biometric data D1 are vital information that directly or indirectly indicates the vital signs of a living organism, or behavioral information that directly or indirectly indicates the behavior of a living organism. The living organism is typically a human. However, the living organism may also be an animal other than a human (e.g., another mammal).

[0048] The vital information is, for example, heart rate information, which is information relating to the heart rate, blood pressure information, respiration information, body temperature information, blood information, or brain wave information.

[0049] In this specification, the term "heartbeat" is not limited to a heartbeat based on an electrocardiogram waveform acquired by electrocardiography (ECG), but also includes a pulse based on a pulse waveform. In other words, in this specification, both a heartbeat based on an electrocardiogram waveform and a pulse based on a pulse waveform are treated as "heartbeats." The method for acquiring the pulse waveform is not particularly limited, and it may be acquired by, for example, PPG.

[0050] The heart rate information may be, for example, the heart rate (HR), the R-R interval, a time domain index related to the heart rate, a frequency domain index related to the heart rate, a nonlinear index related to the heart rate, electrocardiogram waveform information, or pulse wave information. The heart rate is the number of times the heart beats within a certain period of time, and is expressed, for example, in beats per minute (bpm). The R-R interval is the time interval from one QRS wave to the next QRS wave in the electrocardiogram waveform. The time domain index related to the heart rate, the frequency domain index related to the heart rate, and the nonlinear index related to the heart rate are collectively referred to as heart rate variability indexes (HRV indexes). Note that in this specification, for example, the pulse rate (PR) is treated as the heart rate, the pulse interval (PI) is treated as the R-R interval, and pulse rate variability is treated as heart rate variability. Note that HRV stands for heart rate variability.

[0051] The time domain index may be, for example, SDNN (Standard deviation of NN intervals), RMSSD (Root Mean Square of Successive Differences), CVRR (Coefficient of Variation of RR intervals), SDRR (Standard deviation of RR intervals), SDANN (Standard Deviation of the Average NN intervals for each 5-minute segment of a 24-hour HRV recording), SDNN index, NN50 (the number of pairs of successive NN intervals that differ by more than 50 ms), pNN50 (the proportion of NN50 divided by the total number of NN intervals), HR Max, HR Min, (HR Max-HR Min), HTI (HRV Triangular Index), or TINN (Triangular Interpolation of the NN Interval Histogram). The RMSSD is the square root of the mean of the squares of the differences between consecutive adjacent RR intervals.

[0052] The frequency domain index is, for example, LF (power or peak of low frequency components), HF (power or peak of high frequency components), LF / HF, Total Power, LF Norm, HF Norm, ULF (power of extremely low frequency components), or VLF (power of very low frequency components) of heart rate variability.

[0053] Examples of nonlinear indices include entropy, SD1 (standard deviation in the direction perpendicular to y=x in a Poincaré plot), SD2 (standard deviation in the direction along y=x in a Poincaré plot), SD1 / SD2, ApEn (approximate entropy), SampEn (sample entropy), DFA α1 (Detrended Fluctuation Analysis, which describes short-term fluctuations), DFA α2 (Detrended Fluctuation Analysis, which describes long-term fluctuations), (DFA α1) / (DFA α2), CVI (Cardiac Vagal Index), or CSI (Cardiac Sympathetic Index).

[0054] The respiratory information is, for example, a respiratory rate, a respiratory rate, or a respiratory volume. The body temperature information is, for example, a skin temperature, a body temperature, or a core body temperature. The blood information is, for example, an arterial blood oxygen saturation (e.g., SpO ２ ) or blood glucose level. The electroencephalogram information is, for example, an electroencephalogram or electroencephalogram attribute information. The electroencephalogram attribute information indicates attribute values ​​that can be acquired from an electroencephalogram. The electroencephalogram attribute information is, for example, sleep duration or sleep depth.

[0055] The behavior information is, for example, information on the number of steps, sleep information, information on exercise time, information on exercise intensity, information on consumed energy, body movement information, or body physical quantity information.

[0056] Information about the number of steps and information about exercise intensity are types of exercise amount information. The number of steps and exercise intensity may be collectively referred to as exercise amount. Information about the number of steps indicates, for example, the number of steps per minute (spm) within a certain period of time (e.g., one minute). Information about exercise intensity is indicated, for example, by METs (Metabolic equivalents). METs indicate, for example, the number of METs in a certain period of time (e.g., one minute). METs is a unit that represents the intensity of physical activity as a multiple of the intensity at rest. Note that information about exercise intensity is not limited to METs, and may also be indicated, for example, by a relative value to maximum oxygen intake (%VO2max) or a method based on maximum heart rate (%HRmax, %MHR).

[0057] Sleep information is information related to the sleep of a living organism. The sleep information includes at least one of information related to sleep duration, information related to sleep depth, information related to sleep rhythm, and information related to sleep quality based on body movement information. The sleep information indicates, for example, sleep depth, sleep duration, time of falling asleep, time of waking up, or the midpoint between time of falling asleep and time of waking up. Body movement information is information indicating the body movement of a living organism. The body movement information is, for example, the amount and / or frequency of a specific body movement (e.g., turning over in sleep). Body physical quantity information is a physical quantity resulting from body movement. The physical quantity is, for example, acceleration or speed.

[0058] The original biometric data D0 and the biometric data D1 may be data obtained by subjecting vital information or behavioral information to primary processing. Examples of primary processing include logarithmic transformation, sign reversal, removal of outliers, or missing data completion. Furthermore, the original biometric data D0 and the biometric data D1 may be data obtained by subjecting vital information or behavioral information to predetermined processing. The predetermined processing refers to a process of performing statistical processing on the vital information or behavioral information for each of a plurality of time intervals arranged on a time axis and calculating a statistical index for each time interval. The statistical index may be, for example, a mean value, a variance, a standard deviation, or a mean squared continuous error. The time interval is not particularly limited and may be, for example, in units of seconds, minutes, days, weeks, or months. Furthermore, the original biometric data D0 and the biometric data D1 may be data obtained by subjecting vital information or behavioral information to primary processing and then predetermined processing, or data obtained by subjecting the vital information or behavioral information to predetermined processing and then primary processing.

[0059] Furthermore, for information X of the vital information and behavioral information that cannot be directly obtained as time-series information, information X is calculated from multiple pieces of original data Y in each of multiple time intervals (similar to the time intervals described above) arranged on the time axis. As a result, multiple pieces of information X arranged in time series are obtained. Information X is, for example, a time domain index related to the heartbeat, a frequency domain index related to the heartbeat, or a nonlinear index related to the heartbeat. Original data Y is, for example, an R-R interval.

[0060] In the following description, unless otherwise specified, the original biometric data D0 is electrocardiogram waveform data of a subject, and the biometric data D1 is data indicating an R-R interval. The R-R interval is the time interval from one QRS wave to the next QRS wave in an electrocardiogram waveform.

[0061] First, the preprocessing unit 410 will be described with reference to FIGS. 1, 3, and 4. As shown in FIGS. 1 and 3, the preprocessing unit 410 acquires multiple pieces of original biometric data D0 arranged in chronological order from the first database 3. The preprocessing unit 410 then performs preprocessing operations on the multiple pieces of original biometric data D0 to calculate multiple pieces of biometric data D1 from the multiple pieces of original biometric data D0. The biometric data D1 is stored in the first database 3 or the second database 5. The preprocessing operations are operations that generate, from the multiple pieces of original biometric data D0, time-series biometric data D1 to be classified by the classification unit 411. Note that if the original biometric data D0 is the very data to be classified by the classification unit 411, the preprocessing operations are not performed by the preprocessing unit 410. In this case, the classification unit 411 processes the original biometric data D0 as biometric data D1.

[0062] As an example, the preprocessing calculation is a calculation for generating data indicating the R-R interval from the electrocardiogram waveform data. Fig. 4 is a graph showing a plurality of biological data D1 arranged in time series. The horizontal axis represents time t (minutes), and the vertical axis represents the R-R interval (milliseconds).

[0063] Next, the classification unit 411 will be described with reference to Fig. 3, Fig. 5, and Fig. 6. Fig. 5 is a graph for explaining the classification unit 411. The horizontal axis represents time t (minutes), and the vertical axis represents the R-R interval (milliseconds).

[0064] 3 and 5, the classification unit 411 classifies multiple pieces of chronologically arranged biometric data D1 into a first state group G1 indicating a first mental and physical state and a second state group G2 indicating a second mental and physical state. The first state is different from the second state. In FIGS. 5 to 7, black circles indicate the first state and white diamonds indicate the second state.

[0065] The first state is different from the second state. Specifically, the first state indicates a mental or physical state that is incompatible with the second state. For example, the first state may indicate a mental or physical state that is opposite to the second state. For example, the first state indicates either an active or inactive state of the mind or body. The active state includes, for example, a "stress" state, a "fatigue" state, an "exercise" state, an "active" state, or a "sympathetic dominant" state. The inactive state includes, for example, a "relaxation" state, a "recovery" state, a "rest" state, a "quiet" state, or a "parasympathetic dominant" state.

[0066] For example, the first state indicates either a stable state or an unstable state of the mind and body. For example, the first state indicates either an appropriate state or an inappropriate state of the mind and body. The appropriate state includes, for example, a "sufficient" state, an "appropriate load" state, or a "stable" state. The inappropriate state includes, for example, an "excessive load" state or a "deficient" state.

[0067] Below, examples of combinations of the first state and the second state are shown for each type of biometric data D1. When the biometric data D1 is a heart rate or an R-R interval, the first state is a "stress" state or a "fatigue" state, and the second state is a "relaxed" state or a "recovery" state. When the biometric data D1 is the number of steps, the first state is an "exercise" state, and the second state is a "rest" state. When the biometric data D1 is electrocardiogram waveform information, pulse wave information, or acceleration, the first state is an "active" state, and the second state is a "rest" state.

[0068] When the biometric data D1 is "LF / HF of heart rate variability," the first state is a "sympathetic dominant" state, and the second state is a "parasympathetic dominant" state. When the biometric data D1 is "a nonlinear index of heart rate variability (a nonlinear index related to heart rate)," the first state is an "unstable (disorder)" state, and the second state is a "stable" state. When the biometric data D1 is "sleep time," the first state is a "sufficiency" state, and the second state is a "deficiency (debt)" state. When the biometric data D1 is "arterial blood oxygen saturation," the first state is a "stable" state, and the second state is a "deficiency" state.

[0069] Furthermore, if the biometric data D1 is "heart rate" during exercise, the first state is an "excessive load" state and the second state is an "appropriate load" state. For diseases in which excessive load on the body can cause attacks, inflammation, etc., and insufficient exercise can cause reduced cardiopulmonary function and a limited range of joint motion, the "heart rate" is limited to "during exercise" in order to evaluate the degree of accumulation of daily exercise load. If the biometric data D1 is "arterial blood oxygen saturation" during a period in which the number of steps is less than a certain number, the first state is a "stable" state and the second state is a "deficient" state. In addition to oxygen consumption during exercise, the "arterial blood oxygen saturation" is limited to "a period in which the number of steps is less than a certain number" in order to evaluate the state of adaptation to a low-oxygen environment, such as heat stroke, respiratory attacks, and sleep apnea.

[0070] In the example of FIG. 5, the first state indicates an active state of mind and body (for example, stress or fatigue), and the second state indicates an inactive state of mind and body (for example, relaxation or recovery).

[0071] The first state and the second state may be states that indicate different levels of the biological state. For example, the "REM sleep" state and the "non-REM sleep" state indicate different levels of the mental and physical state. In this way, the first state indicates a mental and physical state that is incompatible with the second state.

[0072] Next, the classification unit 411 will be described in detail. The classification unit 411 classifies multiple pieces of biometric data D1 into a first status group G1 and a second status group G2, thereby acquiring (calculating) multiple pieces of first biometric data B1 belonging to the first status group G1 and multiple pieces of second biometric data B2 belonging to the second status group G2. The first biometric data B1 has one of positive and negative signs. The second biometric data B2 has the other of positive and negative signs. FIG. 6 is a graph showing the first biometric data B1 and the second biometric data B2 arranged in chronological order. The horizontal axis represents time t (minutes), and the vertical axis represents the relative value of the R-R interval (= D1 - C1). In the example of FIG. 6, the first biometric data B1 is a positive value with a positive sign. The second biometric data B2 is a negative value with a negative sign.

[0073] 5 , the classification unit 411 calculates the first classification standard information C1 based on the plurality of biometric data D1 during a specific period T1. The specific period T1 is a period for calculating the first classification standard information C1 and a ratio R (described later). The first classification standard information C1 is, for example, the average value, median, mode, or percentile value of the plurality of biometric data D1 during the specific period T1. In the example of FIG. 5 , the first classification standard information C1 indicates the median.

[0074] The classification unit 411 compares the first classification standard information C1 with the biometric data D1 to be classified and classifies the biometric data D1 based on the comparison result. The biometric data D1 to be classified is the biometric data D1 during an observation period T2. The observation period T2 of the biometric data D1 is, in other words, a period during which a plurality of cumulative data D3, described below, is calculated. In this embodiment, as an example, the observation period T2 of the biometric data D1 is the same as the specific period T1 for calculating the first classification standard information C1. As an example, the classification unit 411 subtracts the first classification standard information C1 from each piece of biometric data D1. Then, as shown in FIG. 6 , the classification unit 411 classifies each piece of biometric data D1 into a first status group G1 or a second status group G2 depending on whether each subtraction result (=D1-C1) indicates a positive or negative value. In FIG. 6 , the classification unit 411 sets a subtraction result indicating a positive value to the first biometric data B1 and a subtraction result indicating a negative value to the second biometric data B2.

[0075] The classification unit 411 may subtract each piece of biometric data D1 from the first classification standard information C1.

[0076] Here, the classification unit 411 may include a trained model. The trained model is a computer program. The trained model is constructed by performing machine learning using a plurality of pieces of biometric data D1 arranged in chronological order, a plurality of pieces of first biometric data B1, and a plurality of pieces of second biometric data B2 as training data. The trained model includes, for example, a neural network. The trained model inputs a plurality of pieces of biometric data D1 and outputs a plurality of pieces of first biometric data B1 and a plurality of pieces of second biometric data B2. In other words, the classification unit 411 inputs a plurality of pieces of biometric data D1 to the trained model and acquires a plurality of pieces of first biometric data B1 and a plurality of pieces of second biometric data B2 from the trained model.

[0077] Next, the correction unit 412 will be described with reference to FIGS. 3, 6, and 7. As shown in FIGS. 3 and 6, the correction unit 412 calculates a plurality of time-series corrected data D2 by performing a correction calculation on a plurality of first biometric data B1 and a plurality of second biometric data B2 to be corrected. In this embodiment, the first biometric data B1 and second biometric data B2 to be corrected are the first biometric data B1 and second biometric data B2 during the observation period T2. FIG. 7 is a graph showing the time-series corrected data D2. The horizontal axis represents time t (minutes), and the vertical axis represents the corrected relative value of the R-R interval.

[0078] 6 , the correction calculation is a calculation for adjusting the balance in magnitude between the plurality of first biometric data B1 and the plurality of second biometric data B2. In other words, the correction calculation is a calculation for adjusting the ratio R (=FS2 / FS1) of the sum FS2 of the absolute values ​​of the plurality of second biometric data B2 to the sum FS1 of the absolute values ​​of the plurality of first biometric data B1. Preferably, the correction calculation is a calculation for setting the ratio R (=FS2 / FS1) of the sum FS2 of the absolute values ​​of the plurality of second biometric data B2 to the sum FS1 of the absolute values ​​of the plurality of first biometric data B1 to essentially "1" during the observation period T2.

[0079] In this preferred example, the correction calculation includes a process of calculating the ratio R and a process of correcting the first biometric data B1 or the second biometric data B2 based on the ratio R. That is, the correction unit 412 calculates the ratio R for the observation period T2 (specific period T1). Then, for example, if the sum FS1 is greater than the sum FS2, the correction unit 412 corrects each piece of second biometric data B2 for the observation period T2 by multiplying the reciprocal of the ratio R (1 / R) by the second biometric data B2. Thus, the correction unit 412 obtains multiple corrected pieces of second biometric data B2# as the multiplication result. As a result, after the correction calculation is performed, the ratio R becomes "1." As shown in FIG. 7 , the multiple pieces of first biometric data B1 and the corrected multiple pieces of second biometric data B2# constitute multiple pieces of corrected data D2.

[0080] For example, when the sum FS1 is greater than the sum FS2, the correction unit 412 may multiply each of the first biometric data B1 in the observation period T2 by the ratio R to obtain a plurality of corrected first biometric data B1# (not shown) as a multiplication result. In this case, the corrected plurality of first biometric data B1# and the plurality of second biometric data B2 constitute a plurality of corrected data D2.

[0081] Here, since it is sufficient that the ratio R after the correction calculation is substantially "1," the correction calculation may include a process of calculating a first correction parameter and a second correction parameter based on the ratio (FS1:FS2) between the sum FS1 and the sum FS2, and a process of correcting the first biometric data B1 and the second biometric data B2 based on the first correction parameter and the second correction parameter. In this case, the first biometric data B1 is corrected by multiplying it by the first correction parameter, and the second biometric data B2 is corrected by multiplying it by the second correction parameter. In this case, the corrected plurality of first biometric data B1# and the corrected plurality of second biometric data B2# constitute the plurality of corrected data D2. For example, when FS1:FS2 = 1:2, the first correction parameter is "4" and the second correction parameter is "2."

[0082] The correction unit 412 may also calculate a first approximation function that approximates the plurality of first biometric data B1 and a second approximation function that approximates the plurality of second biometric data B2. The correction unit 412 then analytically calculates a first area based on the first approximation function and analytically calculates a second area based on the second approximation function. Furthermore, the correction unit 412 calculates a ratio R of the absolute value of the second area (which essentially corresponds to the sum of the absolute values ​​of the plurality of second biometric data B2) to the absolute value of the first area (which essentially corresponds to the sum of the absolute values ​​of the plurality of first biometric data B1). The correction unit 412 then corrects the first biometric data B1 or the second biometric data B2 based on the ratio R, as described above. Note that, as described above, the correction calculation may include a process of calculating a first correction parameter and a second correction parameter based on the ratio between the absolute value of the first area and the absolute value of the second area, and a process of correcting the first biometric data B1 and the second biometric data B2 based on the first correction parameter and the second correction parameter.

[0083] The correction unit 412 may divide the observation period T2 into multiple time intervals TG, calculate a representative value of two or more correction data D2 for each time interval TG, and set the representative value as new correction data D2. The representative value may be, for example, an average value, a median value, or a mode value. The time interval TG is not particularly limited and may be, for example, in units of seconds, minutes, days, weeks, or months.

[0084] Next, the accumulator 413 will be described with reference to FIGS. 3, 7, and 8. As shown in FIGS. 3 and 7, the accumulator 413 calculates a plurality of time-series accumulated data D3 by accumulating the time-series corrected data D2 along the time axis. The accumulated data D3 may be referred to as a corrected accumulated index. FIG. 8 is a graph showing the time-series accumulated data D3. The horizontal axis indicates time t (minutes), and the vertical axis indicates the accumulated value of the corrected relative value of the R-R interval. In this embodiment, the corrected data D2 to be accumulated is the corrected data D2 for the observation period T2. Therefore, the classification unit 411 calculates the accumulated data D3 at each time t during the observation period T2. In other words, the accumulated data D3 at each time t indicates the accumulated value of the corrected data D2 from the start of accumulation to that time t.

[0085] As an example, the accumulator 413 calculates a cumulative sum D3j as the cumulative data D3 using equation (1). D3j represents the cumulative sum at time t=j in the observation period T2. D2i represents the corrected data D2 at time t=i. ts represents the start time of the accumulation. The accumulator 413 calculates multiple cumulative sums D3j in time series during the observation period T2 while updating j.

[0086]

[0087] The accumulator 413 may calculate the accumulated data D3 by performing numerical integration on the corrected data D2.

[0088] As described above with reference to FIGS. 3 to 8 , according to this embodiment, the correction calculation by the correction unit 412 is a calculation that adjusts the balance in magnitude between the first biometric data B1 and the second biometric data B2, which represent different states. Therefore, when the accumulator 413 accumulates the corrected data D2 by the correction unit 412, divergence of the accumulated data D3 can be suppressed. Correction that suppresses divergence, i.e., correction that adjusts the balance, can give the accumulated data D3 characteristics similar to the biological mechanism that ensures homeostasis. In other words, it is possible to derive accumulated data D3 that reflects the biological homeostasis. Biological homeostasis refers to the biological tendency to maintain a constant mental and physical state in response to changes within the organism and changes in the external environment.

[0089] In this embodiment, the cumulative data D3 is calculated by accumulating multiple sets of corrected data D2 indicating the first and second mental and physical states on a time axis. Therefore, the cumulative data D3 reflects the mental and physical states (first and second states) accumulated over time. As a result, the cumulative data D3 quantitatively and chronologically represents the mental and physical states (first and second states) accumulated over time.

[0090] For example, living organisms such as humans have a biological mechanism that alternates between a first state, such as stress and fatigue, and a second state, such as relaxation and recovery, thereby ensuring homeostasis, which attempts to maintain a constant mental and physical state. Therefore, if accumulating biological data causes the cumulative data to diverge toward one of the first and second states, the correlation with the biological mechanism is lost. Therefore, in this embodiment, a correction calculation is performed to suppress the divergence of the cumulative data D3 and ensure the correlation between the cumulative data D3 and the biological mechanism. Furthermore, the cumulative data D3 quantitatively and chronologically represents the first state, such as stress and fatigue, and the second state, such as relaxation and recovery, accumulated over time.

[0091] In particular, in this embodiment, the correction calculation refers to a calculation that sets the ratio R of the sum FS2 of the absolute values ​​of the second biometric data B2 to the sum FS1 of the absolute values ​​of the first biometric data B1 to essentially "1." In other words, the correction calculation refers to a calculation that prevents the cumulative data D3 from diverging. In further other words, the correction calculation refers to a calculation that essentially prohibits the cumulative data D3 from diverging. Therefore, it is possible to more effectively prevent the cumulative data D3 from diverging, and to derive cumulative data D3 that better reflects the homeostasis of the living body.

[0092] The ratio R being substantially "1" does not necessarily mean that the ratio R is "1" but may be any ratio R that can reflect constancy in the cumulative data D3. For example, the ratio R may be "0.8 or more and 1.2 or less," or "0.9 or more and 1.1 or less." The closer the ratio R is to "1," the better, but the most preferable case is "1."

[0093] The calculation that substantially prohibits divergence includes not only the case where divergence is prohibited but also the case where divergence is prohibited to the extent that constancy can be reflected in the cumulative data D3. In other words, the calculation that substantially prohibits divergence includes the calculation that suppresses divergence. However, it is most preferable that the correction calculation is the calculation that prohibits divergence.

[0094] In this embodiment, the first state in classification by the classification unit 411 indicates a physical and mental state opposite to the second state. Therefore, for example, the biological mechanism that "a living organism such as a human maintains homeostasis by balancing a first state such as stress and fatigue with a second state such as relaxation and recovery" can be effectively reflected in the correction data D2 and the cumulative data D3.

[0095] In particular, the first state indicates either an active or inactive state of the body or mind, either a stable or unstable state of the body or mind, or either an appropriate or inappropriate state of the body or mind. Furthermore, the second state indicates a state opposite to the first state. Therefore, for example, the biological mechanism that "a living organism such as a human maintains homeostasis by balancing an active state and an inactive state, a stable state and an unstable state, or an appropriate state and an inappropriate state" can be effectively reflected in the correction data D2 and the cumulative data D3.

[0096] Furthermore, in this embodiment, the classification unit 411 compares the first classification standard information C1 with the biometric data D1 and classifies the biometric data D1 based on the comparison result, thereby enabling classification of the biometric data D1 through simple processing.

[0097] Furthermore, in this embodiment, the classification unit 411 compares the biometric data D1 to be classified during the observation period T2, which is the same as the specific period T1 used to calculate the first classification reference information C1, with the first classification reference information C1, and classifies the biometric data D1 based on the comparison result. The correction unit 412 then performs a correction calculation on the multiple first biometric data B1 and multiple second biometric data B2 during the observation period T2 based on the multiple first biometric data B1 and multiple second biometric data B2 during the observation period T2. The accumulation unit 413 then accumulates the correction data D2 to calculate multiple cumulative data D3 in time series at multiple times during the observation period T2. In this manner, in this embodiment, the observation period T2 and the specific period T1 are matched. Therefore, the relative relationship between the first state and the second state within the observation period T2 can be reflected in the cumulative data D3. As a result, for example, periods of high physical and mental stress (first state) can be compared with periods of low physical and mental stress (second state) during the observation period T2.

[0098] Here, the lengths of the specific period T1 and the observation period T2 are not particularly limited, and may be measured in units of seconds, minutes, hours, weeks, months, or years.

[0099] Furthermore, for example, if the observation period T2 changes, the first classification standard information C1 and the ratio R are calculated for the new observation period T2 (i.e., the new specific period T1) even for the same living organism. In this case, for example, multiple consecutive observation periods T2 are set at 30-day intervals, and the calculation of the first classification standard information C1 and the ratio R for 30 days is repeated every 30 days.

[0100] Furthermore, in this embodiment, the specific period T1 and the observation period T2 may be periods during which the mental and physical state is in a specific state. Therefore, the preprocessing unit 410, the classification unit 411, the correction unit 412, and the accumulation unit 413 perform their respective processes during periods during which the mental and physical state is in a specific state. Therefore, the cumulative data D3 is calculated under the special condition of "periods during which the mental and physical state is in a specific state." As a result, cumulative data D3 that is more specifically tailored to the intended use can be obtained. The specific state is not particularly limited, but may be, for example, a sleeping state, a waking state, a state in which the number of steps is less than a certain number, a walking state, a state in which the exercise intensity is equal to or greater than a certain level, a state in which the exercise is being performed, or a state of rest (non-exercise). The specific state may also be a combination of two or more states, such as a state of resting while awake. Thus, the specific state may be, for example, a state related to physical behavior.

[0101] Next, the statistical processing unit 414 will be described with reference to FIGS. 3, 8, and 9. FIG. 9 is a graph showing cumulative data D4 after statistical processing arranged in chronological order. The horizontal axis represents time t (minutes), and the vertical axis represents the statistically processed value relative to the cumulative value. As shown in FIGS. 3, 8, and 9, the statistical processing unit 414 calculates multiple cumulative data D4 after statistical processing by performing statistical processing on multiple cumulative data D3. Therefore, according to this embodiment, it is easy to grasp the trends of the multiple cumulative data D4. The statistical processing may be, for example, a centering process, a standardization process, or a normalization process. The cumulative data D3 to be statistically processed is the cumulative data D3 during the observation period T2.

[0102] Centering is a process of calculating a centered value Cj of cumulative data D3 shown in equation (2). In this case, the centered value Cj is a statistically processed value and represents cumulative data D4 after statistical processing. In equation (2), Cj represents the centered value at time t = j. D3j represents the value of cumulative data D3 at time t = j. Av represents the average value of multiple cumulative data D3 during observation period T2. In centering, difference values ​​are distributed around the average value Av without changing units, making it easy to intuitively grasp cumulative data D4.

[0103] Cj=D3j−Av…(2)

[0104] The standardization process is a process of calculating a standardized value Sj of the cumulative data D3 shown in formula (3). In this case, the standardized value Sj is a statistically processed value and represents the cumulative data D4 after statistical processing. In formula (3), Sj represents the standardized value at time t = j. σ represents the standard deviation of the multiple cumulative data D3 during the observation period T2. In the example of Figure 9, the statistically processed value on the vertical axis is the standardized value. In the standardization process, the unit "1" becomes "1 standard deviation," making the cumulative data D4 easier to understand statistically.

[0105] Sj=(D3j-Av) / σ...(3)

[0106] The normalization process is a process of calculating a normalized value Nj of the cumulative data D3 shown in equation (4). In this case, the normalized value Nj is a statistically processed value and represents the cumulative data D4 after statistical processing. In equation (4), Dmn is the minimum value of the cumulative data D3 during the observation period T2. Dmx is the maximum value of the cumulative data D3 during the observation period T2. In the normalization process, each cumulative data D4 takes a value between 0 and 1. In this case, for example, it can be expressed as a percentage (%), which makes it easier to understand and interpret.

[0107] Nj=(D3j-Dmn) / (Dmx-Dmn)...(4)

[0108] As described above with reference to FIG. 9 , according to this embodiment, statistical processing is performed on multiple cumulative data D3. In this case, as an example, the statistical processing unit 414 performs centering, standardization, or normalization on the cumulative data D3 of a specific living body based on a statistical index calculated from the cumulative data D3 of the specific living body. The statistical index is, for example, the average value Av, the standard deviation σ, the minimum value Dmn, or the maximum value Dmx. According to this preferred example, statistical processing is performed using a statistical index based on the cumulative data D3 of only the specific living body. Therefore, in the cumulative data D4, it is possible to easily determine when a specific living body, for example, an individual subject, is in the first state and when it is in the second state, regardless of other living bodies of the same type.

[0109] As another example, the statistical processing unit 414 may perform centering, standardization, or normalization on the cumulative data D3 of a specific organism among the multiple organisms of the same type, based on a statistical index calculated from the cumulative data D3 of the multiple organisms of the same type. According to this preferred example, the statistical processing is performed using a statistical index based on the cumulative data D3 of a group (multiple organisms of the same type), so that in the time-series cumulative data D4, it is possible to determine when a specific organism, for example, an individual subject, is in the first state and when it is in the second state, based on the group.

[0110] As described above, statistical processing is performed on the cumulative data D3 from various viewpoints, but it is not necessary to perform statistical processing on the cumulative data D3.

[0111] Next, the flow of processing by the biometric data processing device 4 will be described with reference to FIGS. 3 and 10. FIG. 10 is a flowchart showing the "biometric data accumulation processing" of step S3 in FIG. 2. As shown in FIG. 10, the "biometric data accumulation processing" includes steps S11 to S15. A computer program stored in the storage unit 45 causes the control unit 41 to execute steps S11 to S15. In other words, the computer program product realizes steps S11 to S15 when the computer program is executed by the control unit 41. The control unit 41 corresponds to an example of a "computer" in the present disclosure.

[0112] 3 and 10, first, in step S11, the pre-processing unit 410 generates a plurality of pieces of biometric data D1 (FIG. 4) based on a plurality of pieces of original biometric data D0 acquired from the first database 3. That is, the pre-processing unit 410 acquires a plurality of pieces of biometric data D1.

[0113] Next, in step S12, the classification unit 411 classifies the multiple biometric data D1 into a first status group G1 and a second status group G2, thereby calculating multiple first biometric data B1 belonging to the first status group G1 and multiple second biometric data B2 belonging to the second status group G2 (Figure 6).

[0114] Next, in step S13, the correction unit 412 calculates corrected data D2 by executing a correction calculation on the first biometric data B1 and the second biometric data B2 (FIG. 7).

[0115] Next, in step S14, the accumulator 413 accumulates the correction data D2 along the time axis to calculate a plurality of time-series accumulated data D3 (FIG. 8).

[0116] Next, in step S15, the statistical processing unit 414 performs statistical processing on the plurality of cumulative data D3 to calculate cumulative data D4 after statistical processing ( FIG. 9 ). Then, the “biometric data cumulative processing” is completed, and the process proceeds to step S4 in FIG. 2.

[0117] Next, details of the biological data analysis device 6 in Fig. 1 will be described with reference to Fig. 11 to Fig. 14. Fig. 11 is a diagram showing an example of the configuration of the biological data analysis device 6 in Fig. 1. As shown in Fig. 11, the biological data analysis device 6 includes a control unit 61, an input unit 62, a display unit 63, a communication unit 64, and a storage unit 65.

[0118] The hardware configurations of the control unit 61, input unit 62, display unit 63, communication unit 64, and memory unit 65 are similar to the hardware configurations of the control unit 41, input unit 42, display unit 43, communication unit 44, and memory unit 45 in Figure 3, respectively.

[0119] The control unit 61 includes an analysis unit 610 and a support unit 620. For example, the control unit 61 functions as the analysis unit 610 and the support unit 620 by executing a computer program stored in the storage unit 65. The analysis unit 610 includes at least one of a first analysis unit 611, a second analysis unit 612, and a third analysis unit 613. The third analysis unit 613 will be described later in a fifth modified example.

[0120] The first analyzer 611 directly or indirectly performs frequency analysis on a plurality of pieces of time-series cumulative data D3 ( FIG. 8 ). "Direct frequency analysis" refers to directly analyzing the frequency of the cumulative data D3. "Indirect frequency analysis" refers to performing a predetermined process on the cumulative data D3 and performing frequency analysis on the data calculated. In this case, the predetermined process is, for example, statistical processing by the statistical processing unit 414. The first analyzer 611 then performs frequency analysis on a plurality of pieces of statistically processed cumulative data D4 ( FIG. 9 ) calculated by the statistical processing unit 414.

[0121] For example, the first analysis unit 611 performs frequency analysis by applying a discrete Fourier transform, a fast Fourier transform, or an autoregressive model to the cumulative data D3 or the cumulative data D4 after statistical processing.

[0122] According to this embodiment, the cumulative data D3 and cumulative data D4, which represent biological homeostasis, can be analyzed in the frequency domain based on the results of frequency analysis by the first analysis unit 611. As an example, a case will be described in which a woman's menstrual cycle, which has a cycle of 28 days, is calculated. In this case, for example, the observation period T2 (specific period T1) is set to four months. The biological data D1 is the average heart rate during daily sleep. The first analysis unit 611 then performs a discrete Fourier transform on the cumulative data D3 based on the biological data D1 to calculate the power spectral density. As a result, it can be observed that the 28-day cycle wave (power spectral density) is the strongest.

[0123] Another example will be described. Fig. 12 is a graph showing cumulative data D4 after statistical processing. The horizontal axis indicates time t (months), and the vertical axis indicates the statistically processed value. In this case, the statistically processed value indicates a value calculated by performing statistical processing on the cumulative value of the corrected relative values ​​of the R-R interval. In the example of Fig. 12, the statistically processed value is a standardized value.

[0124] The first analyzer 611 calculates the intensity of each wavelength component (each frequency component) by performing a discrete Fourier transform on the multiple pieces of accumulated data D4 in Fig. 12. Fig. 13 is a graph showing the results of the frequency analysis by the first analyzer 611. The horizontal axis represents wavelength (unit: days), and the vertical axis represents the intensity of the wavelength component.

[0125] As can be seen from FIGS. 12 and 13, the first analyzer 611 can detect that the 30-day cycle wave (wavelength component) is the strongest in the RR interval over a one-year period.

[0126] The information providing server 7 (FIG. 1) transmits the result of the frequency analysis by the first analysis unit 611 to the terminal TM, thereby causing the terminal TM to output the result of the frequency analysis as an image or sound (step S7 in FIG. 2). As a result, the user of the terminal TM can understand, use, and analyze the result of the frequency analysis.

[0127] The second analysis unit 612 directly or indirectly analyzes multiple pieces of cumulative data D3 in a time series from the past and evaluates the past or present mental and physical state based on the analysis results. And / or the second analysis unit 612 directly or indirectly analyzes multiple pieces of cumulative data D3 in a time series from the past and predicts the future mental and physical state based on the analysis results. "Direct analysis" refers to directly analyzing the cumulative data D3. "Indirect analysis" refers to analyzing data obtained by performing a predetermined process on the cumulative data D3. In this case, the predetermined process is, for example, statistical processing by the statistical processing unit 414.

[0128] The information providing server 7 transmits information indicating the evaluation result and / or prediction result by the second analysis unit 612 to the terminal TM, thereby causing the terminal TM to output the evaluation result and / or prediction result as an image or sound (step S7 in FIG. 2 ). As a result, according to this embodiment, the user of the terminal TM can understand, use, and analyze the evaluation result and / or prediction result.

[0129] The support unit 620 generates support information for supporting the user based on the result of the frequency analysis by the first analysis unit 611 or the evaluation or prediction result by the second analysis unit 612. The support information is, for example, information that supports the user in maintaining or recovering their health. The support information may include, for example, warning information.

[0130] The information providing server 7 transmits the support information generated by the support unit 620 to the terminal TM, thereby causing the terminal TM to output the support information as an image or sound (step S7 in FIG. 2). As a result, according to this embodiment, the user of the terminal TM can take action based on the support information.

[0131] The information providing server 7 transmits the cumulative data D3 generated by the biometric data processing device 4 and / or information indicating the characteristics of the cumulative data D3 to the terminal TM, thereby causing the terminal TM to output the cumulative data D3 and / or the information indicating the characteristics as an image or sound (step S7 in FIG. 2). As a result, according to this embodiment, the user of the terminal TM can understand, use, and analyze the cumulative data D3 and / or the information indicating the characteristics. The information indicating the characteristics of the cumulative data D3 is, for example, cumulative data D4 after statistical processing.

[0132] Next, the flow of processing by the biological data analysis device 6 will be described with reference to FIGS. 3 and 14. FIG. 14 is a flowchart showing an example of the "accumulated data analysis process" of step S5 in FIG. 2. As shown in FIG. 14, the "accumulated data analysis process" includes steps S21 to S24. A computer program stored in the storage unit 65 causes the control unit 61 to execute steps S21 to S24. In other words, the computer program product realizes steps S21 to S24 when the computer program is executed by the control unit 61. The control unit 61 corresponds to an example of a "computer" in the present disclosure.

[0133] As shown in FIGS. 3 and 14, first, in step S21, the first analyzer 611 directly or indirectly performs frequency analysis on the accumulated data D3.

[0134] Next, in step S22, the second analysis unit 612 directly or indirectly analyzes the cumulative data D3, and evaluates the past or present mental and physical state based on the analysis results.

[0135] Next, in step S23, the second analysis unit 612 directly or indirectly analyzes the cumulative data D3, and predicts the future physical and mental state based on the analysis results.

[0136] Next, in step S24, the support unit 620 generates support information for supporting the user based on the cumulative data D3, the cumulative data D4, the result of the frequency analysis by the first analysis unit 611, the evaluation result by the second analysis unit 612, and / or the prediction result by the second analysis unit 612. Then, the "accumulated data analysis process" is completed, and the process proceeds to step S6 in FIG. 2.

[0137] (First Modification) In the first embodiment, the specific period T1 for calculating the first classification standard information C1 and the ratio R is the same as the observation period T2 of the biometric data D1. In contrast, in a first modification of the first embodiment, the specific period T1 and the observation period T2 are different. The following mainly describes the differences with reference to FIG. 3 .

[0138] Specifically, the classification unit 411 compares the biometric data D1 to be classified during an observation period T2, which is different from the specific period T1, with the first classification standard information C1 calculated during the specific period T1, and classifies the biometric data D1 based on the comparison result. Therefore, during the observation period T2, multiple pieces of first biometric data B1 and multiple pieces of second biometric data B2 to be corrected are calculated.

[0139] Then, the correction unit 412 performs a correction calculation on the plurality of first biometric data B1 and the plurality of second biometric data B2 to be corrected during the observation period T2, based on the plurality of first biometric data B1 and the plurality of second biometric data B2 calculated from the plurality of biometric data D1 during the specific period T1. That is, the correction unit 412 corrects the plurality of first biometric data B1 or the plurality of second biometric data B2 to be corrected during the observation period T2, based on the ratio R calculated from the plurality of first biometric data B1 and the plurality of second biometric data B2 during the specific period T1. Therefore, a plurality of correction data D2 is calculated during the observation period T2.

[0140] The accumulator 413 then accumulates the correction data D2 to calculate multiple pieces of time-series accumulated data D3 at multiple times during the observation period T2.

[0141] As described above, according to the first modification, the accumulated data D3 for the observation period T2 can be derived based on the biometric data D1 for the specific period T1. For example, the biometric data processing device 4 sets the specific period T1 to a certain period before a medical procedure, and sets the observation period T2 to a certain period after the medical procedure. The medical procedure may be, for example, treatment or surgery. Furthermore, for example, the biometric data processing device 4 sets the specific period T1 to "last month" and the observation period T2 to "this month."

[0142] For example, if the specific period T1 is set to a period during which the living body is healthy, the ratio R is calculated as a parameter of homeostasis in the healthy state. Therefore, if the observation period T2 is set to a period during which the living body is ill, classification based on the ratio R in the healthy state is performed during the ill period. As a result, the condition of the ill period can be evaluated based on the healthy state. In this way, the homeostasis parameter (ratio R) established during the specific period T1 can be used to evaluate the disturbance or breakdown of the living body's homeostasis (e.g., illness) from the perspective of homeostasis. From this perspective, even in the first modification, the cumulative data D3 can be considered to indirectly reflect the homeostasis of the living body. Homeostasis may be directly reflected in the cumulative data D3, as in the first embodiment, or indirectly reflected in the cumulative data D3, as in the first modification.

[0143] Furthermore, part or all of the specific period T1 may overlap with part of the observation period T2. For example, the biometric data processing device 4 sets the specific period T1 to "one month in April of each year" and sets the observation period T2 to "the fiscal year from April to March of the following year." Note that, for example, part of the specific period T1 may overlap with the entire observation period T2.

[0144] The length of the specific period T1 and the length of the observation period T2 may be the same or different.

[0145] The specific period T1 is a past period. The observation period T2 is a period that follows the specific period T1 on the time axis. In this case, the observation period T2 is, for example, a past period.

[0146] Furthermore, according to the first modification, the first classification standard information C1 and the ratio R are determined in advance. Therefore, the preprocessing unit 410, the classification unit 411, the correction unit 412, and the accumulating unit 413 can perform their respective processes even during the observation period T2. As a result, the accumulated data D3 can be calculated sequentially even during the observation period T2. In this case, the observation period T2 is a period from the past to the future.

[0147] In particular, during the observation period T2, the preprocessing unit 410, the classification unit 411, the correction unit 412, and the accumulation unit 413 perform real-time processing. As a result, according to the first modification, the accumulated data D3 can be calculated in real time. Real-time processing means that processing is performed immediately in response to reception of the original biometric data D0.

[0148] Specifically, the preprocessing unit 410 calculates biometric data D1 by processing the original biometric data D0 in real time. The classification unit 411 calculates first biometric data B1 and second biometric data B2 by processing the biometric data D1 in real time. The correction unit 412 calculates corrected data D2 by processing the first biometric data B1 and the second biometric data B2 in real time. The accumulation unit 413 calculates accumulated data D3 by processing the corrected data D2 in real time. The statistical processing unit 414 calculates post-statistical-process accumulated data D4 by processing the accumulated data D3 in real time.

[0149] The second analysis unit 612 directly or indirectly analyzes the plurality of pieces of time-series cumulative data D3 or D4 in real time, and evaluates the current mental and physical state based on the analysis results. And / or the second analysis unit 612 directly or indirectly analyzes the plurality of pieces of time-series cumulative data D3 or D4 in real time, and predicts the future mental and physical state based on the analysis results. The "direct analysis" and "indirect analysis" are the same as the "direct analysis" and "indirect analysis" in the first embodiment.

[0150] The information providing server 7 (FIG. 1) transmits information indicating the evaluation result and / or prediction result by the second analysis unit 612 to the terminal TM in real time, thereby causing the terminal TM to output the evaluation result and / or prediction result as an image or sound (step S7 in FIG. 2). As a result, according to the first modification, the user of the terminal TM can understand, use, and analyze the evaluation result and / or prediction result in real time.

[0151] The support unit 620 generates support information for supporting the user in real time based on the result of the frequency analysis by the first analysis unit 611 or the evaluation or prediction result by the second analysis unit 612. The support information is, for example, information that supports the user in maintaining or recovering their health. The support information may include, for example, warning information.

[0152] The information providing server 7 transmits the support information generated by the support unit 620 to the terminal TM in real time, thereby causing the terminal TM to output the support information as an image or sound (step S7 in FIG. 2 ). As a result, according to the first modification, the user of the terminal TM can immediately take action based on the support information (e.g., warning information).

[0153] The observation period T2 may be a period that precedes the specific period T1 on the time axis.

[0154] In the first modification, the specific period T1 is different from the observation period T2. However, the first modification also uses the ratio R, which is a correction parameter calculated in the correction calculation, and therefore the divergence of the cumulative data D3 for the observation period T2 is suppressed.

[0155] (Second Modification) In the first embodiment, the biometric data D1 is classified based on the first classification standard information C1. In contrast, in a second modification of the first embodiment, the biometric data D1 is classified based on predetermined second classification standard information C2. Differences will be mainly described below with reference to Figs. 3 and 15 to 19.

[0156] In the second modified example, for example, the original biometric data D0 is data indicating the number of steps. In this case, for example, preprocessing by the preprocessing unit 410 is not performed. Therefore, the original biometric data D0 is processed as the biometric data D1 to be classified.

[0157] FIG. 15 is a graph showing biometric data D1 according to the second modified example. The vertical axis represents the number of steps. FIG. 16 is a graph showing first biometric data B1 and second biometric data B2 calculated by the classification unit 411 according to the second modified example. The vertical axis represents the number of steps. FIG. 17 is a graph showing corrected data D2 calculated by the correction unit 412 according to the second modified example. The vertical axis represents the number of steps after correction. FIG. 18 is a graph showing cumulative data D3 calculated by the accumulator 413 according to the second modified example. The vertical axis represents the cumulative value of the number of steps after correction. FIG. 19 is a graph showing cumulative data D4 after statistical processing calculated by the statistical processing unit 414 according to the second modified example. The vertical axis represents the statistically processed value for the cumulative value. In the example of FIG. 19, the statistically processed value is a standardized value. In FIGS. 15 to 19, the horizontal axis represents time t (minutes). In addition, in FIGS. 16 and 17, black circles represent the first state, and white diamonds represent the second state. The first state indicates a state in which the living organism is active (active state). The second state indicates a state in which the living organism is inactive (inactive state). Specifically, the first state indicates a state in which the living organism is moving (active state). The second state indicates a state in which the living organism is stationary (inactive state).

[0158] As shown in FIGS. 3 and 15 , the classification unit 411 compares the second classification reference information C2 with the biological data D1 to be classified and classifies the biological data D1 based on the comparison result. The second classification reference information C2 is set to a value of the biological data that substantially indicates a cessation of mental and physical activity. In the example of FIG. 15 , the second classification reference information C2 is set to a value of the number of steps that substantially indicates a cessation of mental and physical activity. In other words, the second classification reference information C2 is set to zero. According to the second modification, the biological data D1 is classified into an active state and an inactive state, and then correction and accumulation are performed. Therefore, the cumulative data D3 reflects the comparison result and balance between the active state and the inactive state. As a result, the cumulative data D3 is more suitable for evaluating the biological mechanism that repeats activity and inactivity.

[0159] 15 and 16, the classification unit 411 classifies, among the plurality of biometric data D1, biometric data D1 that is greater than the second classification standard information C2 into a first status group G1, thereby calculating first biometric data B1 belonging to the first status group G1. Furthermore, the classification unit 411 classifies, among the plurality of biometric data D1, biometric data D1 that is equal to the second classification standard information C2 into a second status group G2, thereby calculating second biometric data B2 belonging to the second status group G2. In this case, the classification unit 411 sets the second biometric data B2 to a predetermined value. In the example of FIG. 16, the predetermined value is "-1."

[0160] The predetermined value is determined experimentally and / or empirically. The classification unit 411 sets the sign of the predetermined value to be opposite to the sign of the first biometric data B1. In the example of FIG. 16, the predetermined value has a negative sign. In the example of FIG. 16, the magnitude of the predetermined value is "1". Note that the correction unit 412 performs a correction calculation such that the ratio R becomes substantially 1, so the magnitude of the predetermined value may be any value.

[0161] 16 and 17 , the correction unit 412 performs a correction operation on the plurality of first biometric data B1 and the plurality of second biometric data B2 to calculate a plurality of time-series corrected data D2. The plurality of corrected data D2 is composed of the plurality of first biometric data B1 and the plurality of corrected second biometric data B2#.

[0162] As shown in FIGS. 17 and 18, the accumulator 413 accumulates the correction data D2 along the time axis to calculate a plurality of pieces of time-series accumulated data D3.

[0163] As shown in FIGS. 18 and 19, the statistical processing unit 414 performs statistical processing on a plurality of pieces of cumulative data D3 to calculate a plurality of pieces of cumulative data D4 after the statistical processing.

[0164] (Third Modification) In the first embodiment, the biometric data D1 is classified according to the first classification standard information C1. In contrast, in a third modification of the first embodiment, the biometric data D1 is classified according to a period. Hereinafter, differences will be mainly described with reference to FIG. 3 .

[0165] The classification unit 411 classifies the biometric data D1 of a first period into a first status group G1, and the biometric data D1 of a second period different from the first period into a second status group G1. The first period indicates a period during which the physical and mental state is a specific state. That is, the first status in the first status group G1 is the specific state. In other words, the first status indicates the physical and mental state during the first period, and the second status indicates the physical and mental state during the second period. The first biometric data B1 and the second biometric data B2 obtained by such classification are corrected, and cumulative data D3 is calculated from the corrected data D2. Therefore, the cumulative data D3 reflects the comparison result and balance between a specific physical and mental state (first status) and another physical and mental state (second status). As a result, according to the third modification, the cumulative data D3 is more suitable for evaluating the mechanisms of a living body.

[0166] For example, the first period may be a certain period before sleep, a period during sleep, a certain period after waking up, a period during exercise, a period during a seizure seen in a specific disease, a period of injury or illness, or a period during clinical trial participation. For example, the second period may be a period other than the first period. Alternatively, for example, the second period may be a period other than the first period during which the physical and mental state is different from the specific state in the first period.

[0167] Specifically, the classification unit 411 calculates first biometric data B1 of the first status group G1 and second biometric data B2 of the second status group G2 by encoding the biometric data D1 depending on whether the biometric data D1 belongs to the first period or the second period. The encoding is a process of setting the biometric data D1 belonging to the first period to a first specific value (first biometric data B1) having a positive or negative sign, and setting the biometric data D1 belonging to the second period to a second specific value (second biometric data B2) having the other positive or negative sign. For example, the first specific value is "+1" and the second specific value is "-1."

[0168] For example, if the first period is a period during sleep and the second period is a period during wakefulness, the classification unit 411 encodes the biometric data D1 of the first period as “+1” and encodes the biometric data D1 of the second period as “−1.” The cumulative data D3 based on such classification can be used to evaluate, for example, the period during which the subject had a sleep debt.

[0169] (Fourth Modification) In the first embodiment, the cumulative sum D3j is expressed by equation (1). In contrast, in a fourth modification of the first embodiment, a limit value (a threshold value Th described later) is set for the cumulative sum D3j. Differences will be mainly described below with reference to FIGS. 3, 20, and 21.

[0170] In the fourth modified example, for example, the original biometric data D0 is data indicating sleep duration. In this case, for example, preprocessing by the preprocessing unit 410 is not performed. Therefore, the original biometric data D0 is processed as the biometric data D1 to be classified. FIG. 20 is a graph showing the biometric data D1 according to the fourth modified example. The vertical axis indicates the daily sleep duration, and the horizontal axis indicates the time t (month). The classification unit 411 performs processing on the biometric data D1 in FIG. 20 to calculate first biometric data B1 and second biometric data B2. In this case, for example, the first state is "sufficient sleep" and the second state is "insufficient sleep." Then, the correction unit 412 performs a correction calculation on the first biometric data B1 and the second biometric data B2 to generate corrected data D2. The accumulator 413 then accumulates the corrected data D2 to calculate multiple cumulative time-series data D3.

[0171] 21 is a graph showing an example of accumulated data D3 according to Modification 4. The vertical axis represents the corrected accumulated value of sleeping time, and the horizontal axis represents time t (months).

[0172] 21 , in the fourth modification, the accumulator 413 compares the cumulative data D3 to be compared with the threshold value Th, and based on the comparison result, sets the threshold value Th to new cumulative data D3 (solid line) instead of the cumulative data D3 to be compared (dashed line). In the example of FIG. 21 , if the cumulative data D3 (dashed line) exceeds the threshold value Th, the accumulator 413 sets the threshold value Th to new cumulative data D3 (solid line) instead of the cumulative data D3 (dashed line). Therefore, according to the fourth modification, by setting the threshold value Th, it is possible to derive cumulative data D3 that more closely resembles the mechanism of a living body.

[0173] For example, in the case where sleep debt occurs due to lack of sleep and sleep time cannot be accumulated, setting a limit on the accumulated data D3 on the assumption that sleep time cannot be accumulated, rather than uniformly accumulating correction data D2, allows for a more appropriate approximation of the biological mechanism. In the example of Figure 21, by setting a threshold value Th as a limit, the characteristics of sleep time that cannot be accumulated are represented by the accumulated data D3. In other words, the sleep debt is represented by the accumulated data D3.

[0174] As an example, the accumulator 413 calculates the cumulative sum D3j as the cumulative data D3 using equations (5) to (7). D3#j indicates the cumulative sum at time t=j. D2i indicates the corrected data D2 at time t=i. ts indicates the start time of the accumulation. The accumulator 413 calculates multiple cumulative sums D3#j in time series while updating j. This is the same as equation (1).

[0175]

[0176] The accumulator 413 determines whether the cumulative sum D3#j exceeds the threshold value Th. If the accumulator 413 determines that the cumulative sum D3#j does not exceed the threshold value Th, it calculates the cumulative sum D3j using equation (6). On the other hand, if the accumulator 413 determines that the cumulative sum D3#j exceeds the threshold value Th, it calculates the cumulative sum D3j using equation (7).

[0177]

[0178]

[0179] In a fifth modification of the first embodiment, the preprocessing unit 410, the classification unit 411, the correction unit 412, the accumulation unit 413, and the statistical processing unit 414 perform processing for each of a plurality of attribute groups having different attributes. Differences from the first embodiment will be mainly described below with reference to FIG. 3 .

[0180] The third analysis unit 613 derives features according to the attributes of the attribute groups by comparing the cumulative data D3 between multiple attribute groups or by comparing information indicating the features of the cumulative data D3 between multiple attribute groups. In this case, the features of the cumulative data D3 are, for example, the results of the frequency analysis by the first analysis unit 611 or the evaluation results or prediction results by the second analysis unit 612.

[0181] The information providing server 7 (FIG. 1) transmits information indicating characteristics according to the attributes of the attribute group to the terminal TM, causing the terminal TM to output the information indicating characteristics according to the attributes of the attribute group as an image or sound (step S7 in FIG. 2). As a result, according to the sixth modification, the user of the terminal TM can grasp, use, and analyze the characteristics of each attribute group. The attributes are, for example, medical history or gender.

[0182] (Sixth Modification) In the first embodiment, the classification unit 411 classified the plurality of biometric data D1 into two status groups. In contrast, in a sixth modification of the first embodiment, the classification unit 411 classifies the plurality of biometric data D1 into three or more status groups. Differences will be mainly described below with reference to FIG. 3 .

[0183] The classification unit 411 classifies the plurality of biometric data D1 into, for example, a first state group G1 indicating a first physical and mental state, a third state group G3 indicating a third physical and mental state, and a second state group G2 indicating a second physical and mental state, thereby calculating first biometric data B1 belonging to the first state group G1, third biometric data B3 belonging to the third state group G3, and second biometric data B2 belonging to the second state group G2. The third state indicates a physical and mental state between the first state and the second state. The third state is, for example, a neutral physical and mental state.

[0184] In the sixth modified example, the first classification standard information C1 includes a first standard value and a second standard value. For example, the first standard value is greater than the second standard value. The classifier 411, for example, subtracts the first standard value from the biometric data D1 and sets the subtraction result, which indicates a positive value, as the first biometric data B1. The classifier 411, for example, subtracts the second standard value from the biometric data D1 and sets the subtraction result, which indicates a negative value, as the second biometric data B2. For example, the classifier 411 sets the biometric data D1 between the first standard value and the second standard value as the third biometric data B3.

[0185] The correction unit 412 calculates corrected data D2 by performing a correction calculation on the first biometric data B1 and the second biometric data B2. The accumulation unit 413 accumulates the corrected data D2 to calculate accumulated data D3. In this way, in the sixth modified example, the third biometric data B3 is excluded from the correction and accumulation targets.

[0186] For example, even when a plurality of biometric data D1 are classified into four or more status groups, the first biometric data B1 and the second biometric data B2 are the objects to be corrected and accumulated.

[0187] (Regarding Various Periods) First, the observation period T2 during which accumulation processing is performed on the biometric data D1 will be described. The observation period T2 can be set to any length, such as one day, one week, one month, one year, or ten years. The observation period T2 includes multiple time intervals TG.

[0188] The observation period T2 may be, for example, a certain period of time going back from the current time point when the biometric data D1 was acquired to a past time point. In this case, the "time point" may be, for example, a "day", but is not particularly limited thereto.

[0189] The specific period T1 for calculating the first classification standard information C1 can be determined in the same manner as the observation period T2.

[0190] Next, the time interval TG constituting the observation period T2 will be described. The time interval TG can be set to any length, such as one second, one minute, one day, one week, one month, or one year. The time interval TG is continuous within the observation period T2.

[0191] For example, if the observation period T2 is one month, the time interval TG is one minute, and the biometric data D1 is acquired in one-minute increments (e.g., heart rate (bpm), number of steps (spm)), accumulation processing is performed on the one-minute biometric data D1 over one month, and cumulative data D3 is calculated sequentially in one-minute increments. This accumulation processing corresponds to the first pattern of accumulation processing in embodiment 2. In this example, the time interval TG is the same as the acquisition interval of the biometric data D1.

[0192] For example, if the observation period T2 is three months, the time interval TG is one day, and the biometric data D1 is acquired in one-minute increments, the preprocessing unit 410 performs a predetermined process on the multiple biometric data D1 for one day, and as a result of the predetermined process, obtains the biometric data for that day (the time interval TG) (hereinafter, biometric data DX). That is, the biometric data DX is obtained on a daily basis. Then, an accumulation process is performed on the daily biometric data DX over three months, and the accumulated data D3 is calculated sequentially on a daily basis. In this case, the predetermined process is, for example, statistical processing. The statistical processing is, for example, processing to calculate a statistical index. The statistical index is, for example, a representative value (e.g., mean value, median, or mode), standard deviation, variance, or mean squared continuous error, but is not particularly limited thereto. In this case, the biometric data DX is a statistical index of the multiple biometric data D1 for the time interval TG. The accumulation process in this example corresponds to the accumulation process of the third pattern in embodiment 2. In this example, the time interval TG is longer than the interval at which the biometric data D1 is acquired, and multiple pieces of biometric data D1 are acquired within the time interval TG.

[0193] For example, a period indicated by a specific condition may be set within the observation period T2. In this case, accumulation processing is performed on the biometric data D1 during the period indicated by the specific condition. The accumulation processing in this example corresponds to the accumulation processing of the second pattern in the second embodiment. In this example, the time interval TG is the same as the acquisition interval of the biometric data D1, and the period indicated by the specific condition includes multiple time intervals TG.

[0194] Furthermore, for example, a period indicated by a specific condition may be set for each time interval TG. In this case, biometric data DX is calculated for each period indicated by the specific condition. Then, during the observation period T2, accumulation processing is performed on the biometric data DX generated for each period indicated by the specific condition. The accumulation processing in this example corresponds to the accumulation processing of the fourth pattern in the second embodiment. Furthermore, in this example, each of the periods indicated by the time interval TG and the specific condition is longer than the acquisition interval of the biometric data D1, and multiple pieces of biometric data D1 are acquired within the period indicated by the specific condition.

[0195] The specific condition may be a sleep-related condition, a wake-related condition, a sun-related time-of-day condition, or an activity-related condition, or a combination of two or more of these conditions. Details of the specific condition will be described in the second embodiment.

[0196] (Embodiment 2) An information processing system according to Embodiment 2 of the present disclosure uses the cumulative data of Embodiment 1 (including Modifications 1 to 6) as input to a learning model. In other words, by using cumulative data reflecting the homeostasis of a living organism as input to the learning model, highly reliable information related to a living organism can be obtained as output from the learning model. Hereinafter, in the description of Embodiment 2, unless otherwise specified, when "Embodiment 1" is mentioned, this includes Modifications 1 to 6.

[0197] Specifically, the information processing system according to the second embodiment uses a learning model to estimate biological-related state information of a subject from cumulative data based on the subject's biological data. The biological-related state information of the subject includes information representing the subject's mental and physical state, or information representing an environmental state that affects the subject's mental and physical state. The learning model is constructed by performing learning using a learning dataset. The learning dataset includes cumulative data based on the biological data of the learning subject and the biological-related state information of the learning subject. The biological-related state information of the learning subject includes information representing the learning subject's mental and physical state, or information representing an environmental state that affects the subject's mental and physical state.

[0198] The "subject" from whom cumulative data is acquired in embodiment 1 corresponds to a "learning subject" when the learning model is trained, and corresponds to a "subject" when the learning model is used, in embodiment 2. In other words, the cumulative data obtained from the training subject and the subject's biological data in embodiment 2 is the cumulative data described in embodiment 1.

[0199] In the first embodiment, the biological data processing system 1 derives cumulative data reflecting the homeostasis of the living body. In the second embodiment, the information processing system uses the cumulative data reflecting the homeostasis of the living body as an input of a learning model, thereby making it possible to obtain highly reliable biological-related state information related to the living body as an output of the learning model.

[0200] An information processing system 1A according to a second embodiment of the present disclosure will be described with reference to FIGS. 22 to 35. FIG. 22 is a block diagram showing an example configuration of the information processing system 1A. As shown in FIG. 22, the information processing system 1A includes a training data creation device 2A, a learning device 3A, and an estimation device 4A. Each of the training data creation device 2A, the learning device 3A, and the estimation device 4A is a computer. The estimation device 4A includes a preprocessing unit 41A, an estimation unit 42A, and a learning model TM1. The estimation device 4A may further include a postprocessing unit 43A.

[0201] The learning data creation device 2A acquires biological state raw data A1 and correct answer information Z1 of a learning subject. The biological state raw data A1 is raw data indicating the biological state of the learning subject. A learning subject is a subject from which raw data for learning is acquired. A learning subject is typically a human. The biological state raw data A1 includes multiple time-series vital raw data and / or behavioral raw data of the learning subject. The vital raw data is raw data indicating vital information of a living organism. The behavioral raw data is raw data indicating behavioral information of a living organism. The correct answer information Z1 indicates biological-related state information.

[0202] In relation to the first embodiment, the biological state raw data A1 is composed of a plurality of original biological data D0 of the time series of the learning subject. The vital raw data of the biological state raw data A1 corresponds to the original biological data D0 indicating the vital information of the learning subject. The behavior raw data of the biological state raw data A1 corresponds to the original biological data D0 indicating the behavior information of the learning subject.

[0203] The training data creation device 2A creates a training data set F1 based on the biological state raw data A1 and the correct answer information Z1. The training data set F1 includes feature information GG1 and a correct answer label BB1 of the training subject. The feature information GG1 includes at least the cumulative biological state information DD1 of the training subject. The feature information GG1 is an explanatory variable. The correct answer label BB1 is a target variable. The correct answer label BB1 is created based on the correct answer information Z1. The correct answer label BB1 includes the biological-related state information M1.

[0204] The biological condition cumulative information DD1 includes cumulative data indicating the homeostasis of the training subject. In relation to the first embodiment, the cumulative data corresponds to, for example, the cumulative data D3 or D4 of the training subject (subject).

[0205] The biological state cumulative information DD1 is created based on the biological state data A10 obtained from the biological state raw data A1. In this case, the biological state raw data A1 may be used as the biological state data A10 as is. For example, the learning data creation device 2A calculates the biological state cumulative information DD1 in the same manner as the biological data processing device 4 of embodiment 1. Note that the learning data creation device 2A may also acquire the biological state cumulative information DD1 of the learning subject from the biological data processing device 4.

[0206] The biological condition data A10 includes a plurality of time-series vital data and / or behavioral data of the learning subject. The vital data is data indicating vital information of the living body. The behavioral data is data indicating behavioral information of the living body.

[0207] In relation to the first embodiment, the biological condition data A10 is composed of a plurality of time-series biological data D1 of the learning subject. The vital data of the biological condition data A10 corresponds to the biological data D1 indicating the vital information of the learning subject. The behavioral data of the biological condition data A10 corresponds to the biological data D1 indicating the behavioral information of the learning subject.

[0208] Hereinafter, vital data and behavioral data may be collectively referred to as biometric data. That is, in the second embodiment, biometric data refers to vital data or behavioral data.

[0209] The learning device 3A executes a machine learning algorithm to perform learning using a learning dataset F1. The learning device 3A generates a learning model TM1 by repeatedly performing learning using a plurality of learning datasets F1. The learning device 3A performs supervised learning. The learning model TM1 outputs output information L2 when input information K2 is input. Typically, the learning model TM1 is a trained model. Furthermore, the learning model TM1 is a computer program.

[0210] On the other hand, the estimation device 4A uses the learning model TM1 to estimate biological-related state information M2 of the subject. The subject is typically a human being.

[0211] Specifically, the preprocessing unit 41A performs preprocessing on the subject's biological state raw data A2 and generates input information K2 as a result of the preprocessing. The biological state raw data A2 is raw data indicating the subject's biological state. The biological state raw data A2 includes multiple time-series vital sign raw data and / or behavioral raw data of the subject.

[0212] In relation to the first embodiment, the biological state raw data A2 is composed of a plurality of original biological data D0 of the subject in time series. The vital raw data of the biological state raw data A2 corresponds to the original biological data D0 indicating the vital information of the subject. The behavior raw data of the biological state raw data A2 corresponds to the original biological data D0 indicating the behavior information of the subject.

[0213] The pre-processing unit 41A includes at least an accumulation processing unit 711. The accumulation processing unit 711 calculates the accumulated biological condition information DD2 of the subject based on the biological condition data A20 obtained from the biological condition raw data A2 of the subject. Therefore, the input information K2 includes at least the accumulated biological condition information DD2 of the subject. Note that the biological condition raw data A2 may be used as the biological condition data A20 as is.

[0214] The biological condition information DD2 includes cumulative data indicating the homeostasis of the subject. In relation to the first embodiment, the cumulative data corresponds to, for example, the cumulative data D3 or D4 of the subject.

[0215] The biological condition data A20 includes multiple pieces of time-series vital data and / or behavior data of the subject.

[0216] In relation to the first embodiment, the biological condition data A20 is composed of a plurality of time-series biological data D1 of the subject. The vital data of the biological condition data A20 corresponds to the biological data D1 indicating the vital information of the subject. The behavior data of the biological condition data A20 corresponds to the biological data D1 indicating the behavior information of the subject.

[0217] Specifically, the accumulation processing unit 711 calculates the biological condition cumulative information DD2 in the same manner as the biological data processing device 4 of embodiment 1. That is, the accumulation processing unit 711 has the same functions as the biological data processing device 4. Note that the pre-processing unit 41A may acquire the biological condition cumulative information DD2 of the subject from the biological data processing device 4. In this case, for example, the information processing system 1A may include the biological data processing device 4.

[0218] The estimation unit 42A inputs input information K2 to the learning model TM1. As a result, the learning model TM1 outputs output information L2. The input information K2 is an explanatory variable. The output information L2 is a target variable. The output information L2 includes biological-related state information M2 of the subject. The estimation unit 42A acquires the output information L2 from the learning model TM1. The post-processing unit 43A may perform post-processing on the output information L2 and generate output information N2, which is the result of the post-processing. The output information N2 includes biological-related state information M3 after post-processing.

[0219] Hereinafter, when there is no need to distinguish between the biological-related state information M2 and M3, the biological-related state information M2 and M3 of the subject may be referred to as "biological-related state information MX."

[0220] In the second embodiment, the estimation device 4A can output the biological-related state information MX with high estimation accuracy by inputting at least the cumulative biological state information DD2 of the subject into the learning model TM1. In this case, for example, the estimation device 4A does not require diagnosis and evaluation by a medical professional, testing using a medical device such as a medical image diagnostic device, or collection and testing of bodily fluids such as blood. The medical professional may be, for example, a doctor or nurse. In addition, the estimation device 4A may automatically and continuously acquire the biological state raw data A2 from a wearable device and / or a mobile terminal. Therefore, the biological-related state information MX can be continuously obtained while reducing the burden on the subject.

[0221] Next, the correlation between the cumulative biological state information DD1 of the training subject and the biological-related state information M1 in the training data set F1 will be described. FIG. 23 is a diagram showing an example of the training data set F1. As shown in FIG. 23, the training data set F1 includes feature information GG1 and a correct label BB1 of the training subject. The feature information GG1 includes at least the cumulative biological state information DD1. Meanwhile, the correct label BB1 includes the biological-related state information M1. In the correct label BB1, the biological-related state information M1 is, for example, an actual measurement value (e.g., a diagnosis result by a doctor, a test result using a medical device, a biopsy result using a biomarker or the like, or an evaluation result using various indices).

[0222] Specifically, the biological-related state information M1 includes information M100 representing the mental and physical state of the learning subject (hereinafter referred to as "mental and physical state information M100"), or information M110 representing the state of the environment that affects the mental and physical state of the learning subject (hereinafter referred to as "mental and physical environment information M110"). On the other hand, the biological state cumulative information DD1 is composed of multiple cumulative data of the learning subject. Therefore, the biological state cumulative information DD1 has characteristics similar to the biological mechanism that ensures homeostasis. In other words, the biological state cumulative information DD1 reflects the homeostasis of the living body. The homeostasis of the living body is the property of the living body to maintain a constant mental and physical state in response to changes inside the living body and changes in the external environment of the living body.

[0223] A living organism in good physical and mental condition maintains homeostasis, for example, by appropriately balancing activity and rest, or between exhaustion and recovery, in order to maintain a constant physical and mental state. On the other hand, for example, accumulation of physical and mental stress or fatigue can disrupt homeostasis, leading to poor physical and mental health. In this way, homeostasis is closely related to the physical and mental state. Therefore, it can be inferred that there is a correlation between the cumulative biological condition information DD1, which reflects the homeostasis of the living organism, and the physical and mental condition information M100. In other words, the cumulative biological condition information DD1 is a highly sensitive indicator of the physical and mental state.

[0224] Furthermore, the homeostasis of a living organism can be indirectly affected by the state of the surrounding environment. For example, when the surrounding environment is good, various mental and physical functions are also good, and homeostasis is easily maintained. However, when the surrounding environment is poor, various mental and physical functions may become impaired, causing homeostasis to be disrupted. Conversely, for example, when homeostasis is disrupted, the ability to adapt to environmental changes is reduced, and environmental factors that would not normally be a problem are more likely to have a significant impact on the mind and body. On the other hand, when homeostasis is maintained, for example, the organism is able to adapt to environmental changes and can avoid significant impacts on the mind and body. From these facts, it can be inferred that there is a correlation between the biological condition cumulative information DD1 and the mental and physical environment information M110. In other words, the biological condition cumulative information DD1 is an index that indirectly represents the state of the environment that affects the mental and physical state.

[0225] Therefore, according to the second embodiment, the learning model TM1 constructed by learning using the learning dataset F1 (accumulated biological state information DD1 and biological-related state information M1) can output (generate) the biological-related state information M2 (FIG. 24) of the subject when the accumulated biological state information DD2 (FIG. 24) of the subject is input. That is, the estimation device 4A can estimate the biological-related state information M2 of the subject by using the learning model TM1.

[0226] In particular, the biological state cumulative information DD1 of the learning subject includes cumulative data calculated by cumulative processing of multiple biological data arranged in time series of the learning subject. Furthermore, the biological state cumulative information DD2 of the subject includes cumulative data calculated by cumulative processing of multiple biological data arranged in time series of the subject. The cumulative processing refers to a process of accumulating biological data under the condition that values ​​do not diverge due to accumulation. By performing such accumulation processing, cumulative data reflecting the homeostasis of the living body can be derived. By inputting the biological state cumulative information DD2 including the cumulative data into the learning model TM1, highly reliable biological-related state information M2 can be acquired from the learning model TM1.

[0227] Specifically, the accumulation process classifies the biological data according to the learning subject or the subject's physical and mental state, corrects the classification results to adjust the balance of values, and accumulates the corrected results along the time axis. By performing this series of processes of classification, correction, and accumulation, the biological state cumulative information DD1 and DD2, which include cumulative data, more accurately reflects the homeostasis of the living body. Therefore, by inputting the subject's biological state cumulative information DD2 into the learning model TM1, the estimation device 4A can obtain more reliable biological-related state information M2 from the learning model TM1.

[0228] Preferably, the biological data to be accumulated indicates vital information or behavioral information of the study subject. The vital information is preferably information related to the heart rate of the study subject (e.g., heart rate or heart rate variability index). The behavioral information is preferably information related to the number of steps taken by the study subject, information related to exercise intensity, or sleep information. The correlation between the biological state accumulated information DD1 and the biological-related state information M1 when these are the accumulation targets will be described.

[0229] Vital and behavioral information reflects physical and mental exhaustion and recovery.

[0230] For example, heart rate indicates the degree of physical or mental activity. Therefore, a high heart rate state is thought to contribute to physical and mental exhaustion, while a low heart rate state is thought to contribute to physical and mental recovery. For example, a heart rate variability index indicates the degree of stress in a living organism. Therefore, physical or mental exhaustion and recovery are expressed depending on the level of a heart rate variability index (e.g., a time domain index related to heart rate such as RMSSD).

[0231] For example, the number of steps and exercise intensity indicate the degree of exercise. A high number of steps and a high exercise intensity are thought to contribute to physical exhaustion, while a low number of steps and a low exercise intensity are thought to contribute to physical recovery. Furthermore, for example, from a mental perspective, appropriate exercise can improve mental exhaustion, while too many steps and too high exercise intensity, or too few steps and too low exercise intensity, can cause mental fatigue and depression. Therefore, the number of steps and exercise intensity, which indicate the degree of exercise, are thought to be related to mental exhaustion and recovery.

[0232] For example, high sleep duration and sleep quality are thought to contribute to physical and mental recovery, while low sleep duration and sleep quality are thought to contribute to physical and mental exhaustion. Sleeping for too long can also cause mental strain, such as depression and fatigue. Therefore, sleep duration and sleep quality are related to physical or mental exhaustion and recovery. Sleep quality can be indirectly indicated by behavioral information, such as body movement information, or by vital signs, such as heart rate information.

[0233] As a result of the above, the vital information and behavioral information represent, for example, physical and mental exhaustion and recovery. Accumulation of a first condition, such as exhaustion, may cause a worsening of physical illness and symptoms, a worsening of mental illness and symptoms, or abnormalities in biomarkers. Conversely, accumulation of a second condition, such as recovery, may result in improvement of physical illness and symptoms, an improvement of mental illness and symptoms, or normalization of biomarkers. Meanwhile, the biological condition cumulative information DD1 (cumulative data) obtained by accumulating biological data indicating vital information or behavioral information quantitatively represents the accumulated mental and physical condition (e.g., exhaustion and recovery) over time. Other characteristics of the biological condition cumulative information DD1 (cumulative data) are the same as those in embodiment 1. Therefore, it can be inferred that there is a correlation between the biological condition cumulative information DD1 (cumulative data) and the biological-related condition information M1.

[0234] Furthermore, in the second embodiment, the biological condition cumulative information DD1 may include cumulative data calculated by cumulative processing of statistical indices (e.g., average values ​​or standard deviations) of multiple biological data during a period indicated by specific conditions (a fourth pattern of cumulative processing described later). In this case, in the time interval TG described in the first embodiment, a period indicated by specific conditions (e.g., a period during sleep) is set. For example, if the observation period T2 is one month, the time interval TG is one day, the period indicated by specific conditions is a one-day period during sleep, and biological data is acquired in one-minute increments, statistical processing is performed on the multiple biological data during the sleep period, and statistical indices (e.g., average values ​​or standard deviations) for the period indicated by the specific conditions are obtained as a result of the statistical processing. As a result, statistical indices are obtained for each period indicated by the specific conditions over one month. Then, cumulative processing is performed on these statistical indices over one month, and cumulative data is sequentially calculated for each period indicated by the specific conditions.

[0235] The specific condition may be a sleep-related condition, a wake-related condition, a sun-related time-of-day condition, or an activity-related condition, or a combination of two or more of these conditions.

[0236] The period indicated by the sleep-related conditions is, for example, a period during sleep, a predetermined period before falling asleep (a predetermined period immediately before falling asleep), or a predetermined period after falling asleep (a predetermined period immediately after falling asleep). The period indicated by the wake-related conditions is, for example, a period during wakefulness, a predetermined period before waking up (a predetermined period immediately before waking up), or a predetermined period after waking up (a predetermined period immediately after waking up). The period indicated by the conditions related to the time of day caused by the sun is, for example, a predetermined period during the day, a predetermined period at night, a predetermined period in the morning, a predetermined period during the day, or a predetermined period at night. The period indicated by the activity-related conditions is a time of rest or a time of exercise. A time of rest is, for example, a time when not walking or exercising. The period indicated by the wake-related conditions and the activity-related conditions is, for example, a period of rest during wakefulness, or a period of exercise during wakefulness.

[0237] As an example, "resting" and "exercising" are determined based on the exercise intensity of the living body (e.g., METs) or the number of steps taken by the living body. For example, if METs exceed a predetermined value, it is determined to be exercising, and if METs are equal to or less than a predetermined value, it is determined to be at rest. For example, if the number of steps exceeds a predetermined value, it is determined to be exercising, and if the number of steps is equal to or less than a predetermined value, it is determined to be at rest. In this case, "exercising" refers to the living body moving, and whether or not the living body is exercising is determined by the exercise intensity or the number of steps. Therefore, the degree of movement in "exercising" can be determined arbitrarily depending on the magnitude of the above-mentioned predetermined value. The same applies to "resting".

[0238] In this case, according to the second embodiment, accumulation processing can be performed on statistical indicators for a period (a period indicated by specific conditions) that accurately reflects the physical and mental state and the environment of the living organism. By using the accumulated data obtained by such accumulation processing as learning data, the estimation accuracy of the living organism-related state information M2 by the learning model TM1 can be further improved.

[0239] The effect of the accumulation processing of the fourth pattern will be illustrated. For example, the transition period from wakefulness to sleep (the period before and after falling asleep) is a time when the sympathetic nervous system switches from a dominant state to a parasympathetic nervous system. The transition period from sleep to wakefulness (the period before and after waking up) is a time when the parasympathetic nervous system switches from a dominant state to a dominant state. Furthermore, the transition period is considered to be highly related to circadian rhythms (biological rhythms that occur with sleep). In other words, the transition period can be a time for capturing the circadian rhythm of a living organism. For these reasons, for example, if an abnormality in the living organism prevents a smooth transition, specific abnormalities may occur in the biological data (e.g., vital signs) during the transition period. Furthermore, in diseases or symptoms with diurnal variation, a unique biological state may be observed during the transition period. Therefore, by accumulating statistical indicators of the biological data during the transition period, the accumulated biological state can be extracted with high sensitivity. Therefore, using the accumulated data as learning data is effective in improving the estimation accuracy of the biological-related state information M2 by the learning model TM1.

[0240] For example, calculating the average heart rate immediately after waking up can capture the inflammatory state, which varies throughout the day. Therefore, by performing accumulation processing on the average heart rate immediately after waking up, inflammatory fatigue can be quantified using the accumulated data. As a result, using the accumulated data as learning data is effective for estimating the biological-related condition information M2, which indicates, for example, the worsening of symptoms of an inflammatory disease.

[0241] For example, if the average heart rate before and after falling asleep is high, it is possible that the transition to rest has not been achieved. When a lack of rest is accumulated due to factors such as alcohol or caffeine intake, high body temperature (falling asleep immediately after bathing), and daytime stress, it can lead to feelings of fatigue and reduced daytime concentration. Therefore, by performing accumulation processing on the average heart rate before and after falling asleep, it is possible to indirectly quantify feelings of fatigue and reduced daytime concentration using the accumulated data. As a result, using the accumulated data as learning data is effective for estimating biological state information M2 that indicates states such as fatigue.

[0242] For example, a state in which the parasympathetic nervous system is temporarily not dominant before and after waking up, or a state in which the sympathetic nervous system index is high, is reflected in the heart rate variability index before and after waking up, and the occurrence of diarrhea or failure to defecate is inferred. If such a state accumulates day by day, it may cause constipation (accumulation of stool) or strain on the digestive system due to diarrhea. Therefore, by performing accumulation processing on the heart rate variability index before and after waking up, it is possible to indirectly quantify constipation or strain on the digestive system using the accumulated data. As a result, using the accumulated data as learning data is effective for estimating, for example, biologically related condition information M2 indicating the occurrence of digestive symptoms.

[0243] Next, the effect of the accumulation process of the fourth pattern will be illustrated. For example, during a resting period while awake, if the biological condition is normal, the biological condition is stable. Therefore, biological data (e.g., vital information, behavioral information) during the resting period while awake may be more likely to sensitively reflect abnormal conditions. Furthermore, during the resting period, the state during exercise can be excluded as noise. Therefore, the resting period while awake is a suitable period for extracting abnormal conditions of the biological condition. Furthermore, during an exercise period while awake, for example, the biological condition is under exercise load. Therefore, "changes in the abnormal condition of the biological condition" due to the exercise load are reflected in biological data (e.g., vital information, behavioral information) during the exercise period while awake. Therefore, the exercise period while awake is a suitable period for extracting abnormal conditions of the biological condition.

[0244] For example, in hyperthyroidism such as Graves' disease, hypermetabolism causes a high heart rate at rest, and even light exercise can increase the heart rate. By accumulating the average heart rate at rest while awake and the average heart rate during exercise, it may be possible to indirectly indicate the burden on the heart and lungs, fatigue, and fatigue susceptibility in various diseases, such as heart failure, autonomic imbalance, and depression, using the accumulated data. As a result, using the accumulated data as learning data is effective for estimating biological condition information M2 that indicates, for example, fatigue and fatigue susceptibility.

[0245] In addition, in the second embodiment, the biological condition cumulative information DD1 may include cumulative data calculated by a cumulative process during a period indicated by a specific condition (a second pattern of cumulative process described later). In this case, the cumulative process sets the period indicated by the specific condition (e.g., a period during sleep) as the observation period T2 described in the first embodiment. For example, if the period indicated by the specific condition (observation period T2) is a period during sleep, the time interval TG is one minute, and biological data is acquired in one-minute increments, the cumulative process is performed on the biological data in one-minute increments during the period during sleep, and cumulative data is calculated sequentially in one-minute increments.

[0246] In this case, according to the second embodiment, accumulation processing can be performed on the biological data for a period (a period indicated by specific conditions) that accurately reflects the physical and mental state and the environment of the living organism. By using the accumulated data obtained by such accumulation processing as learning data, the estimation accuracy of the biological-related state information M2 by the learning model TM1 can be further improved.

[0247] The effect of the cumulative processing of the second pattern will be illustrated. For example, by directly cumulatively processing biological data (e.g., vital information) only during the transition period from wakefulness to sleep (the period before and after falling asleep) or only during the transition period from sleep to wakefulness (the period before and after waking up), abnormalities in the switching function of the autonomic nervous system can be more clearly identified. By cumulatively processing biological data only during the transition period, it is possible to detect the accumulation of abnormal patterns specific to the transition period and quantify subtle changes that would be hidden in data from all time periods. Using the cumulative data obtained by such cumulative processing as learning data is effective, for example, for estimating subtle changes in the biological-related state information M2.

[0248] Next, the effect of the accumulation processing of the second pattern will be illustrated. For example, during periods of rest while awake, biological data (e.g., vital signs, behavioral information) directly reflects alterations in biological functions. Therefore, by accumulating biological data only from periods of rest while awake, subtle changes in condition that would be hidden in data from all time periods can be quantified. Furthermore, for example, during periods of exercise while awake, stress on the organism may become apparent. Therefore, by accumulating biological data only from periods of exercise while awake (e.g., vital signs, behavioral information), subtle changes resulting from potential functional abnormalities can be quantified. Using the accumulated data obtained by such accumulation processing as learning data is effective, for example, for estimating subtle changes in the organism-related state information M2.

[0249] For example, sleep quality can be quantified by calculating a heart rate variability index during sleep from biological data during sleep (e.g., vital information). Then, by performing accumulation processing on the heart rate variability index, sleep debt can be quantified using the accumulated data. As a result, by using the accumulated data as learning data, the estimation accuracy of organism-related state information M2 due to sleep debt can be further improved. Furthermore, sleep quality can be quantified by calculating body movement information during sleep from biological data during sleep (e.g., behavioral information). Then, by performing accumulation processing on the body movement information, sleep debt can be quantified using the accumulated data. As a result, by using the accumulated data as learning data, the estimation accuracy of organism-related state information M2 due to sleep debt can be further improved.

[0250] Here, for example, the mental and physical condition information M100 of the organism-related condition information M1 may include information representing the state of the body (hereinafter referred to as "physical condition information"). Because the cumulative biological condition information DD1 is a quantitative index reflecting the homeostasis of the organism, the cumulative biological condition information DD1 reflects changes in the body's state (e.g., worsening or improvement of physical diseases and symptoms) with high sensitivity. Physical diseases and symptoms include, for example, diseases and symptoms of the circulatory system, nervous and cranial nervous system, respiratory system, endocrine and metabolic system, immune and inflammatory system, musculoskeletal system, sensory system, or gynecological system.

[0251] Physical condition information may include, for example, information regarding diseases or symptoms of the circulatory system (hereinafter referred to as "circulatory system information"), information regarding diseases or symptoms of the nervous and cranial nervous system (hereinafter referred to as "nervous system information"), information regarding diseases or symptoms of the respiratory system (hereinafter referred to as "respiratory system information"), information regarding diseases or symptoms of the endocrine and metabolic system (hereinafter referred to as "endocrine metabolic system information"), information regarding diseases or symptoms of the immune and inflammatory system (hereinafter referred to as "immune inflammatory system information"), information regarding diseases or symptoms of the musculoskeletal system (hereinafter referred to as "musculoskeletal system information"), information regarding diseases or symptoms of the sensory system (hereinafter referred to as "sensory system information"), information regarding diseases or symptoms of the gynecological system (hereinafter referred to as "gynecological information"), or information representing a physical condition that can be indicated by the results of a biopsy (hereinafter referred to as "biopsy information").

[0252] The following describes circulatory system information. For example, prolonged exhaustion such as fatigue or lack of sleep disrupts the balance of the autonomic nervous system, causing sympathetic nervous system dominance and resulting in increased blood pressure. Furthermore, for example, prolonged exhaustion such as fatigue or lack of sleep accumulates damage to blood vessels. Increased blood pressure and accumulated damage to blood vessels may cause circulatory system diseases or symptoms. The accumulated biological condition information DD1 reflects with high sensitivity the accumulated physical condition caused by prolonged exhaustion and lack of sleep, which can cause such increases in blood pressure. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and circulatory system diseases or symptoms (circulatory system information).

[0253] Examples of the cardiovascular disease or condition include hypertension, ischemic heart disease, heart failure, arrhythmia, and arteriosclerosis. Examples of the ischemic heart disease include angina pectoris and myocardial infarction. Examples of the arrhythmia include atrial fibrillation.

[0254] The following describes nervous system information. For example, continued lack of sleep can cause vasoconstriction and changes in intracranial pressure, which can lead to headaches. Furthermore, continued exhaustion, such as fatigue, and continued lack of sleep can cause an imbalance in neurotransmitters, worsening chronic neurological disorders. Vasoconstriction, changes in intracranial pressure, and neurotransmitter imbalances can potentially cause diseases or symptoms of the nerves and cranial nervous system. The accumulated biological condition information DD1 reflects with high sensitivity the accumulated physical condition caused by continued exhaustion and lack of sleep, which can cause such diseases as vasoconstriction. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and diseases or symptoms of the nerves and cranial nervous system (nervous system information).

[0255] Examples of diseases or conditions of the nervous system and cranial nervous system include stroke, migraine, tension headache, seizure disorders, and neurodegenerative disorders. Examples of stroke include cerebral infarction or cerebral hemorrhage. Examples of seizure disorders include epilepsy. Examples of neurodegenerative disorders include Parkinson's disease or ALS.

[0256] Respiratory system information will now be described. For example, continued fatigue or other exhaustion or lack of sleep can weaken the immune system, making one more susceptible to respiratory infections. Furthermore, continued fatigue or other exhaustion or lack of sleep can worsen asthma or dyspnea. A weakened immune system and worsening dyspnea can potentially cause respiratory diseases or symptoms. The accumulated biological condition information DD1 reflects with high sensitivity the accumulated physical condition caused by continued exhaustion or lack of sleep, which can cause such weakened immunity. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and respiratory diseases or symptoms (respiratory system information).

[0257] The respiratory disease or condition is, for example, bronchial asthma, chronic obstructive pulmonary disease (COPD), sleep apnea syndrome (SAS), or an upper respiratory tract infection, such as a cold or bronchitis.

[0258] The endocrine metabolic system information will now be described. For example, continued exhaustion, such as fatigue, or lack of sleep can lead to increased insulin resistance. Furthermore, continued exhaustion, such as fatigue, or lack of sleep can lead to increased cortisol levels, resulting in fat accumulation and elevated blood sugar levels. Increased insulin resistance and increased cortisol levels can potentially cause endocrine and metabolic diseases or symptoms. The accumulated biological condition information DD1 reflects with high sensitivity the accumulated physical condition caused by continued exhaustion and lack of sleep, which can cause such increases in insulin resistance. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and endocrine and metabolic diseases or symptoms (endocrine metabolic system information).

[0259] Examples of endocrine and metabolic diseases or conditions include type 2 diabetes, metabolic syndrome, obesity, and thyroid dysfunction, such as hypothyroidism.

[0260] The immune-inflammatory system information will now be described. For example, continued exhaustion, such as fatigue, or lack of sleep can lead to an increase in inflammatory cytokines. Furthermore, continued exhaustion, such as fatigue, or lack of sleep can lead to a breakdown in homeostasis of the immune system. An increase in inflammatory cytokines and a breakdown in homeostasis of the immune system can potentially cause diseases or symptoms of the immune and inflammatory systems. The cumulative biological condition information DD1 reflects with high sensitivity the accumulated physical condition caused by continued exhaustion and lack of sleep, which can cause such an increase in inflammatory cytokines. Therefore, it can be inferred that there is a correlation between the cumulative biological condition information DD1 and diseases or symptoms of the immune and inflammatory systems (immune-inflammatory system information).

[0261] The disease or condition of the immune and inflammatory system is, for example, chronic fatigue syndrome (CFS / ME), an autoimmune disease, chronic inflammatory bowel disease (IBD), or an allergic condition.

[0262] The musculoskeletal system information will now be described. For example, continued chronic fatigue and sleep deprivation can cause a decrease in pain threshold. Furthermore, continued declines in sleep duration and sleep quality can impede muscle repair. A decrease in pain threshold and inhibition of muscle repair can potentially lead to musculoskeletal diseases or symptoms. The accumulated biological condition information DD1 reflects with high sensitivity the accumulated physical condition caused by continued chronic fatigue and sleep deprivation, which can cause such a decrease in pain threshold. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and musculoskeletal diseases or symptoms (musculoskeletal system information).

[0263] Examples of musculoskeletal diseases or symptoms include fibromyalgia, chronic lower back pain, stiff shoulders, hypertonia due to muscle fatigue, and arthritis.

[0264] The sensory organ system information will now be described. For example, prolonged awake time or continued lack of sleep can lead to symptoms centered on exhaustion, such as fatigue. Furthermore, prolonged awake time or continued lack of sleep can lead to abnormalities in the sense of balance. Exhaustion, such as fatigue, and abnormalities in the sense of balance can potentially cause diseases or symptoms of the sensory organ system. The accumulated biological condition information DD1 reflects with high sensitivity the accumulated physical condition caused by prolonged awake time or continued lack of sleep, which can cause such exhaustion. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and diseases or symptoms of the sensory organ system (sensory organ system information).

[0265] Examples of the disease or symptom of the sensory system include visual fatigue, headache or nausea due to eye strain, tinnitus, dizziness, or unsteadiness. Visual fatigue is, for example, VDT syndrome.

[0266] The gynecological information will now be described. For example, continued exhaustion such as fatigue, lack of sleep, and accumulated stress can lead to hormonal imbalance. Hormonal imbalance can potentially cause gynecological diseases or symptoms. The accumulated biological condition information DD1 reflects with high sensitivity the accumulated physical condition caused by continued exhaustion, lack of sleep, and stress that cause such hormonal imbalance. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and gynecological diseases or symptoms (gynecological information).

[0267] The gynecological disease or condition is, for example, premenstrual syndrome, dysmenorrhea, irregular menstruation, or amenorrhea.

[0268] The biometric information may include information representing a physical condition that can be indicated by, for example, respiratory and circulatory system test results, endocrine metabolic and body tissue system test results, body fluid test results, vision and ophthalmology system test results, or fatigue and stress test results. Body fluids may be, for example, urine or saliva.

[0269] The test results for the respiratory and circulatory systems include, for example, the results of arteriosclerosis tests, polysomnography, or respiratory function tests. For example, in arteriosclerosis tests, damage to blood vessels due to chronic fatigue or sleep deprivation may be detected as increased arterial stiffness. For example, in polysomnography, the accumulation of chronic fatigue or sleep deprivation may result in the occurrence or frequency of awakenings during the night, REM sleep behavior disorder, and the like. For example, in respiratory function tests (spirometry), the effects of chronic fatigue or sleep deprivation on the respiratory system may manifest as decreased ventilation efficiency. Meanwhile, the accumulated biological condition information DD1 reflects with high sensitivity the accumulated physical condition due to chronic fatigue or persistent sleep deprivation. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and information representing the physical condition that can be indicated by the test results for the respiratory and circulatory systems.

[0270] The test results of the endocrine metabolism and body tissue system are, for example, measurement results of BMI (body mass index), visceral fat area, or abdominal circumference. For example, chronic fatigue, lack of sleep, and the resulting stress may promote fat accumulation, which may be detected as an increase in BMI. For example, the accumulation of chronic fatigue and lack of sleep may promote visceral fat accumulation, which may be observed in CT measurements as an increase in visceral fat area. For example, the impact of the accumulation of chronic fatigue and lack of sleep on body tissue may be manifested as an increase in abdominal circumference. Meanwhile, the accumulated biological condition information DD1 reflects with high sensitivity the accumulated physical condition due to the continued chronic fatigue and lack of sleep. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and information representing the physical condition that can be indicated by the test results of the endocrine metabolism and body tissue system.

[0271] The body fluid test results include, for example, measurement results of urinary melatonin, urinary cortisol, salivary amylase, or salivary cortisol. For example, due to exhaustion such as sleep deprivation or fatigue, urinary melatonin secretion may be disrupted, resulting in the detection of abnormal secretion in a urinary melatonin test. For example, in a urinary cortisol test, the cortisol rhythm may remain elevated due to accumulated exhaustion such as sleep deprivation or fatigue, resulting in an increase in urinary cortisol levels. For example, in a saliva test, the effects of stress due to sleep deprivation or accumulated fatigue may manifest as increased activity and activity of salivary amylase or salivary cortisol. Meanwhile, the accumulated biological condition information DD1 reflects with high sensitivity the accumulated physical condition due to persistent sleep deprivation or fatigue. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and information representing the physical condition that can be indicated by the body fluid test results.

[0272] The results of the visual and ophthalmological examinations are, for example, the results of a visual acuity test or an eye movement test. For example, a visual acuity test may detect a decline in visual acuity due to unstable focusing caused by accumulated fatigue or other wear and tear. For example, an eye movement test may detect nystagmus or irregular eye movement due to accumulated sleep deprivation. On the other hand, the accumulated biological condition information DD1 reflects with high sensitivity the accumulated physical condition caused by persistent fatigue or sleep deprivation. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and information representing the physical condition that can be indicated by the results of the visual and ophthalmological examinations.

[0273] The fatigue and stress test results are, for example, test results for oxidative stress, immune function, or cytokines. The oxidative stress test results are, for example, test results for biomarkers such as 8-GHdG. The immune function test results are, for example, test results for biomarkers such as HHV6. The cytokine test results are, for example, test results for biomarkers such as TGF-β.

[0274] For example, continued fatigue and other exhaustion or lack of sleep leads to increased oxidative stress and changes in the immune system. Increased oxidative stress and changes in the immune system affect the above-mentioned biomarkers. Meanwhile, the accumulated biological condition information DD1 reflects with high sensitivity the accumulated physical condition caused by continued fatigue and other exhaustion or lack of sleep. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and information indicating the physical condition that can be indicated by the fatigue and stress test results.

[0275] The physical condition information may include, for example, information on accompanying symptoms of a physical disease or condition, or information on side effects or complications of treatment for a physical disease or condition.

[0276] Here, for example, the mental and physical state information M100 of the biological-related state information M1 may include information representing a mental state (hereinafter referred to as "mental state information"). Because the cumulative biological state information DD1 is a quantitative index reflecting the homeostasis of a living organism, the cumulative biological state information DD1 reflects the mental state (e.g., worsening or improvement of mental disorders and symptoms) with high sensitivity.

[0277] The mental state information may include, for example, information representing a state related to a mental illness (hereinafter referred to as "mental illness information"), information representing a state related to mental stress (hereinafter referred to as "mental stress information"), or information representing a state related to cognitive function (hereinafter referred to as "cognitive function information").

[0278] In this case, the mental disorder is, for example, a disease or symptom in the mood disorder field, an anxiety disorder group, a stress-related disorder group, a disease or symptom in the psychosomatic disorder field, or other mental-related symptoms.

[0279] Mental illness information related to the mood disorder domain will now be described. For example, prolonged sleep deprivation can cause abnormalities in neurotransmitters such as serotonin and noradrenaline due to stress and poor sleep quality. Furthermore, for example, persistent fatigue and other exhaustion can lead to depressed moods and decreased motivation. Furthermore, for example, sleep rhythm disturbances are associated with the onset of mania in bipolar disorder. Neurotransmitter abnormalities, depressed moods, decreased motivation, and sleep rhythm disturbances can potentially cause diseases or symptoms in the mood disorder domain. The accumulated biological condition information DD1 reflects with high sensitivity the accumulated mental and physical state caused by persistent sleep deprivation and exhaustion, which can cause such neurotransmitter abnormalities. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and diseases or symptoms in the mood disorder domain (mental illness information related to the mood disorder domain).

[0280] Diseases or conditions in the mood disorder domain are, for example, depression, bipolar disorder, dysthymia, or premenstrual dysphoric disorder.

[0281] Mental illness information related to the anxiety disorder group will be described. For example, prolonged sleep deprivation leads to sympathetic overactivity. Furthermore, for example, prolonged fatigue or other exhaustion strengthens avoidance behavior and worsens anxiety symptoms. Sympathetic overactivity and strengthened avoidance behavior may cause diseases or symptoms of the anxiety disorder group. The accumulated biological condition information DD1 reflects with high sensitivity the accumulated mental and physical state caused by prolonged sleep deprivation and exhaustion that cause such sympathetic overactivity. Therefore, it can be inferred that there is a correlation between the biological condition information DD1 and the anxiety disorder group (mental illness information related to the anxiety disorder group).

[0282] Diseases or conditions of the anxiety disorder group are, for example, generalized anxiety disorder, panic disorder, social anxiety disorder, or phobias.

[0283] Mental illness information related to the stress-related disorder group will be described. For example, continued lack of sleep leads to a decrease in stress tolerance. Furthermore, continued lack of sleep and fatigue, for example, leads to a decrease in the inhibitory function of the frontal lobe. A decrease in stress tolerance and a decrease in the inhibitory function of the frontal lobe may cause diseases or symptoms of the stress-related disorder group. The accumulated biological condition information DD1 reflects with high sensitivity the accumulated mental and physical state caused by continued lack of sleep and fatigue, which can cause such a decrease in stress tolerance. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and the stress-related disorder group (mental illness information related to the stress-related disorder group).

[0284] Examples of the disease or symptom of the stress-related disorder group include adjustment disorder, post-traumatic stress disorder (PTSD), or complex PTSD. Examples of symptoms of PTSD or complex PTSD include flashbacks or hyperarousal.

[0285] Mental illness information related to the psychosomatic area will be described. For example, continued exhaustion such as fatigue and lack of sleep can cause an imbalance in the autonomic nervous system, leading to the onset of physical symptoms. Furthermore, continued accumulation of exhaustion such as fatigue can cause body hypersensitivity and hyperfocus on pain (a decrease in pain threshold). Autonomic nervous imbalance, body hypersensitivity, and hyperfocus on pain can potentially cause diseases or symptoms in the psychosomatic area. The accumulated biological condition information DD1 reflects with high sensitivity the accumulated mental and physical state caused by continued exhaustion and lack of sleep, which can cause such autonomic nervous imbalance. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and diseases or symptoms in the psychosomatic area (mental illness information related to the psychosomatic area).

[0286] An example of a disease or symptom in the psychosomatic field is somatic symptom disorder.

[0287] Mental illness information related to "other mental illness-related symptoms" will now be described. For example, continued exhaustion, such as fatigue, and lack of sleep can cause abnormalities in bodily functions controlled by the autonomic nervous system, such as the intestines, stomach, heart rate, and sweating. Furthermore, the occurrence of such bodily illness abnormalities can cause stress, leading to further fatigue, lack of sleep, and poor sleep quality. Abnormalities in bodily functions controlled by the autonomic nervous system and the resulting increased stress can potentially cause various mental illness-related symptoms. The accumulated biological condition information DD1 reflects with high sensitivity the accumulated mental and physical state caused by continued exhaustion and lack of sleep, which can cause such bodily illness abnormalities. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and mental illness-related symptoms (mental illness information related to mental illness-related symptoms).

[0288] Examples of "other mental health-related symptoms" include dysautonomia, tension gastritis, functional dyspepsia, sleep disorders, and psychogenic fever. Examples of sleep disorders include insomnia and hypersomnia.

[0289] The mental stress information will now be described. For example, continued exhaustion such as fatigue or lack of sleep leads to increased psychological fatigue and stress. On the other hand, the accumulated physical and mental state due to continued exhaustion or lack of sleep is reflected with high sensitivity in the accumulated biological condition information DD1. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and psychological fatigue and stress (mental stress information).

[0290] The mental stress information may include, for example, information representing a mental state that can be indicated by test results related to mental stress. Test results related to mental stress may be test results based on a psychological stress scale, a fatigue scale, or a lethargy scale. These scales can be used to measure increases in psychological fatigue or stress as numerical values.

[0291] The cognitive function information will now be described. For example, continued exhaustion such as fatigue or lack of sleep leads to a decline in cognitive function, leading to a decline in attention, concentration, and cognitive task efficiency. On the other hand, the accumulated biological condition information DD1 reflects with high sensitivity the accumulated mental and physical state due to continued exhaustion and lack of sleep. Therefore, it can be inferred that there is a correlation between the accumulated biological condition information DD1 and cognitive function (cognitive function information).

[0292] The cognitive function information may include, for example, information representing a mental state that can be indicated by test results related to cognitive function. Test results related to cognitive function include, for example, test results such as PVT (Psychomotor Vigilance Test), Stroop test, Trail Making Test, Digit Span, Working Memory Span, WISC, or WAIS. Cognitive function can be quantitatively measured by these tests.

[0293] The mental state information may include, for example, information about accompanying symptoms of a mental disorder or condition, or information about side effects or complications of treatment for a mental disorder or condition.

[0294] Here, for example, the mental and physical condition information M100 of the organism-related condition information M1 may include information on the quality of life (QOL) of the study subject (hereinafter referred to as "QOL information"). The QOL information indicates, for example, the quality of life related to the physical condition or mental condition.

[0295] For example, continued exhaustion such as fatigue or lack of sleep directly reduces quality of life. Furthermore, the occurrence, worsening, or persistence of the various physical illnesses or symptoms described above, or the various mental illnesses or symptoms described above, indirectly reduces quality of life. The biological condition cumulative information DD1 reflects with high sensitivity the accumulated physical or mental state due to the persistence of exhaustion or illness, which causes a reduction in quality of life. Therefore, it can be inferred that there is a correlation between the biological condition cumulative information DD1 and QOL information.

[0296] Here, for example, the mental and physical environment information M110 of the biological-related state information M1 may include information regarding an abnormality in the environment in which the learning subject is present (hereinafter referred to as "environmental abnormality information"). The environmental abnormality information is, for example, information regarding an abnormality in the components of the gas that constitutes the environment, or an abnormality in an environmental factor (e.g., temperature, humidity, etc.). The mental and physical environment information M110 is expressed, for example, by information indicating whether an abnormality in the environmental state is "present" or "absent," information indicating the degree of the environmental state in stages, or information indicating the environmental state by a score. Note that the content of the mental and physical state information M100 and the content of the mental and physical environment information M110 are different. In other words, the content of the mental and physical state information M100 and the content of the mental and physical environment information M110 do not overlap.

[0297] Here, the organism-related condition information M1 indicates, for example, the results of a diagnosis or evaluation by a medical professional, the patient himself / herself, or a third party, as information indicated by a score, binary-classified information, or multi-class classified information. Diagnosis and evaluation include diagnosis and evaluation based on the results of tests using medical equipment such as a medical image diagnostic device, as well as diagnosis and evaluation based on the results of biometric tests such as biomarkers. In binary classification, the organism-related condition information M1 is, for example, information indicating the "presence" or "absence" of a disease, symptom, or abnormality. In multi-class classification, the organism-related condition information M1 is, for example, information indicating the degree of a disease, symptom, abnormality, quality of life, or environmental condition in stages.

[0298] For example, in the case of hypertension, in binary classification, the biological-related condition information M1 is information indicating "diagnosis of" hypertension (for example, 1) or information indicating "no diagnosis of" hypertension (for example, 0) as a result of a blood pressure test using a blood pressure monitor. In this case, for example, according to the diagnostic criteria of the American Heart Association (AHA) and the American College of Cardiology (ACC), if "systolic blood pressure ≧ 130 mmHg" or "diastolic blood pressure ≧ 80 mmHg", the patient is "diagnosed" of hypertension, and if neither "systolic blood pressure ≧ 130 mmHg" nor "diastolic blood pressure ≧ 80 mmHg" is met, the patient is "no diagnosis of hypertension".

[0299] For example, in the case of hypertension, in multi-class classification, the biological-related condition information M1 indicates severity. In this case, the severity is represented by, for example, information (e.g., 0) indicating a normal state ("systolic blood pressure < 120 mmHg" and "diastolic blood pressure < 80 mmHg"), information (e.g., 1) indicating a pre-hypertensive state ("systolic blood pressure = 120 to 129 mmHg" and "diastolic blood pressure < 80 mmHg"), information (e.g., 2) indicating hypertension stage 1 ("systolic blood pressure = 130 to 139 mmHg" and "diastolic blood pressure = 80 to 89 mmHg"), or information (e.g., 3) indicating hypertension stage 2 ("systolic blood pressure ≥ 140 mmHg" and "diastolic blood pressure ≥ 90 mmHg").

[0300] Furthermore, the biological-related state information M1 may be, for example, the value of an evaluation index or test index for evaluating or testing a disease, symptom, abnormality, quality of life, or environmental state, information obtained by classifying the value into two values, or information obtained by classifying the value into multiple classes. Examples of the evaluation index or test index include a disease activity index, a biopsy test index, a severity score, a functional assessment scale, a prognosis score, and a symptom assessment scale. Examples of the symptom assessment scale include an assessment scale such as a pain scale (e.g., a VAS (Visual Analogue Scale)).

[0301] For example, in the case of sleep apnea syndrome (SAS), in multi-class classification, the biological condition information M1 indicates severity based on the value of the AIH (apnea-hypopnea index). In this case, the severity is represented by, for example, information indicating normal (AHI<5) (e.g., 0), information indicating mild (AHI=5-14) (e.g., 1), information indicating moderate (AHI=15-29) (e.g., 2), or information indicating severe (AHI≧30) (e.g., 3).

[0302] Furthermore, the biological-related condition information M1 may be, for example, a numerical value indicating the result of a biometric test such as a biomarker, information obtained by classifying the numerical value into two values, or information obtained by classifying the numerical value into multiple classes.

[0303] Furthermore, the biological-related state information M1 may be, for example, a comprehensive result obtained by combining the results of the above examples, and may be represented by information indicated by a score, binary-classified information, or multi-class-classified information.

[0304] As described above, the bio-related state information M1 may directly indicate the physical or environmental state in numerical values ​​(direct numerical output), may indicate the physical or environmental state by binary classification (binary classification output), or may indicate the physical or environmental state by multi-class classification (multi-class classification output).

[0305] Continuing with reference to FIG. 23 , the feature information GG1 will be described. The feature information GG1 may further include one or more of vital information H1 of the learning subject, behavioral information J1 of the learning subject, information R1 about the learning subject's environment (hereinafter referred to as "environmental information R1"), self- or third-party report information V1 about the learning subject's physical and mental state, and attribute information Q1 of the learning subject. This is because these pieces of information are correlated with the biological-related state information M1, which can further improve the estimation accuracy of the biological-related state information M2 by the learning model TM1. For example, the feature information GG1 may include vital information H1 and / or behavioral information J1 in addition to the biological state cumulative information DD1.

[0306] First, the correlation between the vital information H1 and the like and the living body-related condition information M1 will be described.

[0307] For example, the vital information H1 is an objective indicator showing the physiological functional state of the study subject and responds directly to changes in the physical state. Since information representing the physical state is reflected in these measurements, the two are closely related. Therefore, it can be inferred that there is a correlation between the vital information H1 and the physical state information of the organism-related state information M1.

[0308] For example, the vital information H1 is an objective index that reflects the autonomic nervous activity of the learning subject, and fluctuates according to changes in the mental state. The mental state influences the vital information through stress responses and emotional changes, so the two are closely related. Therefore, it can be inferred that there is a correlation between the vital information H1 and the mental state information of the organism-related state information M1.

[0309] For example, vital information H1 is an objective indicator that reflects the state of the biological functions of the study subject and is related to changes in quality of life. A decline in quality of life leads to physical and mental stress, which manifests as abnormalities in vital information, and the two influence each other. Therefore, it can be inferred that there is a correlation between the vital information H1 and the QOL information of the biological-related condition information M1.

[0310] For example, the vital information H1 is an objective indicator of the environmental response of the living body of the study subject, and responds sensitively to changes in the environmental condition. Abnormalities in the gas components that make up the living body's environment or abnormalities in environmental factors affect the living body's mind and body and can be observed as fluctuations in the vital information. Therefore, it can be inferred that there is a correlation between the vital information H1 and the environmental abnormality information in the living body-related condition information M1.

[0311] For example, the behavioral information J1 is an objective indicator that indicates the exercise intensity and exercise pattern of the study subject, and is closely related to the physical condition. Physical illnesses and functional disorders are manifested as a decrease in exercise intensity and the number of steps taken, while appropriate physical activity improves physical function. Therefore, it can be inferred that there is a correlation between the behavioral information J1 and the physical condition information of the organism-related condition information M1.

[0312] For example, the behavioral information J1 is an objective index showing the exercise intensity and exercise pattern of the study subject, and reflects the mental state. Changes in mental state such as anxiety and depression appear as changes in exercise intensity and activity rhythm, and are reflected in the fluctuation pattern of the behavioral information. Therefore, it can be inferred that there is a correlation between the behavioral information J1 and the mental state information of the organism-related state information M1.

[0313] For example, the behavioral information J1 is an objective indicator that indicates the exercise intensity and exercise pattern of the study subject, and is directly related to the quality of life. A decline in quality of life appears as a decrease in exercise intensity and a decrease in the number of steps taken, and is clearly reflected in the behavioral information. Therefore, it can be inferred that there is a correlation between the behavioral information J1 and the QOL information of the organism-related state information M1.

[0314] For example, the behavioral information J1 is an objective index indicating the exercise intensity, exercise pattern, etc. of the study subject, and changes depending on environmental conditions. An abnormality in the gas components that make up the living body's environment or an abnormality in environmental factors limits the exercise intensity or the amount of physical activity, and is detected as a change in the behavioral information. Therefore, it can be inferred that there is a correlation between the behavioral information J1 and the environmental abnormality information of the living body-related state information M1.

[0315] For example, the environmental information R1 is an objective indicator of the environmental conditions in which the training subject lives, and is related to the physical condition. Changes in weather conditions and working environment affect the function of the circulatory system and respiratory system, causing fluctuations in the physical condition. Therefore, it can be inferred that there is a correlation between the environmental information R1 and the physical condition information of the organism-related condition information M1.

[0316] For example, the environmental information R1 is an objective indicator of the living environment of the study subject and affects the mental state. Seasonal changes, weather conditions, and stress factors in the work environment affect mood and emotional state, resulting in changes in the mental state. Therefore, it can be inferred that there is a correlation between the environmental information R1 and the mental state information of the biological-related state information M1.

[0317] For example, the environmental information R1 is an objective indicator of the living environment conditions of the study subject and is closely related to quality of life. The quality of weather conditions and working environment directly affect life satisfaction and activity ability, and are manifested as fluctuations in quality of life. Therefore, it can be inferred that there is a correlation between the environmental information R1 and the QOL information of the organism-related state information M1.

[0318] For example, the environmental information R1 is an objective indicator of the external environmental state of the study subject, and is directly related to abnormalities in the gas components that make up the living body's environment or abnormalities in environmental factors. Therefore, it can be inferred that there is a correlation between the environmental information R1 and the environmental abnormality information of the living body-related state information M1.

[0319] The self-reported information V1 from the training subject itself, that is, the self-reported information V1, captures each aspect of the organism-related condition information M1 from a subjective point of view, and it can be assumed that there is a correlation between the two.

[0320] The report information W1 by a third party captures each aspect of the organism-related condition information M1 from an objective point of view, and it can be assumed that there is a correlation between the two.

[0321] The attribute information Q1 is information indicating the basic characteristics of the training subject and indicates the basic background related to the content of the biological-related state information M1. Therefore, it can be inferred that there is a correlation between the attribute information Q1 and the biological-related state information M1.

[0322] Here, particularly when the feature information GG1 further includes one or more pieces of information selected from vital information H1, behavioral information J1, environmental information R1, report information V1, and attribute information Q1 of the study subject, the feature information GG1 (explanatory variable) and the organism-related state information M1 (objective variable) differ in at least content or timing. This is because the objective variable to be estimated and the explanatory variables used to explain the objective variable for estimation differ at least in content or timing. For example, even if the content of the feature information GG1 and the organism-related state information M1 is the same, if the feature information GG1 is past information and the organism-related state information M1 is current or latest information, the feature information GG1 and the organism-related state information M1 differ in timing.

[0323] Next, the contents of the vital information H1 and the like will be described in detail.

[0324] The vital information H1 includes information related to the heartbeat (heartbeat information). The vital information H1 may further include one or more of blood pressure information, respiration information, body temperature information, blood information, and electroencephalogram information. Otherwise, the vital information H1 is similar to the vital information described in the first embodiment.

[0325] The behavior information J1 includes at least one of information related to the number of steps and information related to exercise intensity. The behavior information J1 may further include one or more of information related to sleep, exercise duration, energy consumption, body movement, and body physical quantity. Otherwise, the behavior information J1 is similar to the behavior information described in the first embodiment.

[0326] The behavioral information J1 may include drinking information of the study subject. The drinking information is information related to drinking by a living body. The drinking information is, for example, information indicating whether or not the subject has drunk alcohol, information indicating the amount of alcohol consumed, or information indicating the drinking history. The behavioral information J1 may also include, for example, smoking information. The smoking information is, for example, information indicating whether or not the subject smokes or the smoking history.

[0327] The vital information H1 may include vital information for a period indicated by a specific condition. The behavioral information J1 may include behavioral information for a period indicated by a specific condition. The specific condition is a condition related to sleep, a condition related to wakefulness, a condition related to a time of day due to the sun, or a condition related to activity, or a combination of two or more of these conditions. The specific condition is the same as the specific condition described in relation to the biological condition cumulative information DD1.

[0328] The environmental information R1 is information indicating the environment in which a living organism lives or works. The environmental information R1 includes, for example, one or more of seasonal information, weather information, and employment information. The seasonal information may be indicated by the name of the season, such as winter, or by the month, such as January. The seasonal information includes, for example, information on the amount of ultraviolet light exposure calculated from the season. The weather information includes, for example, information on weather or atmospheric pressure. The employment information includes, for example, one or more of work history, working conditions, work shifts, and the amount of organic solvent exposure due to the occupation.

[0329] The self-report information V1 includes a subjective report of the mental and physical state by the study subject himself / herself. The self-report information V1 includes, for example, a self-report by the study subject regarding the physical state, mental state, or quality of life. For example, the self-report information V1 may be in text format, may be the result of a binary or multi-class classification of a subjective assessment, or may be the result of a rating scale, questionnaire, or survey, etc., expressed numerically. The self-report regarding the physical state includes, but is not limited to, subjective reports regarding, for example, respiratory symptoms such as shortness of breath, dizziness or unsteadiness, fatigue, sleep (quantity and quality), amount of exercise, or the presence or severity of various physical symptoms. The self-report regarding the mental state includes, but is not limited to, subjective reports regarding, for example, anxiety or psychological stress.

[0330] The third-party report information V1 includes an objective report by a third party indicating the physical and mental state of the study subject. The third party may be, for example, a medical professional, a close relative, or a roommate. The third-party report information V1 includes, for example, an objective report by a third party regarding the study subject's physical state, mental state, or quality of life. For example, the third-party report information V1 may be in text format, may be the result of binary or multi-class classification of an objective assessment, or may be the result of a rating scale, questionnaire, or survey, etc., expressed numerically. Objective reports regarding physical states include, but are not limited to, objective reports regarding respiratory symptoms such as shortness of breath, dizziness, unsteadiness, fatigue, sleep (quantity and quality), amount of exercise, or the presence or severity of various physical symptoms. Objective reports regarding mental states include, but are not limited to, objective reports regarding anxiety or psychological stress.

[0331] The self- or third-party report information V1 may be, for example, current or latest information, or may be past information (information reported in the past).

[0332] The attribute information Q1 of the study subject is information indicating the attributes of the study subject. The attribute information Q1 includes, for example, one or more of basic information, anthropometric information, medical history information, diagnostic history information, drug use history information, lifestyle information, exercise-related information, genetic information, and disability-related information. The basic information is, for example, information on age, sex, race, and / or place of residence. The anthropometric information is, for example, information on height, weight, and / or BMI. The medical history information is, for example, information on medical history, surgical history, and / or current illness history. The diagnostic history information is, for example, the results of a health check or information on diagnostic images obtained by a medical image diagnostic device. The drug use history information is, for example, current medication status and / or past medication information. The exercise-related information is, for example, information on whether or not the study subject has an exercise habit, the type of exercise, and / or the frequency of exercise. The disability-related information is, for example, information on whether or not the study subject has a disability, the type of disability, and / or the severity of the disability.

[0333] Next, referring to FIG. 24, the input information K2 input to the learning model TM1 and the output information L2 output from the learning model TM1 will be described. FIG. 24 is a diagram showing an example of the input information K2 and the output information L2. As shown in FIG. 24, the input information K2 includes at least cumulative biological condition information DD2 of the subject. The cumulative biological condition information DD2 includes cumulative data calculated by cumulative processing of multiple biological data arranged in chronological order for the subject. Furthermore, the biological data that is the subject of the cumulative processing indicates vital information or behavioral information of the subject. Otherwise, the cumulative biological condition information DD2 is the same as the cumulative biological condition information DD1 of the learning subject.

[0334] Furthermore, the biological condition cumulative information DD2 may include cumulative data calculated by cumulative processing of statistical indicators of multiple biological data for a period indicated by specific conditions. The biological condition cumulative information DD2 may include cumulative data calculated by cumulative processing for a period indicated by specific conditions. In addition, the period and cumulative data indicated by specific conditions are the same as the period and cumulative data indicated by specific conditions described in relation to the biological condition cumulative information DD1.

[0335] The input information K2 may further include one or more of the subject's vital sign information H2, the subject's behavioral information J2, information R2 about the subject's environment (hereinafter referred to as "environmental information R2"), self-report information V2 about the subject's physical and mental state, and the subject's attribute information Q2. For example, the input information K2 may include the vital sign information H2 and / or the behavioral information J2 in addition to the cumulative biological state information DD2. Note that when the input information K2 includes the vital sign information H2, the feature amount information GG1 ( FIG. 23 ) includes the vital sign information H1; when the input information K2 includes the behavioral information J2, the feature amount information GG1 includes the behavioral information J1; when the input information K2 includes the environmental information R2, the feature amount information GG1 includes the environmental information R1; when the input information K2 includes the self-reported information V2, the feature amount information GG1 includes the report information V1; and when the input information K2 includes the attribute information Q2, the feature amount information GG1 includes the attribute information Q1.

[0336] The vital information H2 includes information about the subject's heartbeat (heartbeat information). The vital information H2 may further include one or more of the subject's blood pressure information, respiration information, body temperature information, blood information, and electroencephalogram information. Otherwise, the vital information H2 is similar to the vital information H1 in FIG. 23 .

[0337] The behavior information J2 includes at least one of information regarding the subject's number of steps and information regarding exercise intensity. The behavior information J2 may further include one or more of information regarding the subject's sleep, exercise duration, energy consumption, body movement, and body physical quantity. Otherwise, the behavior information J2 is similar to the behavior information J1 in FIG. 23 .

[0338] The vital information H2 may include vital information for a period indicated by a specific condition. The behavior information J2 may include behavior information for a period indicated by a specific condition. This is similar to the vital information H1 and the behavior information J1.

[0339] Environmental information R2 is information indicating the environment in which a living organism lives or works. Self-report information V2 includes a subjective report of the subject's physical and mental state. Self-report information V2 may be, for example, current or latest information, or past information (information reported in the past). Attribute information Q2 is information indicating the attributes of the subject. Furthermore, environmental information R2, self-report information V2, and attribute information Q2 in FIG. 24 are similar to environmental information R1, self-report information V1, and attribute information Q1 in FIG. 23, respectively.

[0340] On the other hand, the output information L2 includes the subject's biological-related state information M2. The biological-related state information M2 includes information M200 representing the subject's mental and physical state (hereinafter referred to as "mental and physical state information M200") or information M210 representing the environmental state that affects the subject's mental and physical state (hereinafter referred to as "mental and physical environment information M210"). The mental and physical state information M200 and mental and physical environment information M210 in Fig. 24 are similar to the mental and physical state information M100 and mental and physical environment information M110 in Fig. 23, respectively.

[0341] For example, the mental and physical condition information M200 includes information representing a physical condition (physical condition information), information representing a mental state (mental condition information), or information regarding quality of life (QOL information). QOL information indicates quality of life related to a physical condition or a mental condition. The physical condition information may include, for example, circulatory system information, nervous system information, respiratory system information, endocrine metabolic system information, immune-inflammatory system information, musculoskeletal system information, sensory system information, gynecological system information, or biopsy information. The mental state information may include mental illness information, mental stress information, or cognitive function information. The mental and physical environment information M210 may include environmental abnormality information.

[0342] In the explanation of Figure 23, for example, by replacing the learning subject, feature information GG1, cumulative biological condition information DD1, vital information H1, behavioral information J1, environmental information R1, self-reported information V1, attribute information Q1, correct answer label BB1, biological-related state information M1, mental and physical condition information M100, and mental and physical environment information M110 with the subject, input information K2, cumulative biological condition information DD2, vital information H2, behavioral information J2, environmental information R2, self-reported information V2, attribute information Q2, output information L2, biological-related state information M2, mental and physical condition information M200, and mental and physical environment information M210, respectively, the explanation of the input information K2 and the output information L2 can be substituted.

[0343] Typically, the accumulated biological condition information DD2, the vital information H2, and the behavioral information J2 are information created based on the biological condition raw data A2 acquired from a wearable device (e.g., the biological information detection device 103) and / or a mobile terminal (e.g., the first terminal 102). However, these pieces of information may also be acquired from the wearable device and / or the mobile terminal.

[0344] Next, the utilization stage of the learning model TM1 will be described with reference to Figs. 25 to 31. Fig. 25 is a block diagram showing an example configuration of an estimation system 40A according to the second embodiment. As shown in Fig. 25, the estimation system 40A is connected to a network NW. The estimation system 40A is part of the information processing system 1A in Fig. 1.

[0345] The biological information detection device 103 transmits the biological condition raw data A2 of the subject (plurality of original biological data D0 in time series indicating the biological information of the subject) to the first terminal 102. The biological information detection device 103 may transmit the biological condition raw data A2 to the cloud server 101 or the estimation system 40A via the network NW.

[0346] The first terminal 102 transmits the subject's biological condition raw data A2 to the cloud server 101 via the network NW. The cloud server 101 transmits the biological condition raw data A2 to the estimation system 40A via the network NW. The estimation system 40A processes the biological condition raw data A2. Note that the first terminal 102 may transmit the biological condition raw data A2 to the estimation system 40A via the network NW without providing the cloud server 101. The biological condition raw data A2 may be transmitted directly from the biological information detection device 103 to the estimation system 40A. One or more of the environmental information R2, the self-report information V2, and the attribute information Q2 may be transmitted from the first terminal 102 via the cloud server 101 or directly to the estimation system 40A. A portion of this information may be transmitted from the second terminal 200 via the cloud server 101 or directly to the estimation system 40A.

[0347] More specifically, the estimation system 40A includes a relay server 44A, a first database 45A, an estimation device 4A, a second database 46A, and an information providing server 47A. The relay server 44A, the first database 45A, the estimation device 4A, the second database 46A, and the information providing server 47A are connected to a network NW. The estimation device 4A is, for example, a server. The server is a computer.

[0348] Each of the relay server 44A and the information providing server 47A may include a processing unit, a communication unit, and a storage unit, and may also include an input unit and a display unit. The hardware configurations of the processing unit, communication unit, storage unit, input unit, and display unit are similar to the hardware configurations of the control unit 41, communication unit 44, storage unit 45, input unit 42, and display unit 43 of the biometric data processing device 4 in FIG. 3 . The first database 45A and the second database 46A include at least a storage device such as a hard disk drive. The first database 45A and the second database 46A may have the same hardware configuration as the relay server 44A or the information providing server 47A. The estimation system 40A may not include all or some of the relay server 44A, the first database 45A, the second database 46A, and the information providing server 47A.

[0349] 26 is a flowchart showing an example of an estimation method executed by the estimation system 40A. The estimation method estimates the subject's biological-related state information MX. The estimation method includes steps S51 to S55.

[0350] 25 and 26 , first, in step S51, the relay server 44A receives the subject's biological state raw data A2 (plurality of the subject's time-series original biological data D0) from the cloud server 101. Furthermore, for example, the relay server 44A may receive one or more pieces of information from the subject's environmental information R2, self-report information V2, and attribute information Q2. The relay server 44A is equipped with, for example, an API (Application Programming Interface).

[0351] Next, in step S52, the first database 45A stores the biological state raw data A2 received by the relay server 44A. For example, the first database 45A may store one or more pieces of information selected from the environmental information R2, the self-reported information V2, and the attribute information Q2. Specifically, a record 451 is assigned to each subject in the first database 45A. The record 451 is associated with the subject's identification information (personal identification information). The biological state raw data A2 is recorded in the record 451 of the subject. The record 451 may store one or more pieces of information selected from the environmental information R2, the self-reported information V2, and the attribute information Q2.

[0352] Next, in step S53, the estimation device 4A estimates the subject's biological-related state information MX using the biological state cumulative information DD2 based on the biological state data A20 (plurality of time-series biological data D1) obtained from the biological state raw data A2 stored in the first database 45A and the learning model TM1. Details of the estimation process will be described later. Note that the estimation process may utilize one or more of the environmental information R2, the self-reported information V2, and the attribute information Q2.

[0353] Next, in step S54, the second database 46A stores the subject's biological-related state information MX (output information L2) estimated by the estimation device 4A. Specifically, in the second database 46A, a record 461 is assigned to each subject. The record 461 is associated with the subject's identification information (personal identification information). The biological-related state information MX is recorded in the subject's record 461. The second database 46A also stores input information K2 including accumulated biological state information DD2. Specifically, the input information K2 is recorded in the subject's record 461.

[0354] Next, in step S55, the information providing server 47A transmits the subject's biological-related state information MX stored in the second database 46A to the second terminal 200 and / or the first terminal 102 via the network NW. Specifically, the information providing server 47A displays the subject's biological-related state information MX on the second terminal 200 and / or the first terminal 102. The second terminal 200 is, for example, a terminal of a medical institution, a health management institution, or a research institution. The first terminal 102 is, for example, a terminal of the subject. When step S55 is completed, the estimation method ends.

[0355] In the example of FIG. 25, the biological-related state information MX is transmitted to the first terminal 102 via the cloud server 101, but it may also be transmitted directly to the first terminal 102.

[0356] Fig. 27 is a block diagram showing an example configuration of the estimation device 4A in Fig. 25. As shown in Fig. 27, the estimation device 4A includes a processing unit 400, a communication unit 401, and a storage unit 402. The estimation device 4A may further include an input unit 403 and a display unit 404.

[0357] The input unit 403 is an input device for inputting various information to the processing unit 400. The display unit 404 displays various information. The communication unit 401 is connected to the network NW and communicates with external devices connected to the network NW. The memory unit 402 includes one or more memory devices and stores data and computer programs. The processing unit 400 executes various calculations. The processing unit 400 includes one or more processors. The hardware configurations of the processing unit 400, communication unit 401, memory unit 402, input unit 403, and display unit 404 are similar to the hardware configurations of the control unit 41, communication unit 44, memory unit 45, input unit 42, and display unit 43 of the biometric data processing device 4 in FIG. 3, respectively.

[0358] The memory unit 402 stores a learning model TM1. The learning model TM1 is a trained model. The learning model TM1 is a computer program. The learning model TM1 causes a computer to function so as to estimate biological-related state information M2 of a subject. Specifically, the learning model TM1 causes a computer to function so as to input input information K2 and output output information L2.

[0359] The processing unit 400 includes a pre-processing unit 41A and an estimation unit 42A. The processing unit 400 may further include a post-processing unit 43A. For example, the processor of the processing unit 400 functions as the pre-processing unit 41A, the estimation unit 42A, and the post-processing unit 43A by executing a computer program stored in the storage device of the storage unit 402. The pre-processing unit 41A includes a cumulative processing unit 711. The pre-processing unit 41A may further include a statistical processing unit 712, or the statistical processing unit 712 and a condition processing unit 713. The post-processing unit 43A preferably includes an interpretation unit 431.

[0360] The pre-processing unit 41A acquires biological state raw data A2 of the subject from the first database 45A. The memory unit 402 stores the biological state raw data A2. The biological state raw data A2 includes a plurality of original biological data D0. Furthermore, the pre-processing unit 41A performs pre-processing on the biological state raw data A2 to generate input information K2, which is the result of the pre-processing. The memory unit 402 stores the input information K2. The pre-processing unit 41A may acquire one or more of environmental information R2, self-report information V2, and attribute information Q2 from the first database 45A and include them in the input information K2. The pre-processing unit 41A will be described in detail below.

[0361] 3, the pre-processing unit 41A calculates a plurality of pieces of biometric data D1 from a plurality of pieces of original biometric data D0 constituting the biometric state raw data A2 of the subject. The original biometric data D0 may be used as the biometric data D1 as it is.

[0362] The accumulation processing unit 711 of the pre-processing unit 41A then calculates accumulated data D3 by performing accumulation processing on the multiple biological data D1 arranged in chronological order for the subject. The accumulation processing includes a classification execution process, a correction execution process, and an accumulation execution process. The classification execution process indicates, for example, the process of step S12 in FIG. 10. The classification execution process corresponds to the classification process in embodiment 1. The correction execution process indicates, for example, the process of step S13 in FIG. 10. The correction execution process corresponds to the correction calculation in embodiment 1. The accumulation execution process indicates, for example, the process of step S14 in FIG. 10. In this way, the accumulation processing unit 711 has functions similar to those of the classification unit 411, correction unit 412, and accumulation unit 413 in FIG. 3.

[0363] That is, the classification execution process refers to a process of classifying a plurality of biometric data D1 into at least a first status group G1 indicating a first physical and mental status and a second status group G2 indicating a second physical and mental status, thereby acquiring a plurality of first biometric data B1 belonging to the first status group G1 and a plurality of second biometric data B2 belonging to the second status group G2. The correction execution process refers to a process of calculating a plurality of time-series correction data D2 by adjusting the balance in size between the plurality of first biometric data B1 and the plurality of second biometric data B2. The accumulation execution process refers to a process of calculating cumulative data D3 by accumulating the correction data D2 along the time axis.

[0364] The accumulation process may include, for example, first to fourth patterns. The first accumulation process pattern indicates that the accumulation process is performed on a plurality of pieces of biometric data D1 within the observation period T2 to calculate accumulated data. The second accumulation process pattern indicates that the accumulation process is performed on a plurality of pieces of biometric data D1 within a "period indicated by specific conditions" set within the observation period T2 to calculate accumulated data. The third accumulation process pattern indicates that the accumulation process is performed during the observation period T2 on statistical indices calculated for a plurality of pieces of biometric data D1 within a time section TG in the observation period T2 to calculate accumulated data. The fourth accumulation process pattern indicates that the accumulation process is performed during the observation period T2 on statistical indices calculated for a plurality of pieces of biometric data D1 within a "period indicated by specific conditions" set within the time section TG in the observation period T2 to calculate accumulated data. As described above, the specific conditions are conditions related to sleep, conditions related to wakefulness, conditions related to the time of day due to the sun, conditions related to activity, or a combination of two or more of these conditions.

[0365] The following provides a more detailed example. For example, the subject's original biometric data D0 is raw vital data, and the subject's biometric data D1 is vital data. For example, the subject's original biometric data D0 is raw behavioral data, and the subject's biometric data D1 is behavioral data.

[0366] The vital data may include, for example, heart rate data, and the behavior data may include, for example, at least one of step count data and exercise intensity data.

[0367] The heartbeat data includes, for example, information on the heart rate (bpm). The heartbeat data may include, for example, information on the R-R interval. The heartbeat is measured or aggregated, for example, at a sampling rate of at least once per minute. The heartbeat data is accompanied by the corresponding date, time, minute, and second of measurement and identification information of the subject. The heartbeat data includes information on the heartbeat arranged in chronological order. The heartbeat data may include, for example, information on the R-R interval arranged in chronological order.

[0368] The step count data includes information on the number of steps (spm). The step count is measured or aggregated, for example, at a sampling rate of at least once per minute. The step count data includes the corresponding date, time, minute, and second of measurement and subject identification information. The step count data includes information on the number of steps arranged in chronological order.

[0369] The exercise intensity data includes MET information per fixed time period (1 minute). That is, exercise intensity is typically expressed in METs. The exercise intensity data is measured or aggregated, for example, at a sampling rate of at least once per minute. The exercise intensity data is accompanied by the corresponding date, time, minute, and second of measurement and subject identification information. The exercise intensity data includes MET information arranged in chronological order.

[0370] First, an example of a third pattern of accumulation processing will be described. As an example, the statistical processing unit 712 of the pre-processing unit 41A performs statistical processing on each of the heart rate data, step count data, and exercise intensity data for each time interval TG of the observation period T2. As a result, the statistical processing unit 712 outputs statistical indices, which are the results of the statistical processing, for each of the heart rate data, step count data, and exercise intensity data. Specifically, the statistical processing unit 712 performs statistical processing on multiple pieces of heart rate data included in the time interval TG to calculate statistical indices, performs statistical processing on multiple pieces of step count data included in the time interval TG to calculate statistical indices, and performs statistical processing on multiple pieces of exercise intensity data included in the time interval TG to calculate statistical indices.

[0371] The statistical processing by the statistical processing unit 712 is a mathematical process for quantitatively expressing the characteristics and trends of the data to be processed. Therefore, by performing the statistical processing, the characteristics and trends of the data to be processed (heart rate data, step count data, and exercise intensity data) can be accurately extracted, and the estimation accuracy of the biological-related state information M2 by the learning model TM1 can be further improved.

[0372] Examples of statistical indices include the mean value, sum, standard deviation, variance, mean square error, x% tile value, median, minimum value, maximum value, RMSSD, coefficient of variation CV (= standard deviation / mean value), inverse of the coefficient of variation CV, average change, percentage of changes equal to or greater than a specified value, statistically processed value of the moving median over a time window, and heart rate variability index. RMSSD is calculated not only for heart rate (bpm), but also for steps (SPM) and exercise intensity (METs / min). In this case, RMSSD is the square root of the average of the squares of the differences between consecutive adjacent values. Examples of consecutive adjacent values ​​include consecutive "inverse of heart rate (bpm)," consecutive "steps (SPM)," or consecutive "exercise intensity (METs / min)."

[0373] As described above, the statistical processing unit 712 performs multiple types of statistical processing on the heart rate data and outputs multiple types of statistical indices. For example, the statistical processing unit 712 performs multiple types of statistical processing on the same heart rate data to calculate multiple different statistical indices from the same heart rate data. Therefore, it is possible to extract features of the same heart rate data from different perspectives. This contributes to improving the estimation accuracy of the biological-related state information M2 using the learning model TM1. Furthermore, the statistical processing unit 712 may perform statistical processing on different types of heart rate data (e.g., heart rate, SDNN) to calculate a statistical indices for each type of heart rate data. Similarly, the statistical processing unit 712 may calculate multiple types of statistical indices for each of the step count data and the exercise intensity data.

[0374] Then, as an example, the accumulation processing unit 711 of the pre-processing unit 41A calculates cumulative data by performing accumulation processing on the statistical indices calculated for each time interval TG during the observation period T2. For example, cumulative data for the statistical indices of heart rate data, cumulative data for the statistical indices of step count data, and cumulative data for the statistical indices of exercise intensity data are calculated. Furthermore, the cumulative data is calculated for each different type of statistical indices. Furthermore, for example, cumulative data of statistical indices may be calculated for each different type of heart rate data.

[0375] Note that cumulative data of statistical indicators may be calculated for some of the heart rate data, step count data, and exercise intensity data, or cumulative data may be calculated for one type of statistical indicator.

[0376] Here, for example, a statistical index indicating the degree of dispersion (degree of variation) is an index that quantifies the irregularity of the state of a living organism. The statistical index indicating the degree of dispersion is, for example, standard deviation or variance. An irregular state of a living organism places a burden on the body and mind, and can be a factor that causes disorders in the functioning of the autonomic nervous system and mental state in particular. Therefore, the accumulated physical and mental stresses are reflected in the cumulative data obtained by performing accumulation processing on the statistical index indicating the degree of dispersion. Therefore, by inputting such cumulative data into the learning model TM2, the living organism-related state information M2 can be estimated with high accuracy.

[0377] The statistical processing unit 712 may calculate a statistical index indicating the degree of dispersion for sleep-related data indicating sleep information. The sleep-related data may be, for example, data indicating sleep duration, sleep onset time, wake-up time, or the intermediate time between sleep onset time and wake-up time. For example, the statistical index indicating the degree of dispersion may be calculated using data showing short and long sleep durations over a seven-day period, such as 3 hours, 9 hours, 4 hours, 12 hours, 2 hours, and 14 hours. In this case, the statistical index quantifies the "irregularity of sleep duration over a given week." Such irregularity places a burden on the body and mind, and may particularly cause disruptions to the autonomic nervous system and mental state. Therefore, the cumulative data obtained by performing a cumulative process on the statistical index indicating the degree of dispersion reflects the accumulated physical and mental stress, contributing to improved estimation accuracy of the biological-related state information M2.

[0378] The third pattern of accumulation processing has been described above. In the first pattern of accumulation processing, the accumulation processing unit 711 calculates the accumulated data by performing accumulation processing on a plurality of pieces of heart rate data (e.g., all of the heart rate data) within the observation period T2. In the first pattern, the accumulation processing unit 711 calculates the accumulated data by performing accumulation processing on a plurality of pieces of step count data (e.g., all of the step count data) within the observation period T2. In the first pattern, the accumulation processing unit 711 calculates the accumulated data by performing accumulation processing on a plurality of pieces of exercise intensity data (e.g., all of the exercise intensity data) within the observation period T2. The accumulated data may be calculated for some of the heart rate data, step count data, and exercise intensity data. In the first pattern, the time interval TG is typically the same as the intervals for acquiring the heart rate data, step count data, and exercise intensity data.

[0379] Next, a fourth pattern of accumulation processing will be described. As an example, the condition processing unit 713 of the pre-processing unit 41A performs processing (condition processing) according to specific conditions on each of the heart rate data, step count data, and exercise intensity data for each time interval TG of the observation period T2.

[0380] Specifically, the condition processing unit 713 executes the following processes: extracting heart rate data for a period indicated by a specific condition from a plurality of pieces of heart rate data for the time interval TG; extracting step count data for a period indicated by a specific condition from a plurality of pieces of step count data for the time interval TG; and extracting exercise intensity data for a period indicated by a specific condition from a plurality of pieces of exercise intensity data for the time interval TG. As described above, the specific condition is a sleep-related condition, a wakefulness-related condition, a sun-related condition related to a time of day, or an activity-related condition, or a combination of two or more of these conditions.

[0381] Then, as an example, the statistical processing unit 712 performs statistical processing on each of the extracted heart rate data, extracted step count data, and extracted exercise intensity data. As a result, the statistical processing unit 712 outputs statistical indices, which are the results of the statistical processing, for each of the extracted heart rate data, extracted step count data, and extracted exercise intensity data. That is, the statistical processing unit 712 performs statistical processing on multiple heart rate data included in a period indicated by the specific condition to calculate statistical indices, performs statistical processing on multiple step count data included in the period indicated by the specific condition to calculate statistical indices, and performs statistical processing on multiple exercise intensity data included in the period indicated by the specific condition to calculate statistical indices. Otherwise, the statistical processing performed by the statistical processing unit 712 is similar to the statistical processing performed on the heart rate data, step count data, and exercise intensity data for the time interval TG.

[0382] Then, as an example, the accumulation processing unit 711 calculates cumulative data by performing accumulation processing on the statistical indices calculated for each period indicated by the specific conditions during the observation period T2. As a result, cumulative data for the statistical indices of the heart rate data, cumulative data for the statistical indices of the step count data, and cumulative data for the statistical indices of the exercise intensity data are calculated. Otherwise, the accumulation processing by the accumulation processing unit 711 is similar to the accumulation processing for the statistical indices calculated for each time period TG.

[0383] The fourth pattern of accumulation processing has been described above. In the second pattern of accumulation processing, the accumulation processing unit 711 calculates the accumulated data by performing accumulation processing on a plurality of pieces of heart rate data (e.g., all of the heart rate data) within a "period indicated by specific conditions" set within the observation period T2. In the second pattern, the accumulation processing unit 711 calculates the accumulated data by performing accumulation processing on a plurality of pieces of step count data (e.g., all of the step count data) within the "period indicated by specific conditions" set within the observation period T2. In the second pattern, the accumulation processing unit 711 calculates the accumulated data by performing accumulation processing on a plurality of pieces of exercise intensity data (e.g., all of the exercise intensity data) within the "period indicated by specific conditions" set within the observation period T2. The accumulated data may be calculated for some of the heart rate data, step count data, and exercise intensity data. In the second pattern, the time interval TG is typically the same as the intervals for acquiring heart rate data, step count data, and exercise intensity data.

[0384] The condition processing unit 713 may set two or more different "periods indicated by specific conditions." For example, the condition processing unit 713 extracts heart rate data for each of two different "periods indicated by specific conditions." The two different "periods indicated by specific conditions" may be, for example, daytime and nighttime. The statistical processing unit 712 then calculates statistical indices of the extracted heart rate data for each of the two different "periods indicated by specific conditions." Furthermore, the statistical processing unit 712 calculates the difference between the statistical indices for the two different "periods indicated by specific conditions" (difference processing). The difference can accurately represent, for example, diurnal fluctuations in the biological condition. In the fourth pattern of accumulation processing, when the difference is calculated for each time interval TG, the accumulation processing unit 711 performs accumulation processing on the difference during the observation period T2 to calculate accumulated data. In the second embodiment, the difference in the second pattern is also treated as a type of accumulated data.

[0385] The statistical processing unit 712 may also calculate the proportion of a period satisfying a state condition (a period in an attention state) relative to a period indicated by a specific condition (proportion processing). The state condition may be a condition regarding the magnitude of the step count indicated by the step count data (e.g., 0 steps), a condition regarding the magnitude of the exercise intensity indicated by the exercise intensity data (e.g., METs equal to or greater than a specified value), or a condition regarding information regarding the heart rate indicated by the heart rate data. The proportion can, for example, quantify the duration of the attention state within the period indicated by the specific condition. In other words, the proportion can, for example, capture the activity pattern or rest pattern of a living organism indicated by the attention state. In the fourth pattern of accumulation processing, when the proportion is calculated for each time interval TG, the accumulation processing unit 711 performs accumulation processing on the proportion during the observation period T2 to calculate accumulated data. In the second embodiment, the proportion in the second pattern is also treated as a type of accumulated data.

[0386] Here, the statistical processing unit 712 may calculate a statistical index to be used as the input information K2. In this case, the statistical processing may include, for example, first to fourth patterns. The first pattern of statistical processing indicates that statistical processing is performed on a plurality of pieces of biometric data D1 within the observation period T2 to calculate a statistical index. The second pattern of statistical processing indicates that statistical processing is performed on a plurality of pieces of biometric data D1 within a "period indicated by specific conditions" set within the observation period T2 to calculate a statistical index. The third pattern of statistical processing indicates that statistical processing is performed during the observation period T2 on a statistical index calculated on a plurality of pieces of biometric data D1 within a time section TG within the observation period T2 to calculate a new statistical index. The fourth pattern of statistical processing indicates that statistical processing is performed during the observation period T2 on a statistical index calculated on a plurality of pieces of biometric data D1 within a "period indicated by specific conditions" set within the time section TG within the observation period T2 to calculate a new statistical index.

[0387] First, a first pattern of statistical processing will be described. As an example, the statistical processing unit 712 performs statistical processing on each of the heart rate data, step count data, and exercise intensity data during the observation period T2. Specifically, the statistical processing unit 712 performs statistical processing on multiple pieces of heart rate data (e.g., all heart rate data) included in the observation period T2 to calculate a statistical index, performs statistical processing on multiple pieces of step count data (e.g., all step count data) included in the observation period T2 to calculate a statistical index, and performs statistical processing on multiple pieces of exercise intensity data (e.g., all exercise intensity data) included in the observation period T2 to calculate a statistical index. Note that in the first pattern, the time interval TG is typically the same as the interval between acquisitions of the heart rate data, the step count data, and the exercise intensity data.

[0388] In this case, for example, the statistical processing unit 712 may calculate multiple different statistical indices for the same heart rate data. Furthermore, for example, the statistical processing unit 712 may calculate a statistical indices for each different type of heart rate data. Similarly, the statistical processing unit 712 may calculate multiple types of statistical indices for each of the step count data and the exercise intensity data.

[0389] The first pattern of statistical processing has been described above. In a third pattern of statistical processing, statistical indices are calculated for multiple pieces of heart rate data within each time interval TG, and statistical processing is performed on multiple statistical indices within the observation period T2 to calculate new statistical indices. The same applies to step count data and exercise intensity data. Furthermore, for example, in the third pattern of statistical processing, statistical indices indicating the degree of dispersion may also be calculated for sleep-related data.

[0390] Next, a second pattern of statistical processing will be described. As an example, the condition processing unit 713 performs processing (condition processing) according to specific conditions on each of the heart rate data, step count data, and exercise intensity data during the observation period T2.

[0391] Specifically, the condition processing unit 713 executes a process of extracting heart rate data for a period indicated by a specific condition from multiple heart rate data for the observation period T2, a process of extracting step count data for a period indicated by a specific condition from multiple step count data for the observation period T2, and a process of extracting exercise intensity data for a period indicated by a specific condition from multiple exercise intensity data for the observation period T2.

[0392] Then, as an example, the statistical processing unit 712 performs statistical processing on each of the extracted heart rate data, the extracted step count data, and the extracted exercise intensity data. As a result, the statistical processing unit 712 outputs statistical indices, which are the results of the statistical processing, for each of the heart rate data, step count data, and exercise intensity data during a period of the observation period T2 that is specified by a specific condition. Note that in the second pattern, the time interval TG is typically the same as the interval at which the heart rate data, step count data, and exercise intensity data are acquired.

[0393] The second pattern of statistical processing has been described above. In the fourth pattern of statistical processing, statistical indices are calculated for multiple pieces of heart rate data within a "period indicated by specific conditions" set within each time interval TG, and statistical processing is performed on multiple statistical indices within observation period T2 to calculate new statistical indices. The same applies to step count data and exercise intensity data.

[0394] The statistical processing unit 712 may calculate all or some of the multiple types of statistical indices. Also, the statistical processing unit 712 may perform statistical processing on all or some of the heart rate data, step count data, and exercise intensity data.

[0395] Furthermore, when the statistical processing unit 712 calculates the difference between the statistical indices for two different "periods indicated by specific conditions" (difference processing), the difference may be used as the statistical index (input information K2) for the observation period T2. Furthermore, when the difference is calculated for each time interval TG, a new statistical index (input information K2) may be calculated by statistically processing multiple differences.

[0396] Furthermore, when the statistical processing unit 712 calculates the ratio of the period that satisfies the state condition to the period indicated by the specific condition (ratio processing), the ratio may be used as the statistical index (input information K2) for the observation period T2. Also, when the ratio is calculated for each time interval TG, a new statistical index (input information K2) may be calculated by statistically processing multiple ratios.

[0397] Hereinafter, cumulative data may be referred to as a "cumulative index." The cumulative index is used as input information K2. A statistical index of heart rate data calculated as input information K2 may be referred to as a "heart rate statistical index." A statistical index of step count data calculated as input information K2 may be referred to as a "step count statistical index." A statistical index of exercise intensity data calculated as input information K2 may be referred to as an "exercise intensity statistical index."

[0398] The storage unit 402 stores a cumulative index, a heart rate statistical index, a step count statistical index, and an exercise intensity statistical index as input information K2. The cumulative index is an example of biological condition cumulative information DD2. The heart rate statistical index is an example of vital information H2. The step count statistical index and the exercise intensity statistical index are examples of behavior information J1.

[0399] In the following description, the cumulative index, heart rate statistical index, step count statistical index, exercise intensity statistical index, environmental information R2, self-reported information V2, and attribute information Q2 may be collectively referred to as "features." The number of features is not particularly limited. The "features" may also be referred to as "features FT."

[0400] In the second embodiment, the input information K2 and the feature FT may include at least one cumulative index. In this case, the input information K2 and the feature FT may include two or more different cumulative indexes.

[0401] 27 , the estimation unit 42A estimates the subject's biological-related state information M2 using the learning model TM1. Specifically, the estimation unit 42A inputs input information K2 to the learning model TM1 and obtains output information L2 from the learning model TM1. The storage unit 402 stores the output information L2.

[0402] The learning model TM1 receives input information K2 and outputs output information L2 according to a machine learning algorithm. The machine learning algorithm is not particularly limited, and may be, for example, linear regression, Naive Bayes, Support Vector Machine, neural network, deep neural network (hereinafter referred to as DNN), decision tree, random forest, gradient boosting, or regularized regression. The regularized regression may be, for example, L1 regularized regression or L2 regularized regression. Below, as an example, a case where the machine learning algorithm of the learning model TM1 is DNN will be described.

[0403] FIG. 28 is a diagram schematically illustrating an example of a DNN 50. As shown in FIG. 28, the DNN 50 includes an input layer 51, multiple intermediate layers 52, and an output layer 53. FIG. 28 illustrates an example of a fully connected DNN. The input layer 51 includes at least one node 511. Biological state cumulative information DD2 (cumulative index) is input to at least one node 511. Each of the intermediate layers 52 includes multiple nodes 521. The output layer 53 includes at least one node 531. In the example of FIG. 28, the input layer 51 includes multiple nodes 511, and the output layer 53 includes multiple nodes 531.

[0404] A plurality of feature quantities FT constituting the input information K2 are input to each of the plurality of nodes 511 in the input layer 51. Each node 521 in the intermediate layer 52 converts the output of the previous layer into input to the next layer using trained weights and biases and an activation function. Each node 531 in the output layer 53 outputs a final result (estimated result) based on the output of the previous layer using trained weights and biases and an activation function according to the final output format. In other words, the output layer 53 outputs output information L2. The output information L2 includes a plurality of different types of biological-related state information M2. Note that when the output layer 53 includes one node 531, the output information L2 includes one piece of biological-related state information M2.

[0405] 28, the mental and physical or environmental conditions are directly indicated by numerical values ​​based on the biological-related condition information M2 (direct numerical value output). The number of nodes 531 is the same as the number of biological-related condition information M2 to be output, and may be one, two, or more.

[0406] On the other hand, the post-processing unit 43A in Fig. 27 may output the biological-related condition information M3 by performing post-processing on the biological-related condition information M2. This point will be described with reference to Figs. 29 and 30.

[0407] Fig. 29 is a diagram schematically showing another example of the DNN 50. In the example of Fig. 29, the mental and physical or environmental state is indicated by binary classification using the organism-related state information M3 (binary classification output).

[0408] A first example of binary classification will be described. In the first example, each node 531 directly outputs a numerical value as the biological-related state information M2. In this case, the post-processing unit 43A performs threshold processing on the numerical value indicated by the biological-related state information M2 to perform binary classification. The result of the binary classification is then output as biological-related state information M3.

[0409] A second example of binary classification will be described. In the second example, the biological-related state information M2 output by each node 531 is, for example, a real value between 0 and 1 (hereinafter referred to as a "score value"). For example, if "0" indicates a first state and "1" indicates a second state, the score value indicates a probability value that the mental, physical, or environmental state will be classified as the second state. In this case, the post-processing unit 43A performs probability calibration on the score value to calibrate it to a more reliable probability value. The post-processing unit 43A then performs threshold processing on the score value after the probability calibration (hereinafter referred to as a "calibrated score value"), thereby performing binary classification. The result of the binary classification is then output as biological-related state information M3.

[0410] The number of nodes 531 provided is the same as the number of pieces of biological-related state information M2 to be output, and may be one, two, or more.

[0411] Fig. 30 is a diagram schematically illustrating yet another example of the DNN 50. In the example of Fig. 30, the mental and physical or environmental states are indicated by multi-class classification using the biological-related state information M3 (multi-class classification output). The mental and physical or environmental states are classified into U classes, where U is an integer equal to or greater than 3. The number of classifications (number of classes) U is not particularly limited and can be set arbitrarily.

[0412] As shown in Fig. 30, the DNN 50 includes an input layer 51, multiple intermediate layers 52, and at least one output layer 53 A. In the example of Fig. 30, the DNN 50 includes multiple output layers 53 A.

[0413] The plurality of output layers 53A are provided corresponding to the plurality of different types of biological-related state information M2, respectively. Each output layer 53A outputs output information L2. Specifically, each output layer 53A includes the same number of nodes 531A as the number of classifications (number of classes) U.

[0414] Focus on one output layer 53A. U nodes 531A correspond to U classes, respectively. Each node 531A outputs an output value VL. The output value VL is, for example, a real number between 0 and 1. The sum of the output values ​​VL of the U nodes 531A is "1." Therefore, the output value VL indicates a probability value for classification into the corresponding class. In other words, the class corresponding to the node 531A that outputs the largest output value VL has the highest probability of being correct. However, the post-processing unit 43A performs probability calibration on each output value VL to calibrate each output value VL to a more reliable probability value. The post-processing unit 43A then obtains the largest output value VL among the U output values ​​VL after calibration by probability calibration. The largest output value VL corresponds to the biological-related state information M2. In this way, the U output values ​​VL (output information L2) from the output layer 53A essentially include the biological-related state information M2. Then, the post-processing unit 43A sets the class corresponding to the node 531A that has output the largest output value VL after calibration to the biological-related state information M3. In this manner, the post-processing unit 43A performs multi-class classification.

[0415] The post-processing unit 43A performs multi-class classification for each of the multiple output layers 53A. The number of output layers 53A is the same as the number of pieces of biological-related state information M2 to be output, and may be one or more. When there is one output layer 53A, one piece of biological-related state information M2 and one piece of biological-related state information M3 are output.

[0416] 28 to 30, the input layer 51 only needs to include at least one node 511 for inputting at least one cumulative index.

[0417] Returning to FIG. 27 , the input information K2 input by the estimation unit 42A to the learning model TM1 will be described. Take the cumulative index as an example. The accumulation processing unit 711 performs accumulation processing in units of time corresponding to the observation time unit of the organism-related state information M2 to be estimated. Typically, accumulation processing is performed in units of time equal to the observation time unit. Therefore, the cumulative index in such units of time is input to the learning model TM1. The observation time unit is the time unit used when observing the organism-related state information M2. The observation time unit can be set to any length, such as one minute, one day, one week, one month, or one year. For example, if the organism-related state information M2 is information indicating a change in the state of a disease and the state of the disease is observed on a daily basis, one day is the observation time unit. Therefore, in this case, the time unit for accumulation processing is set to one day. Therefore, the cumulative index obtained on the day the organism-related state information M2 is observed is input to the learning model TM1.

[0418] A more detailed description will be given using heartbeat data as an example. In the accumulation process, the time interval TG is the time unit of the accumulation process. For example, in the first and second accumulation process patterns, if the acquisition interval of heartbeat data is one minute and the observation time unit is one minute, the time interval TG is one minute, and accumulation process is performed in one-minute increments. Therefore, the accumulated data (accumulation index) of the heartbeat data for the time (minutes) during which the biological-related state information M2 is observed is the input information K2. For example, in the third pattern, if the acquisition interval of heartbeat data is one minute and the observation time unit is one day, the time interval TG is one day, and the statistical index of the heartbeat data is calculated in one-day increments, and accumulation process is performed on the statistical index in one-day increments. Therefore, the accumulated data (accumulation index) of the statistical index for the time (days) during which the biological-related state information M2 is observed is the input information K2. For example, in the fourth pattern, if the heart rate data acquisition interval is one minute, the observation time unit is one day, and the period indicated by the specific condition is a one-day sleep period, the time interval TG is one day, the statistical index of the heart rate data is calculated for each sleep period within one day, and accumulation processing is performed on the statistical index on a daily basis. Therefore, the accumulated data (accumulation index) of the statistical index for the time (day) during which the biological-related state information M2 is observed is the input information K2.

[0419] Next, take a heartbeat statistical index as an example of input information K2. The statistical processing unit 712 performs statistical processing in units of time whose length corresponds to the observation time unit of the biological-related state information M2 to be estimated. Typically, the statistical processing is performed in units of time whose length is the same as the observation time unit. In the statistical processing, the time interval TG is the time unit of the statistical processing. Therefore, the heartbeat statistical index in such a time unit is input to the learning model TM1. The same applies to the step count statistical index and the exercise intensity statistical index.

[0420] Continuing with reference to FIG. 27 , the interpretation unit 431 of the post-processing unit 43A calculates contribution information indicating the degree of contribution of multiple feature values ​​FT constituting the input information K2 when the learning model TM1 estimates the biological-related state information M2. The contribution information includes the contribution of each feature value when estimating the biological-related state information M2. Specifically, the interpretation unit 431 calculates the contribution information based on the multiple feature values ​​FT, the biological-related state information M2 that is the estimation result, and information on the learning model TM1. In this case, the interpretation unit 431 calculates the contribution information according to a model interpretation method. The model interpretation method is, for example, Shapley Additive Explanations (SHAP), Individual Conditional Expectation (ICE), Local Interpretable Model-agnostic Explanations (LIME), or Approximate Inverse Model Explanations (AIME). The model interpretation method is not limited to these, as long as it outputs information on indicators that contributed to the estimation result and a breakdown of the contribution. The storage unit 402 stores the contribution information. The second database 46A also stores the contribution information. The information providing server 47A transmits the contribution information stored in the second database 46A together with the biological-related state information MX to the second terminal 200 and / or the first terminal 102 via the network NW. By checking the contribution information, the user can recognize what factors have led to the biological-related state information MX.

[0421] As described above with reference to Figures 22 to 30, according to embodiment 2, the estimation device 4A can obtain highly reliable biological-related state information MX by inputting at least biological state cumulative information DD2 that reflects the homeostasis of the living body into the learning model TM1.

[0422] In particular, in embodiment 2, the input information K2 does not need to include the results of diagnosis and evaluation by a medical professional, the results of tests using medical equipment such as a medical image diagnostic device, or the results of sampling and testing of bodily fluids such as blood. Therefore, in the stage of using the learning model TM1, the estimation device 4A can output the organism-related state information MX with high estimation accuracy without using the results of diagnosis and evaluation by a medical professional, the results of tests using medical equipment such as a medical image diagnostic device, or the results of sampling and testing of bodily fluids such as blood. Therefore, the organism-related state information MX can be obtained while reducing the burden on the subject. In this way, the organism-related state information MX can be obtained without reducing the subject's QOL.

[0423] In addition, the estimation device 4A can continuously acquire the biological state raw data A2 (original biological data D0) from the wearable device (biometric information detection device 103) and / or the mobile terminal (first terminal 102). Therefore, the calculation and monitoring of the biological-related state information MX can be continuously performed. In particular, the biological state raw data A2 is automatically transmitted from the wearable device (biometric information detection device 103) and / or the mobile terminal (first terminal 102) to the estimation system 40A (estimation device 4A). As a result, the burden on the subject is further reduced. In other words, the biological-related state information MX can be obtained while reducing the burden on the subject.

[0424] Next, the estimation process of the biological-related state information M2 executed in step S53 of Fig. 26 will be described with reference to Fig. 27 and Fig. 31. Fig. 31 is a flowchart showing an example of the estimation process. Fig. 31 shows an example in which the accumulation process of the fourth pattern and the statistical process of the first pattern are executed. The estimation process is executed by the estimation device 4A of Fig. 27. As shown in Fig. 31, the estimation process includes steps S71 to S78.

[0425] First, in step S71, the pre-processing unit 41A acquires biological condition raw data A2 (original biological data D0) of the subject from the first database 45A (FIG. 25). The biological condition raw data A2 is stored in the storage unit 402. Furthermore, the pre-processing unit 41A generates biological condition data A20 (biological data D1) based on the biological condition raw data A2, in the same manner as the pre-processing unit 410 in FIG. 3. The biological condition data A20 is stored in the storage unit 402.

[0426] Next, in step S72, the condition processing unit 713 executes processing (condition processing) according to specific conditions on the heart rate data, step count data, and exercise intensity data of the subject's biological condition data A20 stored in the storage unit 402. As a result, the heart rate data, step count data, and exercise intensity data for the period specified by the specific conditions are extracted for each time segment TG of the observation period T2. These data are stored in the storage unit 402.

[0427] Next, in step S73, the statistical processing unit 712 executes statistical processing as pre-processing for the accumulation processing and statistical processing for calculating statistical indices to be used as input information K2.

[0428] That is, the statistical processing unit 712 calculates statistical indices for the heart rate data, the step count data, and the exercise intensity data by performing statistical processing on each of the extracted heart rate data, the extracted step count data, and the extracted exercise intensity data stored in the storage unit 402. These statistical indices are calculated for each time interval TG and stored in the storage unit 402.

[0429] The statistical processing unit 712 also calculates a heart rate statistical index, a step count statistical index, and an exercise intensity statistical index by performing a first pattern of statistical processing on each of the heart rate data for the observation period T2, the step count data for the observation period T2, and the exercise intensity data for the observation period T2 stored in the storage unit 402. These statistical indexes are stored in the storage unit 402 as input information K2.

[0430] Next, in step S74, the accumulation processing unit 711 performs accumulation processing of the fourth pattern during the observation period T2 on the statistical index of the heart rate data for each time interval TG stored in the storage unit 402. The accumulation processing unit 711 performs accumulation processing of the fourth pattern during the observation period T2 on the statistical index of the step count data for each time interval TG stored in the storage unit 402. The accumulation processing unit 711 performs accumulation processing of the fourth pattern during the observation period T2 on the statistical index of the exercise intensity data for each time interval TG stored in the storage unit 402. As a result, cumulative indices for the heart rate data, step count data, and exercise intensity data are obtained. These cumulative indices are stored in the storage unit 402 as input information K2. The cumulative indices are biological condition cumulative information DD2.

[0431] Next, in step S75, the estimation unit 42A inputs the input information K2 (heart rate statistical index, step count statistical index, exercise intensity statistical index, and cumulative index) stored in the storage unit 402 to the learning model TM1. In this case, the input information K2 may include one or more of the environmental information R2, the self-reported information V2, and the attribute information Q2. When the input information K2 is input to the learning model TM1, the learning model TM1 outputs the output information L2.

[0432] Next, in step S76, the estimation unit 42A acquires output information L2 from the learning model TM1. The output information L2 includes biological-related state information M2 of the subject. The output information L2 is stored in the storage unit 402.

[0433] Next, in step S77, the estimation unit 42A determines whether or not post-processing is required for the biological-related state information M2.

[0434] If it is determined in step S77 that post-processing is not required (NO), the estimation process is completed and the process returns to the main routine of Fig. 26. Cases in which post-processing is not required include when the biological-related state information M2 is indicated by direct numerical output (e.g., Fig. 28), or when contribution information is not calculated.

[0435] On the other hand, if it is determined in step S77 that post-processing is necessary (YES), the process proceeds to step S78. Post-processing is necessary, for example, when performing binary classification output or multi-class classification output (e.g., FIGS. 29 and 30), or when calculating contribution information.

[0436] Next, in step S78, the post-processing unit 43A performs post-processing on the biological-related state information M2 and outputs biological-related state information M3. The biological-related state information M3 is stored in the memory unit 402 as output information N2. It is also preferable that the interpretation unit 431 calculates contribution information used when the learning model TM1 estimates the biological-related state information M2. The contribution information is stored in the memory unit 402, for example, as output information L2 or output information N2. The estimation process is then completed, and the process returns to the main routine of FIG. 26.

[0437] 31 , according to the second embodiment, the estimation device 4A can obtain output information L2 (biological-related state information M2) with high estimation accuracy from the learning model TM1 by inputting input information K2 to the learning model TM1. Furthermore, the estimation device 4A can obtain output information N2 (biological-related state information M3) by performing post-processing on the output information L2.

[0438] In this case, the input information K2 only needs to include at least the cumulative index of the subject (biological condition cumulative information DD2). Therefore, the input information K2 does not need to include the results of diagnosis and evaluation by a medical professional, the results of tests using medical equipment such as a medical image diagnostic device, or the results of sampling and testing of bodily fluids such as blood. Therefore, according to the second embodiment, by using the learning model TM1, it is possible to estimate the biological-related condition information M2 with high estimation accuracy while improving the subject's QOL.

[0439] The input information K2 may also include two or more different types of cumulative indices (two or more different types of cumulative data). In this case, the biological-related state information M2 can be estimated with higher accuracy. The input information K2 may also include two or more different types of heart rate statistical indices.

[0440] In step S72, processing according to a specific condition (condition processing) may be performed on at least one of the heart rate data, step count data, and exercise intensity data. In step S73, statistical processing may be performed on at least one of the heart rate data, step count data, and exercise intensity data. In step S74, accumulation processing may be performed on at least one statistical index of the heart rate data, step count data, and exercise intensity data. In step S73, at least one of the heart rate statistical index, step count statistical index, and exercise intensity statistical index may be calculated.

[0441] Here, for example, the accumulation processing unit 711 may execute accumulation processing of the first pattern, the second pattern, or the third pattern in step S74. When executing accumulation processing of the first pattern and the third pattern, step S72 may not be provided.

[0442] Furthermore, for example, in step S73, the statistical processing unit 712 may execute statistical processing of the second, third, or fourth pattern. When executing statistical processing of the first and third patterns, step S72 does not need to be provided.

[0443] In particular, in the second embodiment, it is sufficient to execute the accumulation process for at least one of the first to fourth patterns. Furthermore, the heart rate statistical index, the step count statistical index, and the exercise intensity statistical index may not be calculated, and these indexes may not be included in the input information K2.

[0444] Next, the generation stage of the learning model TM1 will be described with reference to Figures 22, 32, and 33. Figure 32 is a block diagram showing an example configuration of a learning device 3A according to embodiment 2. As shown in Figure 32, the learning device 3A includes a processing unit 31, a communication unit 34, and a storage unit 35. The learning device 3A may also include an input unit 32 and a display unit 33. The hardware configurations of the processing unit 31, the communication unit 34, the storage unit 35, the input unit 32, and the display unit 33 are similar to the hardware configurations of the processing unit 400, the communication unit 401, the storage unit 402, the input unit 403, and the display unit 404 of the estimation device 4A in Figure 27.

[0445] The storage unit 35 stores data and computer programs. The processing unit 31 includes a learning data acquisition unit 310 and a learning unit 311. For example, the processor of the processing unit 31 functions as the learning data acquisition unit 310 and the learning unit 311 by executing a computer program stored in the storage device of the storage unit 35.

[0446] 33 is a flowchart showing an example of a learning method by the learning device 3A. The learning method is an example of a "learning model generation method" of the present disclosure. As shown in FIG. 33, the learning method includes steps S101 to S108.

[0447] First, in step S101, the training data acquisition unit 310 acquires multiple training data sets F1 from the training data creation device 2A ( FIG. 1 ). Specifically, the training data acquisition unit 310 acquires multiple training data sets F1 from the training database DBT ( FIG. 34 ). The storage unit 35 stores the multiple training data sets F1. Some of the multiple training data sets F1 are training data, another part is evaluation data, and still another part is test data.

[0448] Next, in step S102, the learning unit 311 prepares a learning model TM1 before learning. In the learning model TM1 before learning, various parameters are set to initial values.

[0449] Next, in step S103, the learning unit 311 acquires one learning data set F1 from the multiple learning data sets F1 stored in the storage unit 35. In this case, the learning data set F1 is training data.

[0450] Next, in step S104, the learning unit 311 inputs the feature information GG1 included in the learning data set F1 into the learning model TM1 before (or during) learning. As a result, output information is output as an estimation result from the learning model TM1 according to the machine learning algorithm.

[0451] Next, in step S105, the learning unit 311 compares the correct label BB1 included in the learning dataset F1 with the output information output as the estimation result in step S104, and adjusts various parameters based on a machine learning algorithm and a predetermined adjustment method, thereby performing machine learning. As a result, the learning unit 311 causes the learning model TM1 to learn the correlation between the feature information GG1 and the correct label BB1. The predetermined adjustment method for the various parameters is not particularly limited, and may be, for example, the least squares method, maximum likelihood estimation, EM algorithm, gradient descent, backpropagation, or Bayesian estimation.

[0452] Next, in step S106, the learning unit 311 determines whether a learning termination condition is satisfied. The learning termination condition may be, for example, when the evaluation value of a loss function based on the correct label BB1 and the output information output as the estimation result reaches a target value. Alternatively, the learning termination condition may be, for example, when the number of learning iterations (number of epochs) reaches a target number.

[0453] If it is determined in step S106 that the learning termination condition is not satisfied (NO), the process proceeds to step S103. Steps S103 to S105 are repeated until the learning termination condition is satisfied.

[0454] On the other hand, if it is determined in step S106 that the learning termination condition is satisfied (YES), the process proceeds to step S107.

[0455] Next, in step S107, the learning unit 311 adjusts the hyperparameters of the learning model TM1 based on the learning dataset F1 as evaluation data and the input values ​​input by the machine learning engineer via the input unit 32. The adjustment method in this case is not particularly limited, but may be, for example, grid search, random search, or Bayesian optimization.

[0456] Next, in step S108, the learning unit 311 evaluates the estimation accuracy of the learning model TM1 using the learning data set F1 as test data, and the learning method then ends.

[0457] As described above with reference to Figure 33, according to the second embodiment, the learning device 3A performs learning using the learning dataset F1 to generate a learning model TM1 that outputs output information L2 when input information K2 is input. That is, the learning device 3A repeats learning using multiple learning datasets F1 to generate a trained learning model TM1 having various trained parameters. The storage unit 35 stores the trained learning model TM1.

[0458] Next, the generation stage of the training dataset F1 will be described with reference to Fig. 22 , Fig. 34, and Fig. 35. Fig. 34 is a block diagram showing an example configuration of a training data creation device 2A according to embodiment 2. As shown in Fig. 34, the training data creation device 2A includes a processing unit 21, a communication unit 24, and a storage unit 25. The training data creation device 2A may also include an input unit 22 and a display unit 23. The hardware configurations of the processing unit 21, the communication unit 24, the storage unit 25, the input unit 22, and the display unit 23 are similar to the hardware configurations of the processing unit 400, the communication unit 401, the storage unit 402, the input unit 403, and the display unit 404 of the estimation device 4A in Fig. 6.

[0459] The storage unit 25 stores data and computer programs. The storage unit 25 includes a training database DBT. In Fig. 34, the database is abbreviated as DB. The training database DBT stores a plurality of training data sets F1.

[0460] The processing unit 21 includes an accumulation processing unit 211 and a correct label creation unit 216. The processing unit 21 may include one or more of a statistical processing unit 212 and a condition processing unit 213. For example, the processor of the processing unit 400 functions as the accumulation processing unit 211, the correct label creation unit 216, the statistical processing unit 212, and the condition processing unit 213 by executing a computer program stored in the storage device of the storage unit 402.

[0461] The processes of the accumulator 211, the statistical processor 212, and the condition processor 213 are similar to the processes of the accumulator 711, the statistical processor 712, and the condition processor 713 of the estimation device 4A (FIG. 27), respectively.

[0462] For example, in the description of the estimation device 4A, the cumulative processing unit 711 can be replaced with the cumulative processing unit 211, the statistical processing unit 712 with the statistical processing unit 212, the condition processing unit 713 with the condition processing unit 213, the subject with the learning subject, the biological state raw data A2 with the biological state raw data A1, the input information K2 with the feature information GG1, the output information L2 with the correct answer label BB1, the biological state cumulative information DD2 with the biological state cumulative information DD1, the vital information H2 with the vital information H1, the behavioral information J2 with the behavioral information J1, the environmental information R2 with the environmental information R1, the self-reported information V2 with the report information V1, and the biological-related state information M2 with the biological-related state information M1, thereby replacing the description of the cumulative processing unit 211, the statistical processing unit 212, and the condition processing unit 213 of the learning data creation device 2A.

[0463] 35 is a flowchart showing an example of a training data creation method by the training data creation device 2A. Fig. 35 shows an example of executing the accumulation process of the fourth pattern and the statistical process of the first pattern. As shown in Fig. 35, the training data creation method includes steps S201 to S205.

[0464] First, in step S201, the processing unit 21 acquires the biological state raw data A1 (original biological data D0) and the correct answer information Z1 of the learning subject. The biological state raw data A1 and the correct answer information Z1 are stored in the storage unit 25. Furthermore, the processing unit 21 generates biological state data A10 (biological data D1) based on the biological state raw data A1, in the same manner as the pre-processing unit 410 in FIG. 3 . The biological state data A10 is stored in the storage unit 25.

[0465] Next, in step S202, the condition processing unit 213 executes processing (condition processing) according to specific conditions on the heart rate data, step count data, and exercise intensity data of the biological condition data A10 of the learning subject stored in the storage unit 25. As a result, the heart rate data, step count data, and exercise intensity data for the period specified by the specific conditions are extracted for each time interval TG of the observation period T2. These data are stored in the storage unit 402.

[0466] Next, in step S203, the statistical processing unit 210 executes statistical processing as pre-processing of the accumulation processing and statistical processing for calculating statistical indices to be used as the feature amount information GG1.

[0467] That is, the statistical processing unit 210 calculates statistical indices for the heart rate data, the step count data, and the exercise intensity data by performing statistical processing on each of the extracted heart rate data, the extracted step count data, and the extracted exercise intensity data stored in the storage unit 25. These statistical indices are calculated for each time interval TG and stored in the storage unit 25.

[0468] The statistical processing unit 210 also calculates a heart rate statistical index, a step count statistical index, and an exercise intensity statistical index by performing a first pattern of statistical processing on each of the heart rate data for the observation period T2, the step count data for the observation period T2, and the exercise intensity data for the observation period T2 stored in the storage unit 25. These statistical indexes are stored in the storage unit 25 as feature amount information GG1.

[0469] Next, in step S204, the accumulation processing unit 211 performs accumulation processing of the fourth pattern during the observation period T2 on the statistical index of the heart rate data for each time interval TG stored in the storage unit 25. The accumulation processing unit 211 performs accumulation processing of the fourth pattern during the observation period T2 on the statistical index of the step count data for each time interval TG stored in the storage unit 25. The accumulation processing unit 211 performs accumulation processing of the fourth pattern during the observation period T2 on the statistical index of the exercise intensity data for each time interval TG stored in the storage unit 25. As a result, accumulation indices for the heart rate data, step count data, and exercise intensity data are obtained. These accumulation indices are stored in the storage unit 25 as feature amount information GG1. The accumulation indices are biological condition accumulation information DD1.

[0470] Next, in step S205, the correct label creation unit 216 creates a correct label BB1 based on the correct information Z1 and associates it with the feature amount information GG1 (correct label creation process). The correct label BB1 includes the biological-related state information M1. The correct label BB1 is stored in the storage unit 25. As an example, the correct information Z1 is a numerical value that directly indicates the mental, physical, or environmental state. Based on this example, first to third examples will be described.

[0471] As a first example, when the estimation device 4A directly outputs a numerical value (for example, in the case of Figure 28), the correct answer label creation unit 216 associates the correct answer information Z1 with the feature information GG1 as the correct answer label BB1 (bio-related state information M1).

[0472] As a second example, when the estimation device 4A performs binary classification output (e.g., the case of FIG. 29 ), the correct answer label creation unit 216 compares the correct answer information Z1 with a threshold and, based on the comparison result, classifies the correct answer information Z1 (mental, physical, or environmental state) into a first state or a second state. Then, the correct answer label creation unit 216 sets information indicating the first state (e.g., 0) or information indicating the second state (e.g., 1) to the correct answer label BB1 (biological-related state information M1) in accordance with the classification result.

[0473] As a third example, when the estimation device 4A performs multi-class classification (for example, in the case of Figure 30), the correct label creation unit 216 sets an evaluation corresponding to a class for the correct label BB1 (bio-related state information M1) depending on whether the correct information Z1 falls within the numerical range of any of multiple classes.

[0474] In the second and third examples, the correct answer information Z1 after binary classification or multi-class classification may be input via the input unit 22. In this case, the correct answer label creation unit 216 associates the correct answer information Z1 with the feature amount information GG1 as the correct answer label BB1, similarly to the first example.

[0475] When step S205 is completed, the training data creation method ends. Steps S201 to S205 are repeatedly executed to create multiple training data sets F1. The training data creation device 2A stores the training data sets F1 in the training database DBT. The training data creation device 2A may include one or more pieces of information from environmental information R1, report information V1, and attribute information Q1 in the training data set F1.

[0476] As described above with reference to Figure 35, according to embodiment 2, the learning data creation device 2A can create a learning data set F1 suitable for the learning model TM1 that can estimate the biological-related state information M2 while improving the subject's QOL.

[0477] In the second embodiment, the feature amount information GG1 may include at least the cumulative index (biological state cumulative information DD1) of the learning subject.

[0478] The feature information GG1 may also include two or more different types of cumulative indices (two or more different types of cumulative data). In this case, the estimation accuracy of the biological-related state information M2 by the learning model TM1 can be further improved. The feature information GG1 may also include two or more different types of heart rate statistical indices.

[0479] In step S202, processing according to a specific condition (condition processing) may be performed on at least one of the heart rate data, step count data, and exercise intensity data. In step S203, statistical processing may be performed on at least one of the heart rate data, step count data, and exercise intensity data. In step S204, accumulation processing may be performed on at least one statistical index of the heart rate data, step count data, and exercise intensity data. In step S203, at least one of the heart rate statistical index, step count statistical index, and exercise intensity statistical index may be calculated.

[0480] Here, for example, the accumulation processing unit 211 may execute accumulation processing of the first pattern, the second pattern, or the third pattern in step S204. When executing accumulation processing of the first pattern and the third pattern, step S202 may not be provided.

[0481] Furthermore, for example, in step S203, the statistical processing unit 210 may execute statistical processing of the second, third, or fourth pattern. When executing statistical processing of the first and third patterns, step S202 does not need to be provided.

[0482] In particular, in the second embodiment, it is sufficient to execute the accumulation process for at least one of the first to fourth patterns. Furthermore, the heart rate statistical index, the step count statistical index, and the exercise intensity statistical index may not be calculated, and these indexes may not be included in the feature amount information GG1.

[0483] Here, the multiple learning data sets F1 used to train the learning model TM1 are preferably composed of "information representing the physical and mental states of multiple learning subjects belonging to the same group" or "information representing the environmental state that affects the physical and mental states of multiple learning subjects belonging to the same group." A group is a group of learning subjects that satisfy a predetermined condition related to their physical or mental state or their environmental state. The predetermined condition is, for example, a past occurrence of an abnormality in their physical or mental state, such as the onset of a disease or symptom, or a past stay in an environment with an abnormal state, such as staying in an environment with an abnormal atmosphere.

[0484] As described above, it is preferable that the training subjects when acquiring the training dataset F1 belong to the same group. Furthermore, it is preferable that the subject from which the input information K2 to be input to the learning model TM1 is acquired belongs to the same group as the training subjects. In other words, it is preferable that the subject, like the training subjects, also satisfy predetermined conditions related to mental, physical, or environmental conditions. According to this preferable example, for example, it is possible to explain to the subject the validity of estimating the organism-related state information M2 using the learning model TM1, to explain the reliability of the estimation results, and to improve the acceptability of the estimation results. Furthermore, by acquiring the training dataset F1 from training subjects belonging to the same group as the subject, it is possible to improve the estimation accuracy of the organism-related state information M2.

[0485] The above describes the second embodiment of the present disclosure. In the second embodiment, the rheumatoid arthritis index information, blood index information, systemic connective tissue disorder index information, and non-organ-specific systemic autoimmune disease index information may be excluded from the organism-related condition information M1, MX. In other words, the organism-related condition information M1, MX may not include the rheumatoid arthritis index information, blood index information, systemic connective tissue disorder index information, and non-organ-specific systemic autoimmune disease index information. The rheumatoid arthritis index information is an index representing the symptoms or activity of rheumatoid arthritis, or information regarding an index representing the symptoms or activity of rheumatoid arthritis. The blood index information is information that directly or indirectly indicates a blood index. The blood index is an index that quantitatively indicates a blood component, the state of a blood component, a substance in the blood, or the state of a substance in the blood. The systemic connective tissue disorder indicator information is information on an indicator that directly or indirectly indicates the symptoms or activity of a systemic connective tissue disorder, or information on an indicator that directly or indirectly indicates an effect caused by the symptoms or activity of a systemic connective tissue disorder. The non-organ-specific systemic autoimmune disease indicator information is information on an indicator that directly or indirectly indicates the symptoms or activity of a non-organ-specific systemic autoimmune disease, or information on an indicator that directly or indirectly indicates an effect caused by the symptoms or activity of a non-organ-specific systemic autoimmune disease. Systemic connective tissue disorders and non-organ-specific systemic autoimmune diseases include systemic lupus erythematosus.

[0486] Although the preferred embodiments and modifications of the present disclosure have been described in detail above with reference to the accompanying drawings, the technical scope of the present disclosure is not limited to such examples. It is clear that a person skilled in the art of the present disclosure can conceive of various modified or altered examples within the scope of the technical idea described in the claims, and it is understood that these also naturally fall within the technical scope of the present disclosure.

[0487] The devices or systems described herein may be implemented as a single device, or may be implemented by multiple devices (e.g., cloud servers) partially or entirely connected via a network. For example, some or all of the preprocessing unit 410, classification unit 411, correction unit 412, accumulator 413, statistical processing unit 414, analysis unit 610, and support unit 620 may be implemented by the same computer or server. For example, the preprocessing unit 410, classification unit 411, correction unit 412, accumulator 413, statistical processing unit 414, analysis unit 610, and support unit 620 may each be implemented by a separate computer or server. This also applies to the components of the estimation device 4A in FIG. 27 , the learning device 3A in FIG. 32 , and the training data creation device 2A in FIG. 34 . For example, the relay server 2, the biometric data processing device 4, or the biometric data analysis device 6 may have the functionality of the information providing server 7. 25 may have the functions of the information providing server 47A. For example, the first database 3 and the second database 5 may be implemented by a single computer or server. For example, the first database 45A and the second database 46A may be implemented by a single computer or server.

[0488] The series of processes performed by the devices described herein may be implemented using software, hardware, or a combination of software and hardware. A computer program for implementing the functions of the control units 41, 61 and processing units 400, 31, and 21 according to this embodiment may be created and installed on a PC or the like. A computer-readable recording medium storing such a computer program may also be provided. Examples of the recording medium include a magnetic disk, an optical disk, a magneto-optical disk, and a flash memory. The computer program may also be distributed, for example, via a network without using a recording medium. For example, the training database DBT may be located outside the training data creation device 2A. For example, the training model TM1 may be located outside the estimation device 4A.

[0489] Furthermore, the processes described herein using flowchart diagrams do not necessarily have to be performed in the order shown. Some process steps may be performed in parallel. Additional process steps may be employed, and some process steps may be omitted.

[0490] 31 , 33 , and 35 , the processing units 400, 31, and 21 execute the computer programs stored in the storage units 402, 35, and 25 to perform the steps included in the information processing method, estimation processing (estimation method), learning method, and learning data creation method. In other words, the computer programs cause the processing units 400, 31, and 21 to execute the steps included in the information processing method, estimation processing (estimation method), learning method, and learning data creation method. The processing units 400, 31, and 21 correspond to examples of the "computer" in the present disclosure. In other words, the computer program product, when the computer program is executed by the processing units 400, 31, and 21, realizes the steps included in the information processing method, estimation processing (estimation method), learning method, and learning data creation method.

[0491] The estimation device 4A in FIG. 27 may not include, for example, all or part of the statistical processing unit 712 and the condition processing unit 713. The estimation process in FIG. 31 may not include, for example, all or part of steps S72 and S73. Furthermore, the estimation process in FIG. 31 may not include, for example, steps S77 and S78. The training data creation device 2A in FIG. 34 may not include, for example, all or part of the statistical processing unit 212 and the condition processing unit 213. The training data creation method in FIG. 35 may not include, for example, all or part of steps S202 and S203.

[0492] The learning device 3A may generate a new learning model (distilled model) by performing learning using the input information K2 input to the learning model TM1 and the output information L2 or N2 output from the learning model TM1 as a learning data set. The input information K2 and the output information L2, N2 are subject information for the learning model TM1 in the utilization stage. However, in generating a distilled model, the input information K2 and the output information L2, N2 constituting the learning data set correspond to the learning subject information for the learning device 3A.

[0493] Furthermore, the effects described herein are merely descriptive or exemplary and are not limiting. In other words, the technology according to the present disclosure may achieve other effects that will be apparent to those skilled in the art from the description of this specification, in addition to or in place of the above-described effects.

[0494] The following configurations also fall within the technical scope of the present disclosure.

[0495] (Item 1) An estimation device that estimates bio-related state information of a subject, comprising: an estimation unit that inputs input information to a learning model and acquires output information from the learning model; and a storage unit that stores the output information, wherein the output information includes the bio-related state information, and the bio-related state information includes information representing the mental and physical state of the subject, or information representing an environmental state that affects the mental and physical state of the subject, and the input information includes at least cumulative data calculated by cumulative processing of multiple bio-data of the subject arranged in time series, and the cumulative processing indicates a process of accumulating the bio-data under the condition that values ​​do not diverge due to accumulation.

[0496] (Item 2) The estimation device according to Item 1, wherein the accumulation process classifies the biological data according to the physical and mental state of the subject, corrects the classification results to adjust the balance of values, and accumulates the corrected results along a time axis.

[0497] (Item 3) The accumulation process includes a classification execution process, a correction execution process, and an accumulation execution process, wherein the classification execution process represents a process of classifying the plurality of biometric data into at least a first state group indicating a first physical state and a second state group indicating a second physical state, thereby acquiring a plurality of first biometric data belonging to the first state group and a plurality of second biometric data belonging to the second state group, the correction execution process represents a process of adjusting a balance in size between the plurality of first biometric data and the plurality of second biometric data, thereby calculating a plurality of correction data in time series, and the accumulation execution process represents a process of calculating the cumulative data by accumulating the correction data along a time axis.

[0498] (Item 4) The estimation device according to item 3, wherein the correction execution process includes a process of setting a ratio of a sum of absolute values ​​of the plurality of second biometric data to a sum of absolute values ​​of the plurality of first biometric data to substantially 1.

[0499] (Item 5) The estimation device according to Item 3, wherein the correction execution process includes a calculation that prevents the cumulative data from substantially diverging.

[0500] (Item 6) The estimation device according to any one of items 3 to 5, wherein the first state indicates a mental and physical state that is incompatible with the second state.

[0501] (Item 7) The estimation device according to Item 6, wherein the first state indicates either an active state or an inactive state of the body or mind, either a stable state or an unstable state of the body or mind, or either an appropriate state or an inappropriate state of the body or mind.

[0502] (Item 8) In the estimation device according to any one of Items 3 to 7, the classification execution process includes a process of calculating first classification standard information based on a plurality of pieces of biometric data for a specific period, comparing the first classification standard information with the biometric data to be classified, and classifying the biometric data based on a comparison result.

[0503] (Item 9) The classification execution process includes a process of comparing the biometric data to be classified during an observation period that is the same as the specific period with the first classification standard information, and classifying the biometric data based on a comparison result, and the correction execution process includes a process of adjusting a balance in size between the plurality of first biometric data and the plurality of second biometric data during the observation period based on the plurality of first biometric data and the plurality of second biometric data during the observation period.

[0504] (Item 10) The classification execution process includes a process of comparing the biometric data to be classified during an observation period different from the specific period with the first classification standard information, and classifying the biometric data based on a comparison result, and the correction execution process includes a process of adjusting a balance in size between the plurality of first biometric data and the plurality of second biometric data to be corrected during the observation period based on a plurality of first biometric data and a plurality of second biometric data calculated from the plurality of biometric data during the specific period.

[0505] (Item 11) The estimation device according to Item 3, wherein the classification execution process includes a process of comparing second classification standard information set to a value of the biological data that substantially indicates a cessation of mental and physical activity with the biological data to be classified, and classifying the biological data based on a comparison result.

[0506] (Item 12) The classification execution process includes a process of classifying, among the plurality of biological data to be classified, biological data of a first period into the first state group and classifying biological data of a second period different from the first period into the second state group, wherein the first period indicates a period when a physical and mental state is in a specific state.

[0507] (Item 13) The estimation device according to item 3, wherein the accumulation execution process includes a process of comparing the accumulated data with a threshold value and setting the threshold value as new accumulated data instead of the accumulated data based on a comparison result.

[0508] (Item 14) The estimation device described in any one of Items 1 to 13, wherein the biological-related state information includes information representing the subject's mental and physical state, the information representing the subject's mental and physical state includes information representing a physical state, information representing a mental state, or information regarding quality of life, and the information regarding quality of life indicates quality of life related to a physical state or a mental state.

[0509] (Item 15) The estimation device according to Item 14, wherein the information representing the physical condition includes information relating to a disease or symptom of the circulatory system, information relating to a disease or symptom of the nervous and cranial nervous system, information relating to a disease or symptom of the respiratory system, information relating to a disease or symptom of the endocrine and metabolic system, information relating to a disease or symptom of the immune and inflammatory system, information relating to a disease or symptom of the musculoskeletal system, information relating to a disease or symptom of the sensory system, information relating to a disease or symptom of the gynecological system, or information representing a physical condition that can be indicated by the results of a biopsy.

[0510] (Item 16) The estimation device according to Item 14, wherein the information representing the mental state includes information representing a state related to a mental illness or information representing a state related to a mental stress.

[0511] (Item 17) The estimation device according to any one of items 1 to 13, wherein the bio-related state information includes information representing an environmental state that affects the mental and physical state of the subject, and the information representing an environmental state that affects the mental and physical state of the subject includes information regarding abnormalities in gas components that make up the environment or abnormalities in environmental factors.

[0512] (Item 18) The estimation device according to any one of items 1 to 17, wherein the biological data indicates vital information or behavioral information of the subject.

[0513] (Item 19) The estimation device according to any one of Items 1 to 18, wherein the input information further includes one or more of vital information of the subject, behavioral information of the subject, information related to the subject's environment, self-reported information related to the subject's physical and mental state, and attribute information of the subject.

[0514] (Item 20) The estimation device according to any one of items 1 to 19, wherein the cumulative data is cumulative data calculated by the cumulative processing for a period indicated by a specific condition, or cumulative data calculated by the cumulative processing for a statistical index of the biological data for a period indicated by the specific condition, and the specific condition is a condition related to sleep, a condition related to wakefulness, a condition related to a time of day caused by the sun, or a condition related to activity, or a combination of two or more of these conditions.

[0515] (Item 21) The estimation device according to any one of Items ...

Claims

An estimation device for estimating biological-related state information of a subject, comprising: an estimation unit that inputs input information to a learning model and acquires output information from the learning model; a storage unit that stores the output information, the output information includes the biological-related state information, the biological-related state information includes information representing the mental and physical state of the subject, or information representing an environmental state that affects the mental and physical state of the subject, the input information includes at least cumulative data calculated by cumulative processing of a plurality of biological data of the subject arranged in time series, The accumulation process is a process of accumulating the biological data under the condition that the value does not diverge due to accumulation. 　 The estimation device according to claim 1 , wherein the accumulation processing is a process of classifying the biological data according to the physical and mental state of the subject, correcting the classification results to adjust the balance of values, and accumulating the corrected results along a time axis. 　 the accumulation process includes a classification execution process, a correction execution process, and an accumulation execution process, the classification execution process is a process of classifying the plurality of biometric data into at least a first status group indicating a first physical and mental status and a second status group indicating a second physical and mental status, thereby acquiring a plurality of first biometric data belonging to the first status group and a plurality of second biometric data belonging to the second status group; the correction execution process is a process of calculating a plurality of time-series correction data by adjusting a balance between the plurality of first biometric data and the plurality of second biometric data; The estimation device according to claim 1 or 2, wherein the accumulation execution process represents a process of calculating the accumulated data by accumulating the correction data along a time axis. 　 The estimation device according to claim 3 , wherein the correction execution process includes a process of setting a ratio of a sum of absolute values ​​of the plurality of second biometric data to a sum of absolute values ​​of the plurality of first biometric data to substantially 1. 　 The estimation device according to claim 3 , wherein the correction execution process includes a calculation that prevents the cumulative data from substantially diverging. 　 The estimation device of claim 3 , wherein the first state indicates a physical or mental state that is incompatible with the second state. 　 The estimation device according to claim 6 , wherein the first state indicates either an active or inactive state of the body or mind, either a stable or unstable state of the body or mind, or either an appropriate or inappropriate state for the body or mind.

4. The estimation device according to claim 3, wherein the classification execution process includes a process of calculating first classification standard information based on a plurality of biometric data sets during a specific period, comparing the first classification standard information with the biometric data sets to be classified, and classifying the biometric data sets based on a comparison result. 　 the classification execution process includes a process of comparing the biometric data to be classified during an observation period that is the same as the specific period with the first classification standard information, and classifying the biometric data based on a comparison result; The estimation device according to claim 8, wherein the correction execution process includes a process of adjusting a balance in magnitude between the plurality of first biometric data and the plurality of second biometric data during the observation period based on the plurality of first biometric data and the plurality of second biometric data during the observation period. 　 the classification execution process includes a process of comparing the biometric data to be classified during an observation period different from the specific period with the first classification standard information, and classifying the biometric data based on a comparison result; The estimation device of claim 8, wherein the correction execution process includes a process of adjusting the balance in magnitude between the plurality of first biometric data and the plurality of second biometric data to be corrected during the observation period based on a plurality of first biometric data and a plurality of second biometric data calculated from the plurality of biometric data during the specific period. 　 The estimation device according to claim 3, wherein the classification execution process includes a process of comparing the biometric data to be classified with second classification standard information set to a value of the biometric data that essentially indicates a cessation of mental and physical activity, and classifying the biometric data based on a comparison result. 　 the classification execution process includes a process of classifying biometric data of a first period into the first status group, and classifying biometric data of a second period different from the first period into the second status group, among the plurality of biometric data to be classified, The estimation device according to claim 3 , wherein the first period indicates a period during which the mental and physical state is in a specific state. 　 The estimation device according to claim 3 , wherein the accumulation execution process includes a process of comparing the accumulated data with a threshold value and setting the threshold value as new accumulated data instead of the accumulated data based on a comparison result. 　 the biological-related state information includes information representing the mental and physical state of the subject, The information representing the subject's physical and mental state includes information representing a physical state, information representing a mental state, or information relating to quality of life; The estimation device according to claim 1 or 2, wherein the information on quality of life indicates quality of life related to a physical state or a mental state. 　 The estimation device of claim 14, wherein the information representing the physical condition includes information relating to diseases or symptoms of the circulatory system, information relating to diseases or symptoms of the nervous and cranial nervous systems, information relating to diseases or symptoms of the respiratory system, information relating to diseases or symptoms of the endocrine and metabolic systems, information relating to diseases or symptoms of the immune and inflammatory systems, information relating to diseases or symptoms of the musculoskeletal system, information relating to diseases or symptoms of the sensory system, information relating to diseases or symptoms of the gynecological system, or information representing a physical condition that can be indicated by the results of a biopsy. 　 The estimation device according to claim 14 , wherein the information representing the mental state includes information representing a state related to a mental illness or information representing a state related to a mental stress. 　 the biological-related state information includes information representing an environmental state that affects the mental and physical state of the subject, 3. The estimation device according to claim 1, wherein the information representing the state of the environment that affects the physical and mental state of the subject includes information regarding abnormalities in components of gases that constitute the environment or abnormalities in environmental factors. 　 The estimation device according to claim 1 , wherein the biological data indicates vital information or behavioral information of the subject.

3. The estimation device according to claim 1, wherein the input information further includes one or more of vital information of the subject, behavioral information of the subject, information about the subject's environment, self-reported information about the subject's physical and mental state, and attribute information of the subject. 　 the cumulative data is cumulative data calculated by the cumulative processing for a period indicated by a specific condition, or cumulative data calculated by the cumulative processing for a statistical index of the biological data for a period indicated by the specific condition, 3. The estimation device according to claim 1, wherein the specific condition is a sleep-related condition, a wakefulness-related condition, a sun-related condition related to a time of day, or an activity-related condition, or a combination of two or more of these conditions. 　 The learning model is constructed by learning using a learning dataset, the learning data set includes at least cumulative data calculated by cumulative processing of a plurality of biological data of a learning subject arranged in time series, and biological-related state information of the learning subject; the accumulation process is a process of accumulating the biometric data under a condition that the value does not diverge due to accumulation, 3. The estimation device according to claim 1, wherein the biological-related state information of the learning subject includes information representing the mental and physical state of the learning subject, or information representing an environmental state that affects the mental and physical state of the learning subject. 　 A learning model that causes a computer to function to estimate biological-related state information of a subject, causing the computer to receive input information and output output information; the output information includes the biological-related state information, the biological-related state information includes information representing the mental and physical state of the subject, or information representing an environmental state that affects the mental and physical state of the subject, the input information includes at least cumulative data calculated by cumulative processing of a plurality of biological data of the subject arranged in time series, A learning model in which the accumulation process indicates a process of accumulating the biometric data under the condition that the values ​​do not diverge due to accumulation. 　 An estimation method for estimating biological-related state information of a subject, comprising: inputting input information into a learning model; and acquiring output information from the learning model to which the input information has been input, the output information includes the biological-related state information, the biological-related state information includes information representing the mental and physical state of the subject, or information representing an environmental state that affects the mental and physical state of the subject, the input information includes at least cumulative data calculated by cumulative processing of a plurality of biological data of the subject arranged in time series, The accumulation process is a process of accumulating the biological data under the condition that the value does not diverge due to accumulation. 　 A computer program that causes a computer to execute the estimation method according to claim 23. 　 obtaining a training dataset; and generating a learning model that outputs output information when input information is input by performing learning using the learning data set, the learning data set includes at least cumulative data calculated by cumulative processing of a plurality of biological data of a learning subject arranged in time series, and biological-related state information of the learning subject; the biologically-related state information of the learning subject includes information representing the mental and physical state of the learning subject, or information representing an environmental state that affects the mental and physical state of the learning subject, the input information includes at least cumulative data calculated by cumulative processing of a plurality of biological data of a subject arranged in time series, the output information includes biological-related state information of the subject; the biological-related state information of the subject includes information representing a mental and physical state of the subject, or information representing an environmental state that affects the mental and physical state of the subject, A learning model generation method, wherein the accumulation process indicates a process of accumulating the biological data under the condition that the values ​​do not diverge due to accumulation. 　 A computer program that causes a computer to execute the learning model generation method according to claim 25. 　 a processing unit that acquires a plurality of pieces of biometric data arranged in time series; the processing unit executes accumulation processing on the plurality of biometric data; The accumulation process is a process of accumulating the biometric data under the condition that the value does not diverge due to accumulation.

28. The biometric data processing system according to claim 27, wherein the accumulation process classifies the biometric data according to the physical and mental state, corrects the classification results to adjust the balance of values, and accumulates the corrected results along a time axis. 　 The processing unit a classification unit that classifies the plurality of biometric data to be classified, arranged in chronological order, into at least a first status group indicating a first physical or mental status and a second status group indicating a second physical or mental status, thereby acquiring a plurality of first biometric data belonging to the first status group and a plurality of second biometric data belonging to the second status group; a correction unit that calculates a plurality of time-series correction data by performing a correction calculation on the plurality of first biometric data and the plurality of second biometric data; an accumulator that calculates a plurality of time-series accumulated data by accumulating the correction data along a time axis, 29. The biometric data processing system according to claim 27, wherein the correction calculation is a calculation for adjusting a balance in magnitude between the plurality of first biometric data and the plurality of second biometric data.

30. The biometric data processing system according to claim 29, wherein the correction operation is an operation for making a ratio of a sum of absolute values ​​of the plurality of second biometric data to a sum of absolute values ​​of the plurality of first biometric data substantially 1.

30. The biometric data processing system of claim 29, wherein the correction operation indicates an operation that prevents the cumulative data from substantially diverging.

30. The biometric data processing system of claim 29, wherein the first condition indicates a mental or physical state that is incompatible with the second condition.

33. The biometric data processing system of claim 32, wherein the first state indicates either an active or inactive state of the mind and body, either a stable or unstable state of the mind and body, or either an appropriate or inappropriate state for the mind and body. 　 The classification unit Calculating first classification standard information based on a plurality of biometric data sets for a specific period; 30. The biometric data processing system according to claim 29, wherein the first classification standard information is compared with the biometric data to be classified, and the biometric data is classified based on the comparison result. 　 the classification unit compares the biometric data to be classified during the same observation period as the specific period with the first classification standard information, and classifies the biometric data based on a comparison result; the correction unit performs the correction calculation on the plurality of first biometric data and the plurality of second biometric data during the observation period based on the plurality of first biometric data and the plurality of second biometric data during the observation period; 35. The biological data processing system according to claim 34, wherein the accumulator calculates the plurality of time-series cumulative data at a plurality of times during the observation period. 　 the classification unit compares the biometric data to be classified during an observation period different from the specific period with the first classification standard information, and classifies the biometric data based on a comparison result; the correction unit performs the correction calculation on the plurality of first biometric data and the plurality of second biometric data to be corrected during the observation period based on the plurality of first biometric data and the plurality of second biometric data calculated from the plurality of biometric data during the specific period; 35. The biological data processing system according to claim 34, wherein the accumulator calculates the plurality of time-series cumulative data at a plurality of times during the observation period.

30. The biometric data processing system of claim 29, wherein the classification unit compares the biometric data to be classified with second classification standard information set to a value of the biometric data that substantially indicates a cessation of mental and physical activity, and classifies the biometric data based on the comparison result. 　 the classification unit classifies biometric data of a first period into the first status group, and classifies biometric data of a second period different from the first period into the second status group, among the plurality of biometric data to be classified; 30. The biometric data processing system of claim 29, wherein the first period indicates a period during which a physical or mental state is in a particular state.

30. The biological data processing system according to claim 29, wherein the classifying unit, the correcting unit, and the accumulating unit each execute their respective processes during a period when the physical and mental state is in a specific state.

30. The biological data processing system according to claim 29, wherein the accumulator compares the accumulated data with a threshold value, and sets the threshold value as new accumulated data instead of the accumulated data based on a comparison result. 　 a statistical processing unit that performs centering, standardization, or normalization on the cumulative data; 30. The biometric data processing system according to claim 29, further comprising: an output control unit that causes a terminal to output the processing result by said statistical processing unit or information indicating a feature of said processing result.

42. The biometric data processing system of claim 41, wherein the statistical processing unit performs the centering process, the standardization process, or the normalization process on the cumulative data of the specific biological body based on a statistical index calculated from the cumulative data of the specific biological body.

42. The biometric data processing system of claim 41, wherein the statistical processing unit performs the centering process, the standardization process, or the normalization process on the cumulative data of a specific organism among the plurality of same-type organisms based on a statistical index calculated from the cumulative data of the plurality of same-type organisms. 　 the classification unit, the correction unit, and the accumulation unit perform processing for each of a plurality of attribute groups having different attributes; an analysis unit that derives features according to the attributes of the attribute groups by comparing the cumulative data between the plurality of attribute groups or by comparing information indicating features of the cumulative data between the plurality of attribute groups; 30. The biometric data processing system according to claim 29, further comprising: an output control unit that causes a terminal to output information indicating the characteristics according to the attributes of the attribute group. 　 an analysis unit that directly or indirectly performs frequency analysis on the plurality of time-series cumulative data; 30. The biological data processing system according to claim 29, further comprising: an output control unit that outputs the result of the frequency analysis to a terminal. 　 an analysis unit that directly or indirectly analyzes the plurality of cumulative data in a time series in the past and evaluates past or present mental and physical states or predicts future mental and physical states based on the analysis results; 30. The biological data processing system according to claim 29, further comprising: an output control unit that causes the evaluation result or prediction result by said analysis unit to be output to a terminal. 　 the classification unit, the correction unit, and the accumulation unit perform real-time processing; an analysis unit that directly or indirectly analyzes the plurality of time-series cumulative data in real time and evaluates a current mental and physical state or predicts a future mental and physical state based on the analysis result; 30. The biological data processing system according to claim 29, further comprising: an output control unit that causes the evaluation result or prediction result by said analysis unit to be output to a terminal. 　 a support unit that generates support information for supporting a user based on the evaluation result or the prediction result by the analysis unit; 48. The biometric data processing system according to claim 47, wherein the output control unit causes the terminal to output the support information.

30. The biometric data processing system according to claim 29, further comprising an output control unit that causes a terminal to output the plurality of cumulative data or information indicating characteristics of the plurality of cumulative data. 　 acquiring a plurality of biometric data arranged in time series; and performing an accumulation process on the plurality of biometric data. The biometric data processing method, wherein the accumulation process is a process of accumulating the biometric data under the condition that the value does not diverge due to accumulation. 　 On the computer, acquiring a plurality of biometric data arranged in time series; and executing a cumulative process on the plurality of biometric data. The computer program product indicates a process of accumulating the biometric data under the condition that the accumulation does not cause the value to diverge.

Images