Program, information processing method, and information processing device

Abstract

This program causes a computer to execute processing of: acquiring biological data and exercise data of a subject at various points in time in a predetermined period; calculating a parameter related to a heart at the various points in time on the basis of the biological data; calculating an activity amount of the subject at the various points in time on the basis of the exercise data; and calculating an index value related to exacerbation of heart failure on the basis of the parameter and the activity amount at the various points in time.

Description

Program, information processing method and information processing device

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

[0002] Various methods for assisting in the diagnosis of heart failure have been proposed. For example, Patent Literature 1 discloses a device for determining the state of congestive heart failure based on physiological values ​​such as bioimpedance, ECG (Electrocardiogram), and triaxial acceleration.

[0003] Patent No. 6455843

[0004] In one aspect, an object of the present invention is to provide a program or the like capable of calculating an index value related to the worsening of heart failure.

[0005] In one aspect, (1) a program causes a computer to execute a process of acquiring biological data and exercise data of a subject at each time point during a predetermined period, calculating heart-related parameters at each time point based on the biological data, calculating the subject's activity level at each time point based on the exercise data, and calculating an index value related to worsening heart failure based on the parameters and activity level at each time point. Here, in one aspect, (2) the program of (1) calculates an index value related to worsening heart failure based on the correlation between the parameter and a time point or time period at which the value related to the activity level is equal to or greater than a certain value. (3) the program of (2) calculates an index value related to worsening heart failure based on the distribution of the parameter with respect to the activity level at an active time point among the time points at which the activity level is equal to or greater than a certain value. (4) the program of (3) calculates the index value based on the activity level at the active time point and the parameter at a time point a predetermined time after the activity level was measured. (5) The program of (4) acquires the biological data and exercise data for a first period and a second period, calculates the parameter and the activity amount, and compares distribution information on the distribution of the parameter at a time point a predetermined time has elapsed since the activity amount was measured at the active time point in the first period with distribution information on the distribution of the parameter at a time point a predetermined time has elapsed since the activity amount was measured at the active time point in the second period to calculate the index value. (6) The program of (4) calculates a first statistical quantity representing the distribution of the parameter at a time point a predetermined time has elapsed since the activity amount was measured at the active time point as the index value. (7) The program of any of (4) to (6) acquires the biological data and exercise data of the subject for a most recent period to calculate the parameter and the activity amount, derives a regression line for the most recent period that shows the relationship between the activity amount at the active time point and the parameter at a time point a predetermined time has elapsed since the activity amount was measured, and calculates the index value based on the slope of the regression line.(8) The program described in any one of (3) to (7) above acquires the biological data and exercise data for each of a first time period and a second time period to calculate the parameter and activity amount, calculates a difference value between the parameter at each time point in the first time period and a moving average of the parameter for each of the first time period and the second time period, and calculates the index value by comparing difference value distribution information regarding the distribution of the difference values ​​at the active time points in the first time period with difference value distribution information regarding the distribution of the difference values ​​at the active time points in the second time period. (9) The program described in any one of (3) to (7) above calculates a difference value between the parameter at each time point and a moving average of the parameter, and calculates a second statistical quantity representing the distribution of the difference values ​​at the active time points as the index value. (10) The program described in any one of (2) to (9) above identifies a time period in which a statistical quantity representing the distribution of the activity amount is equal to or greater than a certain value from the distribution of the activity amount for each time period, and calculates a statistical quantity regarding the distribution of the parameter for the identified time period as the index value. (11) The program described in any one of (1) to (10) above acquires the biological data and the motion data for a first time period and a second time period, calculates the parameter and the activity amount, separates the parameter and the activity amount into a trend component, a periodic component, and a residual component for each of the first time period and the second time period, extracts periodic components for the same interval from the periodic components of the parameter and the activity amount for each of the first time period and the second time period, and calculates the index value by comparing the extracted periodic components of the parameter and the activity amount between the first time period and the second time period. (12) The program described in any one of (1) to (10) above separates the parameter and the activity amount into a trend component, a periodic component, and a residual component, extracts periodic components of the parameter and the activity amount at each time point included in a predetermined time width, and calculates a correlation coefficient of the cross-correlation between the periodic component of the parameter and the periodic component of the activity amount as the index value based on the extracted periodic components of the parameter and the activity amount.(13) The program described in any of (1) to (12) above acquires the biological data and exercise data of the subject for a most recent period to calculate the parameters and activity level, acquires reference information obtained based on the reference activity level and heart-related reference parameters of a reference subject other than the subject at each reference time point during the reference period, and calculates an index value from the comparison result between the target information based on the parameters and activity level for the most recent period and the reference information.

[0006] In one aspect, (14) an information processing method includes a computer executing a process of acquiring biometric data and exercise data of a subject at each time point during a predetermined period, calculating heart-related parameters at each time point based on the biometric data, calculating the subject's activity level at each time point based on the exercise data, and calculating an index value related to the worsening of heart failure based on the parameters and activity level at each time point.

[0007] In one aspect, (15) an information processing device is an information processing device that includes a control unit, and the control unit acquires biometric data and exercise data of a subject at each time point during a predetermined period, calculates heart-related parameters at each time point based on the biometric data, calculates the subject's activity level at each time point based on the exercise data, and calculates an index value related to the worsening of heart failure based on the parameters and activity level at each time point.

[0008] In one aspect, an index value for worsening heart failure can be calculated.

[0009] FIG. 1 is an explanatory diagram showing an example of the configuration of a heart failure exacerbation determination system. FIG. 2 is a block diagram showing an example of the configuration of a server. FIG. 3 is a block diagram showing an example of the configuration of a wearable device. FIG. 4 is an explanatory diagram relating to a first index value. FIG. 5 is an explanatory diagram relating to a second index value. FIG. 6 is an explanatory diagram relating to a third index value. FIG. 7 is an explanatory diagram relating to a third index value. FIG. 8 is an explanatory diagram relating to a fourth index value. FIG. 9 is an explanatory diagram relating to a fifth index value. FIG. 10 is an explanatory diagram relating to a fifth index value. FIG. 11 is a flowchart showing an example of a processing procedure executed by a server. FIG. 12 is a flowchart showing an example of a processing procedure executed by a server according to a modified example.

[0010] The present invention will be described in detail below with reference to the drawings showing embodiments thereof. (Embodiment 1) Fig. 1 is an explanatory diagram showing an example of the configuration of a heart failure exacerbation determination system. In this embodiment, a heart failure exacerbation determination system that calculates an index value related to the exacerbation of heart failure from biological data and exercise data of a subject will be described. The heart failure exacerbation determination system includes a server 1 and a wearable device 2. Each device is communicatively connected via a network N such as the Internet.

[0011] The server 1 is a server computer capable of various information processing and information transmission and reception, and analyzes the subject's (patient's) biological data and exercise data measured by the wearable device 2. As will be described in detail later, the server 1 calculates heart-related parameters (e.g., pulse rate) from the subject's biological data and calculates the subject's activity level from the exercise data. The server 1 then calculates an index value related to the worsening of heart failure based on the calculated parameters and activity level.

[0012] The wearable device 2 is a device worn on the subject's wrist or the like and measures the subject's biological data and movement data. Specifically, the wearable device 2 measures PPG (Photoplethysmography) signals as biological data and acceleration as movement data. For example, the wearable device 2 includes an LED (Light Emitting Diode) light source and an LED light receiving element, and measures the PPG signal by emitting infrared light or green LED light to the subject's skin and receiving the reflected light of the LED light. The wearable device 2 also has a built-in acceleration sensor that measures acceleration during PPG measurement.

[0013] In this embodiment, PPG is measured as the biological data, but the present embodiment is not limited to this. For example, ECG, respiration, blood pressure, etc. may be measured as the biological data, and heart-related parameters (e.g., heart rate, heart rate variability, etc. for ECG; respiratory rate, respiratory complexity, etc. for respiration; arterial pressure, venous pressure, etc. for blood pressure) may be calculated from these data.

[0014] In addition, although the present embodiment describes measuring acceleration as the motion data, the present embodiment is not limited to this. For example, angular velocity measured by a gyro sensor may be used as the motion data, and the amount of activity of the subject may be calculated from the angular velocity.

[0015] FIG. 2 is a block diagram showing an example configuration of the server 1. The server 1 has a control unit 11 and one or more arithmetic processing devices such as a central processing unit (CPU), a micro-processing unit (MPU), or a graphics processing unit (GPU), and performs various information processing, control processing, and the like by reading and executing programs stored in an 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 communication-related processing 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.

[0016] 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 multiple computers, or may be a virtual machine virtually constructed by software.

[0017] Furthermore, in this embodiment, the server 1 is not limited to the above configuration, and may include, for example, an input unit that accepts operation input, a display unit that displays images, etc. Furthermore, the server 1 may be provided with a reading unit that reads 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.

[0018] FIG. 3 is a block diagram showing an example 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 26. The control unit 21 has one or more processors such as a CPU, 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 a RAM or ROM, and stores programs (program products) and other data required 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 that emits LED light. The light-receiving unit 25 is a light-receiving element that receives the LED light and receives light reflected from the subject's skin. The acceleration detection unit 26 is an acceleration sensor that detects acceleration due to the subject's movements.

[0019] The wearable device 2 may also be provided with 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.

[0020] 4 is an explanatory diagram of the first index value. Fig. 4 shows a scatter diagram in which a parameter related to the heart (pulse rate) calculated from biological data (PPG signal) and the subject's activity amount calculated from exercise data (acceleration) are plotted. An outline of this embodiment will be described below.

[0021] As already described, the server 1 acquires the subject's biological data and exercise data from the wearable device 2 and calculates an index value related to the worsening of heart failure from the biological data and exercise data. For example, the server 1 periodically communicates with the wearable device 2 and acquires the biological data and exercise data at each point in time during a predetermined period.

[0022] The server 1 calculates heart-related parameters at each time point during the measurement period based on the acquired biological data. Specifically, the server 1 acquires a PPG signal as biological data and calculates the pulse rate from the PPG signal.

[0023] The server 1 also calculates the activity amount of the subject at each time point during the measurement period based on the exercise data acquired from the wearable device 2. For example, the server 1 acquires acceleration as exercise data, and calculates a numerical value relating to the activity amount from the acceleration according to the following formula (1).

[0024]

[0025] Here, x, y, and z are accelerations in the x-axis, y-axis, and z-axis directions, respectively, and g is gravitational acceleration. For example, the server 1 counts the number of times the value calculated by formula (1) exceeds a predetermined value within a predetermined time period as activity intensity over time, and calculates the moving average of the activity intensity obtained as time-series data over a predetermined time window width as the activity amount. Here, by using the moving average of activity intensity as the activity amount, the influence of momentary movements such as swinging the arms can be suppressed, and the activity amount can be determined to better reflect whole-body activity that places a strain on the heart.

[0026] The server 1 calculates an index value relating to the worsening of heart failure based on the parameters and the amount of activity calculated as described above. In this embodiment, the server 1 calculates first to fifth index values, which will be described below.

[0027] First, the process for calculating the first index value will be described. Fig. 4 shows scatter diagrams for each of the exacerbation and non-exacerbation periods, plotting the amount of activity and the parameter a predetermined time after (e.g., one hour) has elapsed since the measurement of the amount of activity, based on data from multiple points in time when the amount of activity was equal to or greater than a certain value. The horizontal axis of the scatter diagram represents the amount of activity, and the vertical axis represents the parameter (pulse rate).

[0028] Comparing the two scatter plots, as shown by the dashed rectangular line in Figure 4, the distribution of parameters during the exacerbation phase often takes higher values ​​than the distribution of parameters during the non-exacerbation phase. Therefore, the distribution of the parameters based on data obtained over a predetermined period (e.g., two weeks) can be said to reflect the state of the subject's heart failure over that period. Therefore, the server 1 calculates the difference in the distribution of the parameters as an index by calculating the first index value as follows.

[0029] First, the server 1 calculates the above parameters and activity amounts at each point in time for each of a past period (first period) and a most recent period (second period) from the biological data and exercise data for each of the past period and the most recent period. The past period is, for example, a period from the present time up to a certain time (e.g., six months ago). The most recent period is, for example, a period between the present time and a certain time (e.g., the past two weeks). Furthermore, the past period (first period) is a period that a medical professional has confirmed in advance as being in a non-exacerbation period or an exacerbation period, and by storing the parameters and activity amounts indicating the cardiac condition during that period in the server 1, they serve as a reference for comparing with the most recent period (second period) to calculate an index value.

[0030] The server 1 identifies each time point (hereinafter referred to as an "active time point") at which the amount of activity was equal to or greater than a certain value for each past period and the most recent period, and identifies a parameter (pulse rate) at each time point a predetermined time (e.g., one hour) has elapsed since each identified active time point. In this way, the server 1 generates distribution information regarding the distribution of the parameter after the predetermined time has elapsed since the active time point at which the amount of activity was equal to or greater than a certain value.

[0031] The reason why the parameters at a time point a predetermined time after the active time point are used instead of directly using the parameters at the active time point is that patients with chronic heart failure have reduced cardiac function, so their heart rate does not increase significantly immediately after strenuous activity, and their heart rate does not decrease easily even after time has passed.By calculating an index value based on the amount of activity at the active time point and the parameters at a time point a predetermined time after the measurement of the amount of activity, it is possible to appropriately identify the worsening of heart failure in patients with chronic heart failure.

[0032] The server 1 calculates a first index value by comparing distribution information of the parameter for a past period with distribution information of the parameter for a most recent period. For example, the server 1 calculates KL divergence (Kullback-Leibler divergence), which is a statistical quantity that quantifies the difference between the two distributions. In this way, the server 1 calculates the difference in the distribution of the above parameters as an index.

[0033] Although the KL divergence is calculated as the first index value in the above example, the present embodiment is not limited to this. For example, the server 1 may calculate a basic statistic (first statistic) such as a median or mode representing the distribution of a parameter at a time point a predetermined time after the measurement of the activity amount at the active time point as the first index value. The statistic can also be used to present, as an index, how high the parameter (pulse rate) is, i.e., the increase in the parameter due to the worsening of heart failure.

[0034] Fig. 5 is an explanatory diagram relating to the second index value. Fig. 5 shows a scatter diagram in which the amount of activity and the parameter (pulse rate) are plotted when the amount of activity is equal to or greater than a certain value.

[0035] 5 shows a scatter plot of the activity level versus the above-mentioned parameter (pulse rate) after a predetermined time (e.g., 30 minutes) has elapsed since the activity level was measured for each of the exacerbation and non-exacerbation periods. When the activity level is equal to or greater than a certain value, the regression line of the activity level and the above-mentioned parameter exhibits a steeper overall slope during the exacerbation period. Therefore, the slope of the regression line derived based on data obtained over a predetermined period (e.g., two weeks) can be said to reflect the state of the subject's heart failure during that period.

[0036] The server 1 then calculates a second index value based on the slope of the regression line. First, the server 1 calculates the above parameters and activity amounts from biological data and exercise data for a recent period (e.g., the past two weeks). For the most recent period, the server 1 derives a regression line showing the relationship between activity amounts and parameters based on a distribution (scatter plot) of activity amounts at active times when the activity amounts are equal to or greater than a certain value and parameters at times a predetermined time has elapsed since the measurement of the activity amounts. The server 1 calculates the slope of the derived regression line as the second index value.

[0037] In the above, the slope of the regression line itself is calculated as the second index value, but the present embodiment is not limited to this. For example, when calculating the distribution of activity amount and parameter, the server 1 may obtain multiple distributions by identifying parameters at multiple time points, such as 10 minutes, 30 minutes, 60 minutes, etc., after the activity amount is measured, derive regression lines from each distribution, and calculate the average value, total value, etc. of the slopes of the regression lines as the second index value. In this way, it is sufficient for the server 1 to be able to calculate the second index value based on the slope of the regression line.

[0038] Furthermore, while the second index value is calculated from only the data from the most recent period in the above example, the present embodiment is not limited to this. The second index value may be calculated by comparing the data from the most recent period with data from a previous period (e.g., six months ago). That is, the server 1 calculates the slope of a regression line by taking the distribution of the above parameters relative to the amount of activity from the biometric data and exercise data from the most recent period, and also calculates the slope of the regression line by taking the distribution of parameters relative to the amount of activity from the biometric data and exercise data from a previous period, and then calculates the difference, ratio, or the like between the two slopes as the second index value. In this way, the server 1 may calculate the second index value based on the slope from the most recent period and the slope from the previous period.

[0039] 6 and 7A and 7B are explanatory diagrams relating to the third index value. Fig. 6 shows a graph illustrating the parameter (pulse rate) and the moving average waveform of the parameter. Figs. 7A and 7B show histograms illustrating the distribution of the difference between the parameter and the moving average during an exacerbation period (Fig. 7A) and a non-exacerbation period (Fig. 7B).

[0040] As in the cases of the first and second index values, the server 1 calculates the third index value based on the distribution of the parameter at the active time point when the amount of activity is equal to or greater than a certain value. Here, when dividing the data into exacerbation periods and non-exacerbation periods, it is desirable to examine the relationship with the amount of activity, but there are cases where it is difficult to distinguish between the influence of the long-term trend of the parameter and the influence of the amount of activity (see FIG. 6 ).

[0041] Therefore, the server 1 calculates the difference between the parameter and the moving average of the parameter, thereby eliminating the influence of the long-term trend.

[0042] 7A and 7B show histograms showing the distribution of the difference values ​​during the exacerbation and non-exacerbation periods when the amount of activity is equal to or greater than a certain value. As shown in FIGS. 7A and 7B, when the amount of activity is equal to or greater than a certain value, the histogram shifts to the right during the exacerbation period, and the difference values ​​increase. Therefore, the distribution of the difference values ​​over a predetermined period (e.g., two weeks) can be said to reflect the state of the subject's heart failure during that period.

[0043] Therefore, the server 1 calculates the difference in distribution in the histogram as a third index value. First, the server 1 calculates the above parameters and activity amounts from biological data and exercise data for each of a past period (first period) and a most recent period (second period), and calculates the difference between the above parameters at each time point and the moving average of the parameters for each of the past period and the most recent period. The server 1 generates difference value distribution information regarding the distribution of difference values ​​at active times when the activity amount is equal to or greater than a certain value for each of the past period and the most recent period. The server 1 then calculates the third index value by comparing the difference value distribution information for the past period with the difference value distribution information for the most recent period. Specifically, the server 1 calculates the KL divergence, a statistical quantity that quantifies the difference in distribution, as the third index value, in the same way as when calculating the first index value.

[0044] Although the KL divergence is calculated as the third index value in the above example, the present embodiment is not limited to this. For example, the server 1 may calculate a basic statistic (second statistic) such as a median or mode representing the distribution of difference values ​​at active time points as the index value. The statistic may also be used to determine a shift in the histogram due to worsening heart failure. Furthermore, when generating the difference value distribution information, instead of calculating the difference value at each time point, active time points at which the activity amount is equal to or greater than a certain value may be identified, and only the difference values ​​at the identified active time points may be calculated.

[0045] Fig. 8 is an explanatory diagram relating to the fourth index value. Fig. 8 shows a violin plot representing the distribution of activity levels for each time period (e.g., each time period obtained by dividing a day into two hours) during a predetermined period, and a violin plot representing the distribution of activity levels and the above-mentioned parameter (pulse rate) for each time period (e.g., each time period obtained by dividing a day into two hours) during an exacerbation period and a non-exacerbation period within the predetermined period.

[0046] As shown in Figure 8, the distribution of activity levels at 4:00 p.m. is biased higher than in other time periods, making it the time period in which the subject is most active during the day. During this time period, the distribution of the parameter (pulse rate) during the exacerbation period is higher than during the non-exacerbation period. Therefore, the distribution of the parameter during this time period can be considered to reflect the state of the subject's heart failure. Therefore, the server 1 calculates a fourth index value from the distribution of activity levels and the parameter for each time period.

[0047] First, the server 1 calculates the amount of activity from exercise data for a predetermined period, such as a past period (e.g., six months ago) or a recent period (e.g., the past two weeks), and identifies the distribution of the amount of activity by time slot. From the identified distribution of the amount of activity by time slot, the server 1 identifies active time slots in which statistics (median, mode, etc.) representing the distribution of the amount of activity are equal to or greater than a certain value. Next, the server 1 calculates the above parameters from the biological data for the most recent period (e.g., the past two weeks) and identifies the distribution of the parameters during the active time slots. The server 1 then calculates statistics (e.g., variance, mode, etc.) representing the distribution of the above parameters during the active time slots as a fourth index value.

[0048] Although the fourth index value is calculated from only data from the most recent period in the above example, this embodiment is not limited to this. The fourth index value may be calculated by comparing data from a past period with data from the most recent period. Specifically, the server 1 calculates statistics representing the distribution of the parameters during an active time period (e.g., 4:00 p.m.) identified based on exercise data from a predetermined period for each of a past period (e.g., six months ago) and the most recent period. The server 1 then calculates the fourth index value by, for example, calculating the difference between the statistical values ​​of the parameters for the past period and the most recent period. Alternatively, the server 1 may calculate the fourth index value (e.g., KL divergence, a statistical value that quantifies the difference in distribution) by comparing distribution information of the parameters during the active time period for the past period with distribution information of the parameters during the active time period for the most recent period. In this way, the server 1 may calculate the fourth index value by comparing data from a past period with data from the most recent period.

[0049] Furthermore, as shown in FIG. 8 , during an active time period, such as from 8:00 to 20:00, the above parameter (pulse rate) is generally high during the exacerbation period, even though the activity level generally tends to decrease. Therefore, the distribution of the activity level and the above parameter during that time period can be said to reflect the subject's state of heart failure. Therefore, the server 1 may calculate the fourth index value by comparing the distribution of the activity level and the above parameter for each time period between data from a past period and data from a recent period. Specifically, the server 1 determines the distribution of the activity level and the parameter calculated from the biological data and exercise data for each of the past period (e.g., six months ago) and the most recent period during an active time period (e.g., from 8:00 to 20:00) based on the exercise data for a predetermined period. The server 1 then calculates the KL divergence, a statistical quantity that quantifies the difference in distribution of the activity level and the parameter during the active time period between the past period and the most recent period, and sums these values ​​to obtain the fourth index value. The distributions of activity amounts and parameters for the past and most recent periods may represent distributions for the entire active time period (e.g., 12 hours), and the KL divergence may be calculated using these distributions. Alternatively, the distributions of activity amounts and parameters for the past and most recent periods may represent distributions obtained by dividing the active time period into predetermined time intervals (e.g., 2 hours), as shown in FIG. 8 , and the KL divergence may be calculated for each time interval and then summed to calculate the fourth index value. In this specific example, the "constant value of a statistic representing the distribution of activity amounts (e.g., median, mode, etc.)" used to identify the active time period may be a value lower than that in the previously described specific example and greater than a statistic representing the distribution of activity amounts during bedtime (e.g., from 10:00 PM to 6:00 AM).

[0050] 9 to 11 are explanatory diagrams relating to the fifth index value. FIG. 9 illustrates how the original data (Observed) for each of the above parameters (pulse rate) and activity amounts are separated into trend components (Trend), periodic components (Seasonal), and residual components (Residual), and how the periodic components for the same section are extracted to generate histograms. FIG. 10 illustrates how a histogram for a past period (shown hatched) is compared with a histogram for a most recent period (shown in white). FIG. 11 illustrates how, similar to FIG. 9, the original data for each of the above parameters and activity amounts are separated into trend components, periodic components, and residual components, and how the periodic components for the same section are compared to calculate a correlation coefficient.

[0051] The server 1 calculates a fifth index value by comparing the periodic components of the parameters and activity amounts for a past period (first period) and a most recent period (second period). First, the server 1 calculates the parameters and activity amounts from the biological data and exercise data for each of the past period and the most recent period. As shown in FIG. 9 , the server 1 separates the parameters and activity amounts (first graph from the top) for each of the past period and the most recent period into a trend component (second graph from the top), a periodic component (third graph from the top), and a residual component (fourth graph from the top) using, for example, the STL (Seasonal-Trend Decomposition Procedure Based on LOESS) method.

[0052] As shown in FIG. 9 , the server 1 extracts periodic components for the same interval from the periodic components of the parameters and activity amounts for each of the past and most recent periods. The extracted time-series periodic component data is converted into a histogram representing the distribution of the data. If the periodic components are daily, the same interval refers to, for example, an interval for one day (24 hours). In this way, the same interval may be the same as the period used when separating periodic components from the original data. Alternatively, the same interval may be, for example, an interval for half a day (12 hours) from 6:00 a.m. to 6:00 p.m.

[0053] 10 , the server 1 calculates the fifth index value by comparing the periodic components of the extracted parameters and activity amounts between a past period and a recent period. For example, the server 1 calculates KL divergence, which is a statistic that quantifies the difference in distribution between the distribution (histogram) of the periodic components of the parameters and the distribution of the periodic components of the activity amounts between the past period and the recent period, and then sums these to obtain the fifth index value. Alternatively, the server 1 may calculate the difference between a basic statistic (such as a median) representing the distribution of the periodic components of the parameters and a basic statistic representing the distribution of the periodic components of the activity amounts as the fifth index value.

[0054] Alternatively, the server 1 may calculate the fifth index value by directly comparing the time series data of the periodic component of the parameter with the time series data of the periodic component of the activity amount, as shown in FIG. 11 . Specifically, the server 1 separates the parameter and the activity amount into a trend component, a periodic component, and a residual component, and then extracts the periodic components of the parameter and the activity amount at each time point included in a predetermined time span. The server 1 then calculates, as the fifth index value, a correlation coefficient of the cross-correlation between the periodic component of the parameter and the periodic component of the activity amount based on the extracted periodic components of the parameter and the activity amount. Note that if the periodic component is a daily cycle, the predetermined time span refers to, for example, one day (24 hours). The predetermined time span may be shorter than one day, such as half a day (12 hours) from 6:00 a.m. to 6:00 p.m., or may be two or more days.

[0055] As described above, the server 1 calculates the first to fifth index values ​​as index values ​​related to the worsening of heart failure. The server 1 may calculate all of the first to fifth index values, or may calculate only some of the index values. This makes it possible to provide indexes related to the worsening of heart failure to medical professionals and the like.

[0056] Fig. 12 is a flowchart showing an example of a processing procedure executed by the server 1. The processing executed by the server 1 will be described with reference to Fig. 12. The control unit 11 of the server 1 acquires biological data and exercise data of the subject at each point in time during a predetermined period from the wearable device 2 (step S11). The biological data is, for example, a PPG signal. The exercise data is, for example, acceleration.

[0057] The control unit 11 calculates a parameter related to the subject's heart at each time point based on the acquired exercise data (step S12). For example, the control unit 11 calculates the pulse rate as the parameter. The control unit 11 also calculates the subject's activity amount at each time point based on the acquired exercise data (step S13). For example, the control unit 11 counts the number of times the value calculated by Equation (1) exceeds a predetermined value within a predetermined time period as activity intensity over time, and calculates the moving average of the activity intensity obtained as time-series data over a predetermined time window width as the activity amount.

[0058] The control unit 11 calculates an index value related to the worsening of heart failure based on the calculated parameters and the amount of activity (step S14). Specifically, like the first to fourth index values ​​described above, the control unit 11 calculates the index value related to the worsening of heart failure based on the correlation between the parameters and the time points or time periods when the value related to the amount of activity is equal to or greater than a certain value. In this case, like the first to third index values ​​described above, the control unit 11 calculates the index value related to the worsening of heart failure based on the distribution of parameters with respect to the amount of activity at active times when the amount of activity is equal to or greater than a certain value. Furthermore, like the first and second index values, the control unit 11 calculates the index value based on the amount of activity at active times and the parameters at a time point a predetermined time after the measurement of the amount of activity.

[0059] For example, the control unit 11 calculates the first index value by comparing distribution information regarding the distribution of parameters at a time point a predetermined time has elapsed since the measurement of the activity amount at a time point when the user was active in a past period (first period) with distribution information regarding the distribution of parameters at a time point a predetermined time has elapsed since the measurement of the activity amount at a time point when the user was active in a most recent period (second period). Alternatively, the control unit 11 may calculate, as the first index value, a statistic (first statistic) representing the distribution of parameters at a time point a predetermined time has elapsed since the measurement of the activity amount at a time point when the user was active.

[0060] Furthermore, for example, the control unit 11 derives a regression line showing the relationship between the amount of activity at the time of activity for the most recent period and the parameter at a time a predetermined time has elapsed since the time the amount of activity was measured, and calculates a second index value based on the slope of the regression line.

[0061] Furthermore, for example, the control unit 11 calculates the difference between the parameter at each time point and the moving average of the parameter for each past period (first period) and most recent period (second period), and compares difference value distribution information regarding the distribution of difference values ​​at active time points in the past period with difference value distribution information regarding the distribution of difference values ​​at active time points in the most recent period to calculate the third index value. Alternatively, the control unit 11 may calculate a statistic (second statistic) representing the distribution of difference values ​​at active time points as the third index value.

[0062] Furthermore, for example, the control unit 11 identifies the distribution of the above activity levels for each time period, and from the distribution of each identified time period, identifies time periods in which a statistical quantity representing the distribution of the activity levels (e.g., a median value) is equal to or greater than a certain value, and calculates a statistical quantity regarding the distribution of the above parameters in the identified time period as a fourth index value.

[0063] Furthermore, for example, the control unit 11 may separate each of the parameters and activity amounts into trend components, periodic components, and residual components for a past period (first period) and a most recent period (second period), extract periodic components for the same interval from each of the periodic components of the parameters and activity amounts, and compare the extracted periodic components of each of the parameters and activity amounts between the past period and the most recent period to calculate the fifth index value. Alternatively, the control unit 11 may separate each of the parameters and activity amounts into trend components, periodic components, and residual components, extract periodic components of each of the parameters and activity amounts at each time point included in a predetermined time width, and calculate the correlation coefficient of the cross-correlation between the extracted periodic components of the parameters and the periodic components of the activity amounts as the fifth index value.

[0064] The control unit 11 may calculate all of the first to fifth index values, or may calculate only some of the index values, and then the control unit 11 ends the series of processes.

[0065] In the above description, the server 1 on the cloud analyzes the biometric data and exercise data to calculate the index value, but the index value may also be calculated by the local wearable device 2.

[0066] As described above, according to this embodiment, an index value relating to the worsening of heart failure can be calculated and provided to medical professionals and the like.

[0067] (Modification) In the above embodiment, the index value is calculated only from the subject's own data. However, the present embodiment is not limited to this, and the index value may be calculated by comparing the subject's data with data of a third party other than the subject (hereinafter referred to as a "reference subject").

[0068] For example, in the above-described embodiment, the first, third, and fifth index values ​​are calculated by comparing data from a past period (first period) with data from a most recent period (second period). In this modification, data from a reference subject other than the subject is used as the data from the past period.

[0069] That is, the server 1 calculates the above parameters (pulse rate) and activity amount from the subject's biological data and exercise data for the most recent period, and also acquires reference information including activity amounts (hereinafter referred to as "reference activity amounts") and heart-related parameters (hereinafter referred to as "reference parameters") of one or more reference subjects other than the subject at each time point (hereinafter referred to as "reference time points") in a past period (hereinafter referred to as "reference period"). For example, the server 1 stores reference information constructed based on the parameters and activity amounts of persons other than the subject in the auxiliary storage unit 14, and reads the reference information from the auxiliary storage unit 14.

[0070] The server 1 calculates an index value related to the worsening of heart failure by comparing target information based on the subject's parameters and activity amount for the most recent period with the reference information. For example, when calculating a first index value as an index value, the server 1 calculates the first index value by comparing parameter distribution information at a time point a predetermined time has elapsed since the measurement of the subject's activity amount at an active time point in the most recent period with reference parameter distribution information at a time point a predetermined time has elapsed since the measurement of the reference activity amount at an active time point in a past period of the reference subject. Similarly, when calculating the third and fifth index values, the server 1 calculates the index value by comparing the subject's target information with the reference information of the reference subject.

[0071] FIG. 13 is a flowchart showing an example of a processing procedure executed by the server 1 according to the modified example. After calculating the subject's heart-related parameters (step S12) and the subject's activity level (step S13), the server 1 executes the following processing. The control unit 11 of the server 1 acquires reference information based on the reference activity level and the heart-related reference parameters of a reference subject other than the subject at each reference time point during the reference period (step S201). The control unit 11 calculates an index value related to the worsening of heart failure based on the comparison result between the subject information based on the subject's parameters and activity level for the most recent period and the reference information acquired in step S201 (step S202). For example, the control unit 11 calculates the first, third, and / or fifth index values ​​using the reference information of the reference subject as data for the past period. The control unit 11 then terminates the series of processing.

[0072] The auxiliary storage unit 14 may store parameters and activity amounts of a reference subject other than the subject as reference data, and the server 1 may acquire the reference information by having the control unit 11 of the server 1 read the reference data from the auxiliary storage unit 14 and construct reference information based on the read reference data in step S201. Specifically, when calculating a first index value as an index value, for example, in step S201, the control unit 11 of the server 1 constructs distribution information of the reference parameters from the reference data read from the auxiliary storage unit 14, and in step S202, calculates the first index value from a result of comparing the distribution information based on the subject's parameters and activity amounts for the most recent period with the constructed distribution information of the reference subject.

[0073] As described above, according to this modification, it is also possible to calculate an index value using data of a reference subject other than the subject.

[0074] The embodiments disclosed herein are illustrative in all respects and should not be considered 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 meaning and scope of the claims.

[0075] The matters described in each embodiment can be combined with each other. Furthermore, the independent claims and dependent claims described in the claims can be combined with each other in any combination, regardless of the reference format. Furthermore, the claims use a format in which a claim references two or more other claims (multiple claim format), but this is not limited to this. A multiple claim (multi-multi claim) that references at least one other multiple claim may also be used.

[0076] REFERENCE SIGNS LIST 1 Server 11 Control unit 12 Main memory unit 13 Communication unit 14 Auxiliary memory unit 2 Wearable device 21 Control unit 22 Memory unit 23 Communication unit 24 Light emitting unit 25 Light receiving unit 26 Acceleration detection unit

Claims

1. A program that causes a computer to execute the following processes: acquire biometric data and exercise data of a subject at each time point during a specified period; calculate heart-related parameters at each time point based on the biometric data; calculate the subject's activity level at each time point based on the exercise data; and calculate an index value related to the worsening of heart failure based on the parameters and activity level at each time point.

2. The program according to claim 1, further comprising: calculating an index value relating to the worsening of heart failure based on the correlation between a time point or a time period when the value relating to the amount of activity is equal to or greater than a certain value and the parameter.

3. The program according to claim 2, further comprising: calculating an index value relating to the worsening of heart failure based on a distribution of the parameter for the amount of activity at an active time point where the amount of activity is equal to or greater than a certain value among the time points.

4. The program according to claim 3, further comprising: calculating the index value based on the amount of activity at the active time point and the parameter at a time point a predetermined time after the amount of activity was measured.

5. The program of claim 4, which acquires the biological data and exercise data for each of a first period and a second period to calculate the parameters and activity amount, and calculates the index value by comparing distribution information regarding the distribution of the parameters at a time point a predetermined time has elapsed since the measurement of the activity amount at the active time point in the first period with distribution information regarding the distribution of the parameters at a time point a predetermined time has elapsed since the measurement of the activity amount at the active time point in the second period.

6. The program according to claim 4, further comprising: calculating, as an index value, a first statistical quantity that represents a distribution of the parameter at a time point a predetermined time has elapsed since the measurement of the amount of activity at the active time point.

7. The program according to claim 4, further comprising: acquiring the biometric data and exercise data of the subject for a most recent period to calculate the parameters and activity amount; deriving a regression line showing the relationship between the activity amount at the active time for the most recent period and the parameters at a predetermined time after the activity amount was measured; and calculating the index value based on the slope of the regression line.

8. The program of claim 3, which acquires the biological data and exercise data for each of a first period and a second period to calculate the parameters and activity amount, calculates a difference value between the parameter at each time point and a moving average of the parameter for each of the first period and the second period, and calculates the index value by comparing difference value distribution information regarding the distribution of the difference values ​​at the active time points in the first period with difference value distribution information regarding the distribution of the difference values ​​at the active time points in the second period.

9. The program according to claim 3, further comprising: calculating a difference value between the parameter at each of the time points and a moving average of the parameter; and calculating, as the index value, a second statistic that represents a distribution of the difference values ​​at the active time points.

10. The program according to claim 2, further comprising: identifying a time period in which a statistical quantity representing the distribution of the amount of activity is equal to or greater than a certain value from the distribution of the amount of activity for each time period; and calculating, as the index value, a statistical quantity relating to the distribution of the parameter in the identified time period.

11. The program according to claim 1, which acquires the biological data and movement data for a first period and a second period, respectively, to calculate the parameters and activity amounts, separates the parameters and activity amounts for the first period and the second period, respectively, into trend components, periodic components, and residual components, extracts periodic components for the same interval from the periodic components of the parameters and activity amounts for the first period and the second period, respectively, and calculates the index value by comparing the periodic components of the parameters and activity amounts between the first period and the second period.

12. The program according to claim 1, further comprising: separating each of the parameters and activity amounts into a trend component, a periodic component, and a residual component; extracting the periodic components of each of the parameters and activity amounts at each of the time points included in a specified time span; and calculating, as the index value, a correlation coefficient in the cross-correlation between the periodic component of the parameter and the periodic component of the activity amount based on the extracted periodic components of each of the parameters and activity amounts.

13. The program according to claim 1, further comprising: acquiring the biological data and exercise data of the subject for a most recent period to calculate the parameters and activity amount; acquiring reference information based on a reference activity amount and a reference parameter related to the heart at each reference time point in the reference period of a reference subject other than the subject; and calculating an index value from the result of comparing the subject information based on the parameters and activity amount for the most recent period with the reference information.

14. An information processing method in which a computer executes the following processes: acquiring biometric data and exercise data of a subject at each time point during a specified period; calculating heart-related parameters at each time point based on the biometric data; calculating the subject's activity level at each time point based on the exercise data; and calculating an index value related to the worsening of heart failure based on the parameters and activity level at each time point.

15. An information processing device having a control unit, wherein the control unit acquires biometric data and exercise data of a subject at each time point during a predetermined period, calculates heart-related parameters at each time point based on the biometric data, calculates the subject's activity level at each time point based on the exercise data, and calculates an index value related to the worsening of heart failure based on the parameters and activity level at each time point.

Images