Heart rate data analysis device and program

Abstract

To provide a heart rate data analysis device capable of estimating an autonomic nerve activity amount from heart rate data of a subject in consideration of an environmental change.SOLUTION: A heart rate data analysis device performs first processing (21) of reading an RR interval calculated from heart rate data of a subject for each window, and generating an intermediate function whose value is determined for each one beating of the heart, second processing (22, 23, 24) of calculating a direct current component energy as a change amount from a state (11, 12) of a unique heart rate or a heart rate at the time of resting, and an alternate current component energy from the intermediate function, and third processing (25) of estimating an autonomic nerve activity amount of the subject corresponding to the window as a sum of the direct current component energy and the alternate current component energy.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a heartbeat data analysis device and a program. [Background technology]

[0002] As an example of a conventional technique for evaluating the autonomic nervous system, Patent Document 1 uses the amount of autonomic nervous activity (ccvTP) as an index for objectively evaluating fatigue, calculated by correcting the sum of the LF value and the HF value by the heart rate during a time period. Specifically, the square root of the total power (TP) is divided by the average heart rate during a time period (window), "average (RR)," to calculate ccvTP as shown in formula (PR1). TP is calculated as the sum of the low-frequency component LF and the high-frequency component HF, "TP = LF + HF." Formulas (PR2) to (PR4) are used for the low-frequency and high-frequency components LF and HF. In formula (PR4), C(t) is the autocorrelation function of the RR interval.

[0003]

number

[0004] Here, the conventional technology is based on the idea of ​​evaluating heart rate variability from a nearly constant average heart rate. Here, as cited below from Non-Patent Document 1, it is medically known that heart rate variability reflects the activity of the sympathetic and parasympathetic nervous systems, and the conventional technology makes use of this knowledge to perform evaluation.

[0005] "When the innervation of the heart is blocked, the heart continues to beat at the intrinsic firing frequency of the sinoatrial node cells. This constant frequency is called the intrinsic heart rate. The fluctuation component of the heartbeat interval relative to the average heart rate, which fluctuates due to the action of the sympathetic and parasympathetic nervous systems, is called heart rate variability." (Quoted from Non-Patent Document 1) [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Publication No. 2018-149262 [Non-patent literature]

[0007] [Non-Patent Document 1] Yamamoto, Y. Information theory of heart rate: Toward field physiology. Micromechatronics, 1999, 43.4: 9-17. Summary of the Invention [Problem to be solved by the invention]

[0008] However, conventional techniques have not been able to perform real-time evaluation in environments where the premise of a "constant average heart rate" does not hold.

[0009] For example, as a subject goes through the following process (1) before entering the sauna → (2) while in the sauna → (3) after leaving the sauna, the environment the subject is placed in changes drastically, such as (1) a room temperature 25°C environment outside the sauna → (2) a high temperature 100°C environment inside the sauna → (3) a room temperature 25°C environment outside the sauna. When considering real-time evaluation of the subject's autonomic nervous activity as it goes through such a process, in order to analyze the subject's heart rate data and appropriately evaluate the autonomic nervous activity, it is necessary to analyze the heart rate data while taking into account the effect of changes in average heart rate due to environmental changes, but this measure has not been taken in conventional technology.

[0010] For example, if the ambient temperature rises from 25°C to 100°C, the average heart rate will rise. However, conventional technology evaluates autonomic nervous activity based only on the fluctuation component of the heartbeat interval without taking into account the increase in the average heart rate itself, resulting in an overly small amount of activity.

[0011] That is, the processing system for quantitatively assessing fatigue level in Patent Document 1 assumes that the environment in which the heartbeat interval is acquired is constant, such as when the subject is at rest or in a sitting position. According to paragraph

[0028] , the amount of autonomic nervous activity is compared with a database to determine the deviation. Therefore, this system is not suitable for determining the "change" in this amount in real time. Furthermore, [Equation 3] (the aforementioned equation (PR-4)) performs a Fourier transform of the autocorrelation function C(t). When the observation region (window) is short, the side lobe effect becomes significant. Since a relatively long window is required to reduce the side lobe effect, delay times tend to increase, making real-time evaluation difficult.

[0012] In view of the above-mentioned problems with the conventional technology, an object of the present invention is to provide a heart rate data analysis device and program that can estimate autonomic nervous activity from a subject's heart rate data while also taking environmental changes into consideration. [Means for solving the problem]

[0013] In order to achieve the above object, the present invention is a heartbeat data analysis device or program that executes the following steps: a first process of reading, for each window, an RR interval calculated from the heartbeat data of a subject, and generating an intermediate function whose value is determined for each heartbeat; a second process of calculating, from the intermediate function, a DC component energy and an AC component energy as changes from the intrinsic heart rate or the resting heart rate; and a third process of estimating the autonomic nervous activity of the subject corresponding to the window as the sum of the DC component energy and the AC component energy. [Effects of the Invention]

[0014] According to the present invention, an intermediate function whose value is determined for each heartbeat is analyzed, and the DC component energy is calculated as the change from the intrinsic heart rate or the resting heart rate, and then the autonomic nervous activity is estimated. Therefore, the autonomic nervous activity can be estimated from the subject's heart rate data while taking environmental changes into consideration. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a functional block diagram of a heartbeat data analysis device according to an embodiment. [Figure 2] 10 is a flowchart illustrating an operation of a heartbeat data analysis device according to an embodiment. [Figure 3] FIG. 10 is a diagram showing a schematic graph example of an intermediate function as an example for explaining the present embodiment. [Figure 4] 10A and 10B are diagrams illustrating an example of setting timing and window width when processing in real time. [Figure 5] FIG. 10 is a diagram listing examples of other available intermediate functions. [Figure 6] FIG. 1 is a diagram illustrating a hardware configuration of a typical computer. DETAILED DESCRIPTION OF THE INVENTION

[0016] 1 is a functional block diagram of a heartbeat data analysis device 30 according to one embodiment. The heartbeat data analysis device 30 includes an autonomic nerve blocking state estimation unit 10 including a resting mean interval estimation unit 11 and an intrinsic heart rate estimation unit 12, and an autonomic nerve active state estimation unit 20 including an intermediate function generation unit 21, a weighted average calculation unit 22, a DC energy calculation unit 23, an AC energy calculation unit 24, and an autonomic nerve activity amount estimation unit 25.

[0017] 2 is a flowchart of the operation of a heartbeat data analysis device 30 according to one embodiment. Step S1 is a step for preparing in advance information required for performing real-time analysis processing in the next step S2. Step S1 is executed by the autonomic nerve blocking estimation unit 10 to acquire resting heartbeat data (RR intervals) of a subject who is in a resting state in a fixed environment, and estimate the subject's intrinsic heart rate. Step S2 is executed by the autonomic nerve activity estimation unit 20, and by utilizing the information on the intrinsic heart rate estimated in step S1, acquires in real time heartbeat data (RR intervals) of the subject who is in a changeable environment, for example, in the form of the subject entering and leaving a sauna, where the ambient temperature changes drastically, and estimates the subject's autonomic nerve activity in real time.

[0018] The following describes in detail the processing of the units 11 and 12 of the autonomic nerve blocking state estimating unit 10 that executes step S1, and the units 21 to 25 of the autonomic nerve active state estimating unit 20 that executes step S2.

[0019] <Resting average interval estimation unit 11> The resting mean interval estimator 11 acquires heart rate data over a certain period of time from a subject in a stable environment who is in a resting state and is not stressed by either external factors due to the environment or internal factors due to the subject's own psychological state, calculates the mean value of the R-R intervals (referred to as RestRRI), and outputs the calculated mean value to the intrinsic heart rate estimator 12. As is known in the art, heart rate data can be acquired by, for example, attaching an electrocardiogram sensor to the subject to obtain a time series of electrocardiogram data, and acquiring the R-R intervals as the intervals between R waves in the form of a total of N pieces of sequence data (a sequence {RRIi|i=1, 2, 3, ..., N} consisting of values ​​RRIi of the i-th R-R intervals) in which the R-R intervals are arranged in the order of the time series of the electrocardiogram data, and then calculating the RestRRI as the mean of these N (resting) R-R intervals.

[0020] <Intrinsic heart rate estimation unit 12> The intrinsic heart rate estimation unit 12 estimates the intrinsic heart rate interval (referred to as UniqueRRI) by multiplying RestRRI by a constant k according to the following equation (12). Note that the constant is set to, for example, k=1.4, and model parameter values ​​for this estimation are prepared in advance. UniqueRRI = RestRRI × k … (12)

[0021] This intrinsic heart rate interval is the RR interval when the heart is at the intrinsic heart rate in the above-cited Non-Patent Document 1, and since it is practically impossible to "block the autonomic nerve control over the heart," in this embodiment, the value that would be taken if the control were blocked is estimated from RestRRI, which is data from the subject at rest.

[0022] As described above, the autonomic nerve blocking state estimation unit 10 calculates the UniqueRRI as an estimated value through the processing of the respective units 10 and 11, and outputs the calculated value to the autonomic nerve active state estimation unit 20, which performs analysis processing in real time. The respective units 21 to 25 of the autonomic nerve active state estimation unit 20 perform the following processing in this order.

[0023] <Intermediate function generation unit 21> The intermediate function generating unit 21 receives as input in real time RR intervals obtained from an electrocardiogram data time series acquired in real time as heartbeat data from a subject placed in a changeable environment, generates intermediate functions within a window W of a predetermined width, and outputs these intermediate functions to the weighted average calculating unit 22 and subsequent units 22 to 24. Here, the sensors and signal processing used to obtain RR intervals from the heartbeat data of the subject may be any existing method, similar to those described for the resting mean interval estimating unit 11.

[0024] 3 is a schematic graph of an intermediate function within a window W to explain the intermediate function generated from the RR intervals by the intermediate function generation unit 21. Because the RR intervals are simply a sequence along the time axis, they are converted into intermediate functions, which are time-series data, in order to perform frequency analysis (however, in this embodiment, direct frequency analysis calculations are not required, as will be described later). Many types of intermediate functions have been proposed depending on the purpose of analyzing heartbeat data, and examples of intermediate functions include intermediate functions that associate reciprocals P1=1 / RRI1, P2=1 / RRI2, P3=1 / RRI3, ... as constant values ​​with the RR intervals RRI1, RRI2, RRI3, ... (widths that appear sequentially along the time axis).

[0025] In this embodiment, the value of the i-th RR interval is expressed as RRI i Assuming that time t is RRI i The value of the intermediate function P when it is within i By defining P(t) as in the following equation (21a), an intermediate function is generated within the window W so that the value P(t) at time t is as in the following equation (21b), as shown schematically in FIG. 3. (That is, when time t is the i-th RR interval RRI i If the time range is within P(t), then P(t)=P i The intermediate function P(t) is defined as the value of the function (t). Note that in equation (21b), if i=1, this means that P(t)=P1 for 0≦t≦RRI1.

[0026]

number

[0027] In equation (21a), Co is a coefficient used under the assumption that the energy per beat changes. This coefficient is determined by the cardiac stroke volume and vascular resistance. If it is assumed that there is no change in stroke volume or vascular resistance, Co=1 should be used. The following explanation will be given assuming that the size of the window W (window width) is set to 60 seconds, as shown in the schematic example of Figure 3.

[0028] In this embodiment, the intermediate functions defined by the reciprocal of the square root of the RR interval, as defined by equations (21a) and (21b), are similar to the method disclosed in Japanese Patent No. 7221195 by the present applicant, and can achieve the following effects as described in paragraph

[0020] of the same document. "The time domain representation of the time series data is optimized so that the sum of the power spectral densities per heartbeat is equal across the entire time series data. This makes it possible to equalize the sum of the power spectral densities for each interval of a given number of heartbeats in the time series data, and to calculate each power spectral density under this condition. Therefore, simply by comparing these power spectral densities, it becomes possible to correctly evaluate changes over time in the activity of the autonomic nervous system."

[0029] The above effect is due to the fact that the integral of the square of the value (amplitude) of equation (21b), which is the energy per beat, is a constant value Co, as shown in the following equation (21c).

[0030]

number

[0031] In addition, each of the sections 21 to 25 of the autonomic nerve activity estimation section 20 sets a window for the RR interval input in real time in this order, processes the intermediate function generated by the intermediate function generation section 21 for that window, and can calculate the amount of autonomic nerve activity within that window in real time.

[0032] The timing and window width for processing the RR intervals input in real time in this way may be set using various predetermined settings, as shown in the schematic examples of Fig. 4. That is, as shown in example EX1 of Fig. 4, at times t1, t2, t3, ..., which are sequential real-time processing timings, windows W1, W2, W3, ... may be read up to the respective times, and the windows may be arranged to be adjacent (the end time of one window coincides with the start time of the next window). Also, as shown in example EX2, at times t4, t5, t6, ..., which are sequential real-time processing timings, windows W4, W5, W6, ... may be read up to the respective times, and the windows may be arranged to be separated (there is an interval between the end time of one window and the start time of the next window). Also, as shown in example EX3, at times t7, t8, t9, ..., which are sequential real-time processing timings, windows W7, W8, W9, ... may be read up to the respective times, and the windows may be arranged to be partially overlapping (the end part of one window overlaps with the start part of the next window). In Example EX3, the overlapping ratio between adjacent windows is about 20% of the window width, but this overlapping ratio can also be set to any value. (For example, the overlapping ratio can be 90% of the window width, and each window can overlap with windows further away than the two adjacent windows on either side.)

[0033] <Weighted average calculation unit 22> As a preprocessing step to realize AC-DC separation of the intermediate function in each of the subsequent units 23 and 24, a weighted average DCmean of the intermediate function within the window W (the average within the window W of the value of the intermediate function multiplied by the duration of the value as a weight) is calculated as shown in the following equation (22), and this is output to each of the subsequent units 23 and 24.

[0034]

number

[0035] Note that Equation (22) represents the value of the time-length weighted average for the intermediate function illustrated in FIG. 3 (i.e., Co = 1 in Equation (21a)) and the case of the window W position (i.e., the range of time t [seconds] is 0 ≤ t ≤ 60, and within this range, all of the first to nth RR intervals, RRI1 to RRIn, and a part of the (n + 1)th RR interval, RRIn+1, are included). In the right side, the first term of the numerator is the area of the time range from RRI1 to RRIn of the intermediate function (the sum over k = 1, 2, …, n of the value “P k = 1 / √RRI k ” multiplied by the time length RRI k to obtain the value “√RRI k ”), and the second term of the numerator is the area of the portion of the time range of the last RRIn+1 of the intermediate function that falls within the window range (0 ≤ t ≤ 60). The 60 in the denominator is the window width for obtaining the average of the said area. Similarly, for a general case where the window width, window position, etc. are different from the individual specific cases in FIG. 3, the weighted average DCmean can be obtained as the time average within the window of the intermediate function, and with the window width as W, it is as shown in the following Equation (22b). In FIG. 3, the position of the obtained weighted average DCmean (the vertical axis position on the intermediate function graph) is also schematically shown.

[0036]

Equation

[0037] <DC Energy Calculation Unit 23> The DC Energy Calculation Unit 23 calculates the DC component energy DCcomponent of the intermediate function from the following equation using the weighted average DCmean and the unique heart rate interval UniqueRRI, and outputs it to the subsequent units 24 and 25.

[0038]

Equation

[0039] In this embodiment, according to this formula (23), from the weighted average energy, the energy at the unique R - R interval (UniqueRRI) of the heart rate (since it is a constant R - R interval, the corresponding intermediate function has only a DC component and no AC component, so this energy is also composed of only the DC component), "Co / UniqueRRI" (this value is the square of "√(Co / UniqueRRI)" and becomes the energy possessed by the intrinsic heart rate) is subtracted, and the state of the intrinsic heart rate is regarded as zero (the origin), and the DC component energy can be calculated. In each case of the magnitude relationship between the two terms inside the absolute value of formula (23), they correspond to the following states respectively. ● When the square of DCmean > Co / UniqueRRI The average heart rate is higher than the intrinsic heart rate. That is, the sympathetic nerve is working. ● When the square of DCmean < Co / UniqueRRI The average heart rate is lower than the intrinsic heart rate. That is, the parasympathetic nerve is working. ● When the square of DCmean = Co / UniqueRRI Neither the sympathetic nerve nor the parasympathetic nerve is working. That is, it is equivalent to the case where the autonomic nerve is blocked.

[0040] <AC energy calculation unit 24> The AC energy calculation unit 24 calculates the AC component energy ACcomponent of the intermediate function from the following formula (24) using the weighted average DCmean, and outputs it to the autonomic nerve activity amount estimation unit 25.

[0041]

Equation

[0042] The first line of equation (24) is the Parseval equation, which shows the relationship between an arbitrary AC time function f(t) and its Fourier transform F(f), and indicates that the sum of the squares of the time function f(t) is equal to the sum of the squares of the Fourier transform F(f). Furthermore, the second line of equation (24) substitutes the intermediate function within the window into this arbitrary AC function f(t). In this embodiment, by substituting the intermediate function within the window into the arbitrary AC function f(t), the AC component energy can be calculated using only time-domain calculations for the intermediate function, which is a time function, based on the relationship of the Parseval equation, without performing a Fourier transform (conversion to the frequency domain). Note that, like equation (22), the second line of equation (24) also shows a specific equation for the example in FIG. 3, but similar calculations can be made for general cases where the window position, etc., differs from that in FIG. 3. That is, when the window width W is general, equation (24) becomes the following equation (24-b). (Note that in equation (24), DCmean is subtracted from each term, but in equation (24b), the same subtraction is performed collectively by squaring it.)

[0043]

number

[0044] <Autonomic nervous activity estimation unit 25> The autonomic nerve activity amount estimating unit 25 estimates the autonomic nerve activity amount Total power in real time as the sum of the DC component energy and the AC component energy, as shown in the following equation (25). Totalpower=ACcomponent+DCcomponent

[0045] This autonomic nervous activity total power is a numerical representation of the energy generated when the autonomic nervous system is activated, and primarily represents the level of physical stress a person is under. For example, it has been confirmed that the following trends occur when the temperature environment changes:

[0046] ●High temperature: Heart rate increases significantly, blood pressure rises slightly, the autonomic nervous system is activated, and the body is under stress (sympathetic nervous system is dominant). In low temperatures: Heart rate does not change much, but blood pressure rises significantly, the autonomic nervous system is activated, and the body is under stress (a state in which the sympathetic nervous system is dominant). At room temperature: If you are relaxed, the stress on your body will be reduced. In this case, the parasympathetic nervous system will be dominant.

[0047] As described above, according to the embodiment of the present invention, the following effects can be achieved. Quantitative evaluation is possible even when subjects are placed in different environments and the average heart rates are significantly different. This is because it is expressed in beat-based energy, i.e., it uses the intermediate function of equations (21a) and (21b), which consider one heart beat as energy, to add up the energy equivalent to the difference in average heart rates. The autonomic nervous activity estimation unit 20 performs calculations in the order indicated by the reference symbols of its respective units 21 to 25, resulting in linear processing. Therefore, the results of calculations performed with a time shift of one beat at a time can be quantitatively compared. This allows real-time processing and the time change in autonomic nervous activity can be determined. ●Since calculations are performed in the time domain, the window size can be made smaller, which shortens the delay time and ensures real-time performance. ●Since the energy of the intrinsic heart rate is set to 0, the amount of autonomic nervous activity increases even when the parasympathetic nervous system is dominant. This is a result based on the reality of the human body.

[0048] Various supplementary, additional, and alternative examples will be described below.

[0049] (1) According to an embodiment of the present invention, it is possible to grasp autonomic nervous system indices in real time with high accuracy. For example, by using this information to prevent illnesses caused by mental stress, it is possible to contribute to Goal 3 of the United Nations-led Sustainable Development Goals (SDGs), which is to "Ensure healthy lives and promote well-being for all at all ages." (Note that if the arrhythmia that is a cause of missing data is not accidental, separate medical treatment is recommended.)

[0050] (2) In equation (12), the intrinsic heart rate interval UniqueRRI is estimated from the resting RR interval RestRRI using the coefficient k, but k may be set to 1. In other words, by using the resting RR interval RestRRI instead of the intrinsic heart rate interval UniqueRRI in equation (23), the state at rest, rather than the state at autonomic nerve block, may be used as the state regarded as zero (origin) when calculating the DC component energy.

[0051] (3) Although the intermediate functions defined by the formulas (21a) and (21b) have been described, any other intermediate functions whose values ​​are determined for each heartbeat with a time interval (horizontal axis) and / or a value (vertical axis) that matches the RR interval of the heartbeat data may be used. Figure 5 shows examples of other usable intermediate functions. (In the figure, I k is the kth RR interval RRI k ) Example (a) is an intermediate function with a unit impulse, and the interval I k The value (unit value: 1) is determined for each beat of the heart. Example (b) is an intermediate function based on a beat function, where the time interval is equal to Δk and the value I is determined for each interval Δk. k Example (c) is an intermediate function by spline interpolation, with interval I k A value of I appears for each heartbeat. k The points are given and then spline interpolated. Example (d) is an intermediate function using linear interpolation, which is the same as example (c) except that linear interpolation is used instead of spline interpolation. Example (e) is an intermediate function using 1 / f fluctuation, with interval I k Each heartbeat has a value of 1 / I k appear consecutively. Example (f) is as defined by equations (21a) and (21b), and is shown again in Figure 12 to allow comparison with the other examples.

[0052] (4) FIG. 6 is a diagram showing an example of the hardware configuration of a general computer device 70. The heartbeat data analysis device 30 can be realized as one or more computers 70 having such a configuration. When the heartbeat data analysis device 30 is realized using two or more computers 70, information required for processing may be transmitted and received via a network. The computer device 70 includes a CPU (Central Processing Unit) 71 that executes predetermined instructions, a GPU (Graphics Processing Unit) 72 as a dedicated processor that executes some or all of the CPU 71's execution instructions in place of or in cooperation with the CPU 71, a RAM 73 as a main storage device that provides a work area for the CPU 71 (and GPU 72), a ROM 74 as an auxiliary storage device (e.g., an SSD that can be read and written as flash ROM, but may also be an HDD), a communication interface 75, a display 76, an input interface 77 that accepts user input via a mouse, keyboard, touch panel, etc., a heartbeat sensor 78, and a bus BS for transmitting and receiving data among them.

[0053] Each functional unit of the heart rate data analysis device 30 can be realized by a CPU 71 and / or a GPU 72 that reads a predetermined program corresponding to the function of each unit from a ROM 74 and executes it. Reference data required for program execution may be read from a database created in the ROM 74 as an auxiliary storage device. Both the CPU 71 and the GPU 72 are types of arithmetic units (processors). Here, when display-related processing is performed, a display 76 also operates in conjunction with the CPU 71 and the GPU 72, and when communication-related processing related to data transmission and reception is performed, a communication interface 75 also operates in conjunction with the CPU 71 and the GPU 72.

[0054] When the heartbeat data (including the case where the data is processed and processed down to the RR interval) is acquired by the heartbeat data analysis device 30 itself, rather than acquired as an externally measured value, any type of sensor may be used as the sensor that provides the heartbeat measurement function, and the heartbeat sensor 78 may be configured as an ECG sensor that acquires an electrocardiogram signal, a PPG sensor that acquires a pulse wave signal, a BCG sensor that acquires a ballistocardiogram signal, or the like. [Explanation of symbols]

[0055] 30...heartbeat data analysis device, 10...autonomic nerve blockage time estimation unit, 20...autonomic nerve activity time estimation unit 11...resting average interval estimation unit, 12...intrinsic heart rate estimation unit 21... intermediate function generation unit, 22... weighted average calculation unit, 23... DC energy calculation unit, 24... AC energy calculation unit, 25... autonomic nerve activity estimation unit

Claims

1. a first process for reading the RR interval calculated from the subject's heart rate data for each window and generating an intermediate function whose value is determined for each heart beat; a second process of calculating, from the intermediate function, DC component energy and AC component energy as changes from the intrinsic heart rate or the resting heart rate; a third process of estimating the autonomic nerve activity of the subject corresponding to the window as the sum of the DC component energy and the AC component energy; In the second processing, a time average within a window of the intermediate function is calculated, and then the time average is subtracted from the intermediate function to obtain an AC function in the time domain, and Parseval's equation is applied to the AC function, thereby calculating the AC component energy without converting it to the frequency domain.

2. 2. The heartbeat data analysis device according to claim 1, wherein the second processing calculates the DC component energy as the change by subtracting the energy of the intrinsic heart rate or the resting heart rate from the energy calculated from the time average within the window of the intermediate function.

3. The heartbeat data analysis device according to claim 1, characterized in that in the second processing, a DC component energy is calculated as a change from a state of an intrinsic heart rate, and a state estimated from a state of a resting heart rate is used as the state of the intrinsic heart rate.

4. 4. A program that causes a computer to function as the heartbeat data analysis device according to claim 1.

Images