Program, information processing method, and information processing device

Abstract

To provide a program, etc. that can determine an appropriate measurement condition of PPG.SOLUTION: A program acquires a photoelectric pulse wave signal of a plurality of patterns in which output frequencies and output voltages of an LED light source, detected by a photoelectric pulse wave meter, are changed, calculates a score on a photoelectric pulse wave waveform on the basis of the photoelectric pulse wave signal of each pattern, and causes a computer to execute processing for selecting the combination of the output frequency and the output voltage of the photoelectric pulse wave meter on the basis of the score. Preferably, the program calculates a sum total of auto-correlation coefficients of waveform data based on the photoelectric pulse wave signal, or an average of peak values of the auto-correlation coefficients extracted in each section dividing a lag at a predetermined interval as the score.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present invention relates to a program, an information processing method, and an information processing device. [Background technology]

[0002] Photoplethysmography (PPG) is a technique used to measure pulse rate and is a method for predicting chronic diseases such as heart failure.

[0003] For example, Patent Document 1 discloses a physiological index sensor that measures a physiological index using light of two wavelengths: a first target wavelength of 650 nm to 670 nm and a second target wavelength of 840 nm to 860 nm. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Special Publication No. 2018-536516 Summary of the Invention [Problem to be solved by the invention]

[0005] In one aspect, an object of the present invention is to provide a program or the like capable of determining appropriate measurement conditions for PPG. [Means for solving the problem]

[0006] In one aspect, (1) a program causes a computer to execute a process of acquiring multiple patterns of photoplethysmography signals detected by a photoplethysmography meter, in which the output frequency and output voltage of an LED light source are changed, calculating a score related to the photoplethysmography waveform based on the photoplethysmography signal of each pattern, and selecting a combination of the output frequency and output voltage of the photoplethysmography meter based on the score. Here, in one aspect, (2) The program in (1) above calculates the score using an autocorrelation coefficient of waveform data based on the photoplethysmographic signal. (3) The program in (2) above calculates, as the score, the sum of autocorrelation coefficients obtained by adding up the autocorrelation coefficients at all lags of the waveform data based on the photoplethysmographic signal, or the average of the peak values ​​of the autocorrelation coefficients extracted in each section obtained by dividing the lags into predetermined intervals. (4) When calculating the autocorrelation coefficient for each lag, the program in (3) above always calculates the autocorrelation coefficient assuming that the number of data items in the photoplethysmogram signal is the same as the number of data items in each lag shifted in time from the photoplethysmogram signal. (5) The program (3) or (4) above converts the photoplethysmographic signal at time t into X t , photoplethysmogram signal X t When the number of data is n and the lag is k, the autocorrelation coefficient at each lag is calculated using the following formula (1).

number

[0007] In one aspect, (11) a computer executes a process of acquiring multiple patterns of photoplethysmography signals detected by a photoplethysmography meter, in which the output frequency and output voltage of an LED light source are changed, calculating a score related to the photoplethysmography waveform based on the photoplethysmography signal of each pattern, and selecting a combination of the output frequency and output voltage of the photoplethysmography meter based on the score.

[0008] In one aspect, (12) an information processing device including a control unit, the control unit acquires multiple patterns of photoplethysmographic signals detected by a photoplethysmograph in which the output frequency and output voltage of an LED light source are changed, calculates a score related to the photoplethysmographic waveform based on the photoplethysmographic signal of each pattern, and selects a combination of the output frequency and output voltage of the photoplethysmograph based on the score. Effect of the Invention

[0009] In one aspect, appropriate measurement conditions for PPG can be determined. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is an explanatory diagram illustrating an example of the configuration of a PPG measurement system. [Diagram 2] FIG. 2 is a block diagram showing an example of the configuration of a server. [Diagram 3] FIG. 2 is a block diagram showing a configuration example of a wearable device. [Figure 4] 1 is an explanatory diagram showing a PPG waveform according to an output frequency and an output voltage of an LED light source. [Diagram 5] FIG. 2 is an explanatory diagram showing an example of a correlogram based on a PPG signal. [Figure 6]FIG. 2 is an explanatory diagram showing an example of a correlogram based on the formula (1). [Figure 7] 11 is a flowchart illustrating an example of a processing procedure executed by the wearable device. [Figure 8] FIG. 4 is an explanatory diagram showing an example of waveform data based on a PPG signal. [Figure 9] 11 is a flowchart showing an example of a processing procedure executed by the wearable device according to the second embodiment. [Figure 10] FIG. 11 is an explanatory diagram showing an overview of a third embodiment. [Figure 11] 11 is a flowchart showing a procedure for generating a calculation model. [Figure 12] 13 is a flowchart showing an example of a processing procedure executed by the wearable device according to the third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] The present invention will now be described in detail with reference to the drawings showing embodiments thereof. (Embodiment 1) 1 is an explanatory diagram showing an example of the configuration of a PPG measurement system. In this embodiment, a PPG measurement system is described that determines measurement conditions (output frequency and output voltage of an LED (Light Emitting Diode) light source) when measuring the PPG of a subject, and measures the PPG under the determined measurement conditions. The PPG measurement system includes a server 1 and a wearable device 2. Each device is communicatively connected via a network N such as the Internet.

[0012] Wearable device 2 is a device worn on the wrist or the like of the subject, and is a photoplethysmograph that measures the subject's PPG. For example, wearable device 2 includes an LED light source and an LED light receiving element, and measures the PPG by emitting infrared light or green LED light to the subject's skin and receiving the reflected light of the LED light.

[0013] However, when measuring PPG, there is a problem that the measured PPG waveform is affected by the skin color of the subject, ambient light, etc. Therefore, when a single light source is used, there may be subjects whose skin color makes it difficult to measure, or there may be times when measurement is impossible due to the ambient light.

[0014] Therefore, in this embodiment, before starting PPG measurement (hereinafter referred to as "main measurement") for a predetermined period of time (e.g., several tens of minutes) continuously by wearable device 2, calibration is performed to adjust the output frequency (color) and output voltage (output strength) of the LED light source so that the main measurement can be performed without being affected by the characteristics of the subject or ambient light. This process will be described in detail later.

[0015] The server 1 is a server computer capable of various information processing and sending and receiving information, and analyzes the PPG waveform measured by the wearable device 2.

[0016] 2 is a block diagram showing an example of the configuration of the server 1. The server 1 includes a control unit 11, a main storage unit 12, a communication unit 13, and an auxiliary storage unit . The control unit 11 has one or more arithmetic processing devices such as a central processing unit (CPU), a micro-processing unit (MPU), a graphics processing unit (GPU), etc., and performs various information processing, control processing, etc. by reading and executing programs stored in the auxiliary storage unit 14. The main storage unit 12 is a temporary storage area such as a static random access memory (SRAM) or a dynamic random access memory (DRAM), and temporarily stores data required for the control unit 11 to execute arithmetic processing. The communication unit 13 is a communication module for performing processing related to communication, and transmits and receives information to and from the outside. The auxiliary storage unit 14 is a non-volatile storage area such as a large-capacity memory or a hard disk, and stores programs (program products) and other data required for the control unit 11 to execute processing.

[0017] The auxiliary storage unit 14 may be an external storage device connected to the server 1. The server 1 may be a multi-computer consisting of a plurality of computers, or may be a virtual machine virtually constructed by software.

[0018] In the present embodiment, the server 1 is not limited to the above configuration, and may include, for example, an input unit for accepting operation input, a display unit for displaying images, etc. The server 1 may also include a reading unit for reading a portable storage medium 1a such as a CD (Compact Disk)-ROM or a DVD (Digital Versatile Disc)-ROM, and may read and execute a program from the portable storage medium 1a.

[0019] 3 is a block diagram showing an example of the configuration of the wearable device 2. The wearable device 2 includes a control unit 21, a storage unit 22, a communication unit 23, a light emitting unit 24, a light receiving unit 25, and an acceleration detection unit . The control unit 21 has one or more processors such as CPUs, and performs various information processing by reading and executing programs stored in the storage unit 22. The storage unit 22 is a memory such as RAM or ROM, and stores programs (program products) and other data necessary for the control unit 21 to execute processing. The communication unit 23 is a communication module for performing communication-related processing, and transmits and receives information to and from the outside. The light-emitting unit 24 is an LED light source, and emits LED light. The light-receiving unit 25 is a light-receiving element that receives the LED light, and receives the light reflected from the skin of the person being measured. The acceleration detection unit 26 is an acceleration sensor, and detects acceleration due to the movement of the person being measured.

[0020] The wearable device 2 may include a reading unit that reads a portable storage medium 2a such as a CD-ROM, and may read and execute a program from the portable storage medium 2a.

[0021] Fig. 4 is an explanatory diagram showing a PPG waveform according to the output frequency and output voltage of an LED light source. Fig. 4 shows six patterns of PPG waveforms with the LED light set to "infrared light" or "green" and the output set to "Low", "Middle" or "High". The following describes an outline of this embodiment.

[0022] In this embodiment, wearable device 2 measures multiple patterns of PPG signals with different output frequencies and output voltages in order to calibrate the output frequency and output voltage before performing the main measurement. Specifically, as shown in Fig. 4, six patterns of PPG signals are measured with the LED light set to "infrared light" or "green" and the output set to "Low", "Middle" or "High". Wearable device 2 drives the LED light source for several seconds with the output frequency and output voltage of each pattern and receives the LED light to obtain a PPG signal corresponding to each pattern.

[0023] In order to determine which of the measurement conditions for each pattern is the optimal measurement condition, the wearable device 2 calculates a score for evaluating the suitability of the PPG waveform from the PPG signal for each pattern.

[0024] Specifically, the wearable device 2 performs an autocorrelation analysis on the PPG signal of each pattern and calculates a score using the autocorrelation coefficient of the waveform data based on the PPG signal. For example, as described below, the wearable device 2 calculates the score as the sum of the autocorrelation coefficients obtained by adding up the autocorrelation coefficients at all lags of the waveform data based on the PPG signal.

[0025] FIG. 5 is an explanatory diagram showing an example of a correlogram based on a PPG signal. A correlogram is a graph that visualizes the results of an autocorrelation analysis, and is a graph that shows the autocorrelation coefficient between original data and data that is shifted in time from the original data (lag). The horizontal axis of the correlogram shows the lag, and the vertical axis shows the autocorrelation coefficient. The lag represents the degree of time shift from the original data. In general, the autocorrelation coefficient for each lag is calculated by the following formula (2).

[0026]

number

[0027] In addition, X t is the PPG signal at time t, and n is X t where k is the number of data points, and k is the lag. When periodicity is observed in the waveform data based on the PPG signal, that is, when a suitable PPG waveform is shown, the sum of the autocorrelation coefficients for each lag becomes large. Therefore, in this embodiment, the "sum of autocorrelation coefficients" obtained by adding up the autocorrelation coefficients for all lags is used as the score.

[0028] However, when the sum of autocorrelation coefficients calculated based on formula (2) is used, the following problems arise. As the first problem, since the PPG signal used to calculate the score is short (several seconds), as shown by the arrow in Fig. 5, the number of data for observing the autocorrelation relationship decreases toward the latter lag, resulting in a smaller autocorrelation coefficient. As the second problem, the data is weighted toward the high frequency band. As the third problem, since the autocorrelation coefficient is calculated as an absolute value, a positive value is calculated even in the case of a negative correlation. As the fourth problem, if periodic high-frequency noise is added to the PPG waveform, the score becomes high.

[0029] Therefore, in this embodiment, the autocorrelation coefficient is calculated based on the following formula (1) instead of formula (2).

[0030]

number

[0031] Comparing formula (2) and formula (1), it is noteworthy that while formula (2) uses the sum of t=k+1 to n in the numerator, formula (1) uses the sum of t=k+1 to k+n. In other words, formula (2) calculates the autocorrelation coefficient with the number of data of the original PPG signal and the number of data of each lag always being the same. This solves the problem of the number of data decreasing towards the later lags.

[0032] In addition, while the absolute value is taken in formula (2), the absolute value is removed in formula (1) to calculate the autocorrelation coefficient. This makes it possible to calculate both positive and negative correlation coefficients.

[0033] FIG. 6 is an explanatory diagram showing an example of a correlogram based on Equation (1). As shown in FIG. 6, when the autocorrelation coefficient is calculated based on Equation (1), the autocorrelation coefficient periodically takes positive and negative values. Also, as shown by the arrow in FIG. 6, the autocorrelation coefficient is high even in the latter lag. The wearable device 2 calculates the sum of the autocorrelation coefficients in the correlogram shown in FIG. 6 as a score.

[0034] In this embodiment, the sum of the autocorrelation coefficients is calculated as the score, but the present embodiment is not limited to this. For example, the wearable device 2 may calculate, as the score, the average value of the peak values ​​(indicated by dashed circles in FIG. 6) of the autocorrelation coefficients extracted in each section obtained by dividing the lag at a predetermined interval. When periodicity is observed in the waveform data based on the PPG signal, the periodically appearing peak values ​​are generally high, so that the PPG signal can be appropriately evaluated by calculating the average value of the peak values ​​as the score. In addition, the total value of the sum of the autocorrelation coefficients at a predetermined number of lags among all the lags may be calculated as the score. In particular, by calculating the total value of the sum of the autocorrelation coefficients at a lag corresponding to the expected period of the PPG waveform and the lags before and after it as the score, it is possible to appropriately evaluate whether or not the PPG signal includes a PPG waveform.

[0035] The wearable device 2 calculates the autocorrelation coefficient of each lag for the PPG signal of each pattern in which the output frequency and output voltage of the LED light source are changed based on the formula (1), and calculates the sum of the autocorrelation coefficients as a score. Then, the wearable device 2 selects an optimal combination of the output frequency and output voltage based on the calculated score. For example, the wearable device 2 selects the pattern with the maximum score as the optimal combination of the output frequency and output voltage.

[0036] The wearable device 2 drives the LED light source at the selected output frequency and output voltage, and performs a main measurement to measure the PPG for a predetermined time (e.g., several tens of minutes). After completing the main measurement, the wearable device 2 transmits the PPG signal to the server 1, and the server 1 analyzes the PPG signal (e.g., extracts the pulse rate).

[0037] 7 is a flowchart showing an example of a processing procedure executed by the wearable device 2. The processing contents executed by the wearable device 2 will be described with reference to FIG. The control unit 21 of the wearable device 2 sets an output frequency and an output voltage of the LED light source (light-emitting unit 24) (step S11). The control unit 21 drives the LED light source based on the set output frequency and output voltage (step S12). The control unit 21 receives reflected light of the LED light emitted from the LED light source to obtain a PPG signal (step S13).

[0038] The control unit 21 performs an autocorrelation analysis of the acquired PPG signal (step S14). That is, the control unit 21 calculates the autocorrelation coefficient between the original PPG signal and the PPG signal at each lag shifted by a certain time from the PPG signal. In this case, the control unit 21 calculates the autocorrelation coefficient by always assuming the number of data of the original PPG signal and the number of data of each lag shifted by a certain time from the PPG signal to be the same number of data. Specifically, as described above, the control unit 21 calculates the autocorrelation coefficient according to the formula (1).

[0039] The control unit 21 calculates a score for the PPG waveform based on the result of the autocorrelation analysis (step S15). Specifically, the control unit 21 calculates the sum of the autocorrelation coefficients of the waveform data based on the PPG signal as the score. As described above, the control unit 21 may calculate the average value of the peak values ​​of the autocorrelation coefficients extracted in each section obtained by dividing the lag at a predetermined interval as the score.

[0040] The control unit 21 determines whether or not the scores have been calculated for all combinations of output frequencies and output voltages (step S16). If it is determined that the scores have not been calculated for all combinations (S16: NO), the control unit 21 returns the process to step S11. In this case, the control unit 21 obtains PPG signals for different patterns of output frequencies and output voltages, and calculates the scores.

[0041] When it is determined that the scores have been calculated for all combinations (S16: YES), the control unit 21 selects an optimal combination of output frequency and output voltage of the photoplethysmograph (wearable device 2) based on the score calculated for each pattern (step S17). The control unit 21 sets the LED light source to the output frequency and output voltage of the selected combination, and executes the main measurement of the PPG signal (step S18). The control unit 21 ends the series of processes.

[0042] In the above description, the wearable device 2 is described as performing a series of calibration processes, but the server 1 on the cloud may acquire a PPG signal from the wearable device 2, calculate a score, select an optimal combination of output frequency and output voltage, and cause the wearable device 2 to perform the actual measurement.

[0043] In addition, in the present embodiment, calculation of a score and the like is performed as pre-processing before performing the main measurement, but the present embodiment is not limited to this. For example, the wearable device 2 may perform the main measurement under a plurality of measurement conditions in which the output frequency and output voltage of the LED light source are changed, calculate a score from the main measurement data of each pattern, and select the main measurement data to be adopted as the measurement result based on the calculated score. In other words, the wearable device 2 may perform a series of processes as post-processing for selecting the main measurement data to be adopted as the measurement result.

[0044] As described above, according to the first embodiment, appropriate measurement conditions for PPG can be determined.

[0045] (Embodiment 2) In this embodiment, a description will be given of a form in which the influence of amplitude changes is eliminated based on the envelope of the PPG signal so that the periodicity of the PPG waveform can be appropriately evaluated. Note that the same reference numerals are used to designate the same contents as in the first embodiment, and the description will be omitted.

[0046] Fig. 8 is an explanatory diagram showing an example of waveform data based on a PPG signal. Fig. 8 shows waveform data (solid line) of a PPG signal obtained from a subject and an envelope curve (dashed line) of the waveform data. An outline of the present embodiment will be described with reference to Fig. 8.

[0047] In the present embodiment, as in the first embodiment, the wearable device 2 performs an autocorrelation analysis of the PPG signal and calculates a score for evaluating the appropriateness of the PPG waveform. Here, in the present embodiment, the wearable device 2 derives the envelope of the PPG signal and generates a correlogram based on waveform data obtained by dividing the original PPG signal by the envelope.

[0048] As described in the first embodiment, when the waveform data based on the PPG signal shows a periodicity in time, the sum of the autocorrelation coefficients becomes large. On the other hand, when the amplitude change also has a periodicity in addition to the periodicity in time of the waveform itself, the sum of the autocorrelation coefficients becomes large due to the influence of the periodicity. Therefore, in this embodiment, the influence of the amplitude change is eliminated by dividing the original PPG signal by the envelope.

[0049] When the wearable device 2 acquires the PPG signal of each pattern in which the output frequency and output voltage of the LED light source are changed, the wearable device 2 derives the envelope of the PPG signal. The wearable device 2 then divides the value at each time point on the waveform data based on the PPG signal by the value at each time point on the envelope to derive waveform data from which the influence of the change in amplitude has been removed. The subsequent processing is the same as in the first embodiment, and the wearable device 2 performs an autocorrelation analysis of the waveform data, calculates the sum of the autocorrelation coefficients as a score, and selects the pattern with the largest calculated score as the optimal combination of output frequency and output voltage.

[0050] 9 is a flowchart showing an example of a processing procedure executed by the wearable device 2 according to embodiment 2. After acquiring the PPG signal (step S13), the wearable device 2 executes the following processing. The control unit 21 of the wearable device 2 derives the envelope of the PPG signal (step S201). The control unit 21 performs an autocorrelation analysis based on waveform data obtained by dividing the original PPG signal by the envelope (step S202). The control unit 21 calculates a score based on the result of the autocorrelation analysis (step S203). The control unit 21 transitions the process to step S16.

[0051] As described above, according to the second embodiment, the periodicity of the PPG waveform can be evaluated more appropriately.

[0052] (Embodiment 3) In this embodiment, a form will be described in which an optimal combination of an output frequency and an output voltage is selected using a machine learning model.

[0053] Fig. 10 is an explanatory diagram showing an overview of the third embodiment. Fig. 10 conceptually illustrates how a score related to a PPG waveform is calculated by inputting various data into a calculation model 50 constructed by machine learning. The overview of the present embodiment will be described with reference to Fig. 10.

[0054] In this embodiment, a calculation model 50 that calculates a score when a feature amount such as a PPG signal is input is constructed, and a score related to a PPG waveform is calculated using the calculation model 50. The calculation model 50 is a machine learning model generated by learning predetermined training data, for example, XGBoost. Note that the calculation model 50 is not limited to XGBoost, and may be other machine learning models such as logistic regression and AutoML (Automated Machine Learning).

[0055] The server 1 generates the calculation model 50 by using training data in which correct scores are associated with a large amount of PPG signals (data measured in the past) provided for training. Specifically, in addition to the PPG signal, the training data includes the measurement time of the PPG signal, acceleration measured by the wearable device 2 simultaneously with the PPG signal, and subject information on the subject of the measurement of the PPG signal. The server 1 extracts feature amounts to be used as input parameters for the calculation model 50 from these data.

[0056] Specifically, the server 1 performs a Fast Fourier Transform (FFT) on the PPG signal and extracts the highest peak frequency, the amplitude at the peak frequency, velocity, etc. as feature quantities. The server 1 also extracts the posture and resultant acceleration of the subject from the acceleration. The server 1 also extracts data such as the time period, day of the week, season, etc. from the measurement time. The server 1 also extracts data such as the gender, race, etc. of the subject from the subject information.

[0057] Note that the above is an example of the input parameters, and the present embodiment is not limited to this. For example, the server 1 may extract the calibration setting values ​​themselves (output frequency and output voltage of the LED light source) as feature quantities and use them as input parameters.

[0058] The server 1 generates the calculation model 50 based on the extracted feature amount and the correct score. That is, the server 1 calculates a score by inputting the feature amount extracted from the training PPG signal or the like to the calculation model 50, and trains the calculation model 50 so that the calculated score approximates the correct score.

[0059] The data of the calculation model 50 generated by the server 1 is installed in the wearable device 2 via the network N, for example.

[0060] When actually calculating the score, the wearable device 2 drives the LED light source to obtain a PPG signal, and measures acceleration with an acceleration sensor (acceleration detection unit 26). Furthermore, for example, the wearable device 2 obtains subject information from the server 1. In addition, the wearable device 2 obtains the current measurement time.

[0061] Then, the wearable device 2 extracts features from data such as the PPG signal, the measurement time, the acceleration, and the subject information, and calculates a score by inputting the features to the calculation model 50. The wearable device 2 calculates a score by performing similar processing for each pattern of measurement conditions (combination of output frequency and output voltage), and selects the pattern with the maximum score as the optimal combination of output frequency and output voltage.

[0062] Fig. 11 is a flowchart showing the procedure of the generation process of the calculation model 50. The process content when generating the calculation model 50 by machine learning will be described with reference to Fig. 11. The control unit 11 of the server 1 acquires training data for generating the calculation model 50 (step S301). The training data is data in which a correct score is associated with a training PPG signal, a measurement time of the PPG signal, acceleration measured simultaneously with the PPG signal, subject information, and the like.

[0063] The control unit 11 extracts predetermined feature amounts from the training PPG signal and the like (step S302). For example, the control unit 11 performs a fast Fourier transform on the PPG signal and extracts the highest peak frequency, the amplitude of the peak frequency, the velocity, and the like. The control unit 11 also extracts the posture and resultant acceleration of the subject from the acceleration. The control unit 11 also extracts data such as the time period, day of the week, season, and the like from the measurement time. The control unit 11 also extracts data such as the sex and race of the subject from the subject information.

[0064] The control unit 11 generates a trained calculation model 50 to calculate a score when a feature amount such as a PPG signal is input based on the feature amount extracted in step S302 and the correct score included in the training data acquired in step S301 (step S303). For example, the control unit 11 performs training using XGBoost or the like to generate the calculation model 50. The control unit 11 ends the series of processes.

[0065] 12 is a flowchart showing an example of a processing procedure executed by the wearable device 2 according to embodiment 3. After acquiring the PPG signal (step S13), the wearable device 2 executes the following processing. The control unit 21 of the wearable device 2 acquires the measurement time of the PPG signal, the acceleration measured simultaneously with the PPG signal, subject information, etc. (step S321). The control unit 21 extracts predetermined feature amounts from the PPG signal acquired in steps S13 and S321, etc. (step S322). The control unit 21 inputs the extracted feature amounts into the calculation model 50 to calculate a score related to the PPG waveform (step S323). The control unit 21 transitions the process to step S16.

[0066] As described above, according to the third embodiment, appropriate measurement conditions for PPG can be determined.

[0067] The embodiments disclosed herein are illustrative in all respects and should not be considered as limiting. The scope of the present invention is defined by the claims, not by the above meaning, and is intended to include all modifications within the scope and meaning equivalent to the claims.

[0068] The matters described in each embodiment can be combined with each other. In addition, the independent claims and dependent claims described in the claims can be combined with each other in any and all combinations regardless of the citation format. Furthermore, the claims use a format in which a claim cites two or more other claims (multi-claim format), but this is not limited to this. They may also be written using a format in which a multiple claim cites at least one other multiple claim (multi-multi claim). [Explanation of symbols]

[0069] 1 Server 11 Control section 12 Main memory 13. Communications Department 14 Auxiliary storage 2. Wearable Devices 21 Control section 22 Memory section 23 Communications Department 24 Light emitting part 25 Light receiving part 26 Acceleration detection section

Claims

1. A plurality of patterns of photoplethysmographic signals are obtained by changing the output frequency and output voltage of the LED light source, which are detected by a photoplethysmograph. calculating a score for the photoplethysmographic waveform based on the photoplethysmographic signal of each pattern; A combination of the output frequency and the output voltage of the photoplethysmograph is selected based on the score. A program that causes a computer to carry out processing.

2. The score is calculated using an autocorrelation coefficient of waveform data based on the photoplethysmographic signal. The program according to claim 1.

3. The score is calculated as the sum of autocorrelation coefficients obtained by adding up the autocorrelation coefficients at all lags of the waveform data based on the photoplethysmographic signal, or the average value of peak values ​​of autocorrelation coefficients extracted in each section obtained by dividing the lags into predetermined intervals. The program according to claim 2.

4. When calculating the autocorrelation coefficient for each lag, the number of data items in the photoelectric pulse wave signal and the number of data items in each lag shifted in time from the photoelectric pulse wave signal are always set to be the same number of data items in the calculation of the autocorrelation coefficient. The program according to claim 3.

5. Photoplethysmography signal at time t is X ｔ , photoplethysmogram signal X ｔ When the number of data is n and the lag is k, the autocorrelation coefficient at each lag is calculated by the following formula (1): The program according to claim 3. [0010]

6. Deriving an envelope of the photoplethysmographic signal; The score is calculated based on the waveform data obtained by dividing the photoplethysmographic signal by the envelope. The program according to claim 3.

7. extracting a predetermined feature value from the photoplethysmographic signal; The extracted feature quantity is input to a model that has been trained to calculate the score when the feature quantity is input, thereby calculating the score. The program according to claim 1.

8. Acquire a measurement time of the photoplethysmogram signal, an acceleration measured simultaneously with the photoplethysmogram signal, or subject information regarding a subject to be measured by the photoplethysmogram meter; extracting feature amounts from the photoplethysmographic signal and the measurement time, acceleration, or subject information; The extracted feature amount is input to the model to calculate the score. The program according to claim 7.

9. Based on the selected combination of output frequency and output voltage, a photoplethysmogram is measured. The program according to any one of claims 1 to 8.

10. Each of the plurality of patterns of photoelectric pulse wave signals is measurement data obtained by continuing measurement of a photoelectric pulse wave for a predetermined period of time, By selecting a combination of the output frequency and the output voltage, the measurement data to be adopted as the measurement result is selected. The program according to any one of claims 1 to 8.

11. A plurality of patterns of photoplethysmographic signals are obtained by changing the output frequency and output voltage of the LED light source, which are detected by a photoplethysmograph. calculating a score for the photoplethysmographic waveform based on the photoplethysmographic signal of each pattern; A combination of the output frequency and the output voltage of the photoplethysmograph is selected based on the score. An information processing method in which processing is performed by a computer.

12. An information processing device including a control unit, The control unit: A plurality of patterns of photoplethysmographic signals are obtained by changing the output frequency and output voltage of the LED light source, which are detected by a photoplethysmograph. calculating a score for the photoplethysmographic waveform based on the photoplethysmographic signal of each pattern; A combination of the output frequency and the output voltage of the photoplethysmograph is selected based on the score. Information processing device.

Images