Data analysis system, data analysis method, and program

Abstract

To provide a data analysis system capable of analyzing a symptom of a heart disease by associating an action in a life with a function of a heart.SOLUTION: The data analysis system includes an acceleration sensor that acquires acceleration data of a user's torso, a heart rate sensor that acquires heart rate data of the user, and an analysis device that analyzes that a heart disease symptom has improved or is in a good state when a value of a slope of a regression line of a point group with the acceleration data or a value based on the acceleration data on a horizontal axis or a vertical axis and the heart rate data on the vertical axis or the horizontal axis is within a predetermined range.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a data analysis system, a data analysis method, and a program for analyzing acquired sensor data. [Background technology]

[0002] Cardiac disease can be prevented to some extent by cardiac rehabilitation, such as moderate exercise. However, to continuously understand the state of heart disease, expensive tests such as blood tests must be performed regularly, and they cannot be visualized at a low cost in daily life. In addition, it is difficult to judge the state of heart disease based on the individual's awareness in daily life.

[0003] Currently, there is no technology to monitor cardiac disease symptoms in an unobserved environment, which means that cardiac rehabilitation cannot always be carried out at the appropriate time and with the appropriate content.

[0004] Body weight is an index that can be used to easily monitor heart disease symptoms on a daily basis. Non-Patent Document 1 shows that there is a correlation between changes in body weight and in-hospital mortality in patients with heart disease. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] Masaaki Konishi et al., “Association of weight change and in-hospital mortality in patients with repeated hospitalization for heart failure,” [online], December 23, 2020, Journal of Cachexia, Sarcopenia, and Muscle, [searched June 19, 2020], Internet<URL:https: / / onlinelibrary.wiley.com / doi / 10.1002 / jcsm.13170> Summary of the Invention [Problem to be solved by the invention]

[0006] However, weight gain or loss does not effectively reflect the condition of the heart or the quality of daily activity, and is also an indicator that changes depending on various other factors, making it difficult to accurately predict the continuous improvement or worsening of heart disease symptoms based on weight.

[0007] On the other hand, it is possible to associate daily activities with cardiac function and analyze symptoms of heart disease by combining sensor data obtained from wearable devices that do not interfere with daily life, but such technology does not currently exist.

[0008] Therefore, an object of the present disclosure is to provide a data analysis system that can associate daily activities with cardiac function and analyze symptoms of heart disease. [Means for solving the problem]

[0009] The data analysis system disclosed herein includes an acceleration sensor, a heart rate sensor, and an analysis device. The acceleration sensor acquires acceleration data of a user's torso. The heart rate sensor acquires heart rate data of the user. The analysis device analyzes that a heart disease symptom has improved or is in good condition if the slope of a regression line of a point cloud with the acceleration data or a value based on the acceleration data as the horizontal or vertical axis and the heart rate data as the vertical or horizontal axis is within a predetermined range. [Effects of the Invention]

[0010] According to the data analysis system of the present disclosure, it is possible to associate daily activities with cardiac function and analyze symptoms of heart disease. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a block diagram showing the device configuration of a data analysis system according to a first embodiment. [Figure 2] FIG. 2 is a block diagram showing the functional configuration of the analysis device of the first embodiment. [Figure 3] 4 is a flowchart showing the operation of each device in the data analysis system according to the first embodiment. [Figure 4] A diagram conceptually showing the relationship between the slope of the regression line and heart disease symptoms. [Figure 5] FIG. 10 is a block diagram showing the functional configuration of an analysis device according to a second embodiment. [Figure 6] 10 is a flowchart showing the operation of each device in the data analysis system according to the second embodiment. [Figure 7] FIG. 10 is a block diagram showing the functional configuration of an analysis device according to a third embodiment. [Figure 8] 10 is a flowchart showing the operation of each device in the data analysis system according to the third embodiment. [Figure 9] FIG. 11 is a block diagram showing the functional configuration of an analysis device according to a modified example of the third embodiment. [Figure 10] 10 is a flowchart showing the operation of each device in the data analysis system according to a modification of the third embodiment. [Figure 11] A graph showing an example of monthly fluctuations in a patient's resting index. [Figure 12] A graph showing an example of monthly fluctuations in a patient's proBNP levels. [Figure 13] FIG. 2 is a diagram showing an example of the functional configuration of a computer. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, embodiments of the present disclosure will be described in detail. Note that components having the same functions are assigned the same numbers, and redundant explanations will be omitted. [Embodiment 1]

[0013] The device configuration of the data analysis system of this embodiment will be described below with reference to Fig. 1. As shown in the figure, the data analysis system 1 of this embodiment includes an acceleration sensor 11, a heart rate sensor 12, and an analysis device 13. The specifications of each device will be described below.

[0014] <Acceleration sensor 11, heart rate sensor 12> Acceleration sensor 11 is preferably a three-axis acceleration sensor. Acceleration sensor 11 is used to detect the movement of the user's torso, so it is preferably fixed to an attachment (such as a rubber band or belt) that can be attached to the user's torso. Heart rate sensor 12 is used to detect the user's heartbeat (pulse), so it is preferably fixed to an attachment (such as a rubber band or belt) that can be attached to a part of the user's body that can detect the heartbeat (pulse).

[0015] Furthermore, acceleration sensor 11 and heart rate sensor 12 may have a built-in small transmitter for transmitting acceleration data and heart rate data to other communication devices, or may be connected to a small transmitter. For example, acceleration sensor 11 and heart rate sensor 12 may transmit acceleration data and heart rate data to a mobile phone such as a smartphone or a portable communication terminal via a transmitter, and the mobile phone or communication terminal may transmit the acceleration data and heart rate data to analysis device 13 described below (the mobile phone or communication terminal itself may be analysis device 13). Furthermore, acceleration sensor 11 and heart rate sensor 12 may transmit acceleration data and heart rate data directly to analysis device 13 via a transmitter.

[0016] The acceleration sensor 11 and the heart rate sensor 12 may be realized as different functions of the same device, or may be realized as different devices.

[0017] <Analyzer 13> The analysis device 13 can be realized by a general-purpose PC or the like, but as described above, it may also be a mobile phone such as a smartphone, a portable communication terminal, or a wearable device. All of the functions of the acceleration sensor 11, heart rate sensor 12, and analysis device 13 described above may be realized on a single wearable device.

[0018] <Functional configuration of the analysis device 13> The functional configuration of the analysis device 13 will be described with reference to Fig. 2. As shown in the figure, the analysis device 13 of this embodiment includes an acceleration data acquisition unit 131, a heart rate data acquisition unit 132, a regression line derivation unit 133, a cardiac disease symptom analysis unit 134, and a parameter storage unit 135. Below, the operation of each sensor and each component of the analysis device 13 will be described with reference to Fig. 3.

[0019] The acceleration sensor 11 acquires acceleration data of the user's trunk (S11), and the heart rate sensor 12 acquires heart rate data of the user (S12).

[0020] The acceleration data acquisition unit 131 of the analysis device 13 acquires acceleration data from the acceleration sensor 11 (S131). The acceleration data acquisition unit 131 may perform filtering (high-pass filtering or average deviation calculation) to remove the influence of gravitational acceleration and noise from the acquired acceleration data, and output the norm of the vector after filtering.

[0021] The heart rate data acquiring unit 132 acquires heart rate data from the heart rate sensor 12 (S132). The heart rate data acquiring unit 132 may use the heart rate data (hr) acquired from the heart rate sensor 12 as is.

[0022] The regression line derivation unit 133 derives a regression line of the point cloud with the acceleration data or a value based on the acceleration data as the horizontal or vertical axis and the heart rate data as the vertical or horizontal axis (S133). For example, the regression line derivation unit 133 may calculate a cumulative or weighted average of the acceleration data at time t-Δt and treat it as a point cloud for deriving the regression line together with the heart rate at time t.

[0023] The cardiac symptom analysis unit 134 analyzes that the cardiac symptom has improved or is in a good state when the value of the slope of the regression line is within a predetermined range, and analyzes that the cardiac symptom has worsened or is in a bad state when the value of the slope is outside the predetermined range (S134). The parameters that determine the predetermined range of the slope value are stored in advance in the parameter storage unit 135.

[0024] <Relationship between the slope of the regression line and heart disease symptoms> When the acceleration of the user's trunk is high, it generally means that the user is engaged in vigorous activity or exercise. When the user's activity or exercise is vigorous, the user's heart rate also increases. Therefore, as shown by the solid line in the graph in Figure 4, it can be said that a normal physical reaction is when the acceleration of the user's trunk and the user's heart rate are proportional to each other within a certain slope range (for example, within the hatched area in the figure).

[0025] On the other hand, if the slope is too steep, as shown by the dashed line in the graph, it means that the heart's function per beat is declining and the heart needs to beat excessively. On the other hand, if the slope is too shallow, as shown by the dashed line in the graph, it means that the heart is not responding adequately to the energy expenditure caused by exercise.

[0026] Therefore, if the slope of the regression line falls within a predetermined range, it can be said that the symptoms of heart disease have improved or are in a good condition, but if the slope of the regression line falls outside the predetermined range, it is expected that the symptoms of heart disease have worsened or are in a worse condition.

[0027] In this way, according to the data analysis system 1 of this embodiment, the acceleration data of the user's trunk and the heart rate data are analyzed in combination, thereby making it possible to continuously estimate symptoms of heart disease.

[0028] According to the data analysis system 1 of this embodiment, by using both acceleration data of the user's torso and heart rate data, the state of heart disease can be estimated at low cost throughout daily life without the user having to be particularly conscious or undergoing any specific tests or actions.

[0029] Furthermore, the data analysis system 1 of this embodiment uses a single conversion formula that requires an extremely small amount of calculation, so that the analysis device 13 and the entire data analysis system 1 can be realized by a small device with few resources. [Embodiment 2]

[0030] The data analysis system 1 of the first embodiment may not be able to derive an accurate regression line for users with a low level of physical activity (for example, users who are bedridden or hospitalized) because the change in acceleration in their daily lives is too small.

[0031] Therefore, in the second embodiment, a data analysis system is disclosed that can continuously analyze the heart disease symptoms of a user with a low physical activity level by deriving a regression line focusing on the user's trunk angle.

[0032] Although not shown, the data analysis system 2 of this embodiment includes the same acceleration sensor 11 and heart rate sensor 12 as in the first embodiment, and an analysis device 23 different from that in the first embodiment.

[0033] The functional configuration of the analysis device 23 of this embodiment will be described with reference to Fig. 5. As shown in the figure, the analysis device 23 of this embodiment includes an acceleration data acquisition unit 131, a heart rate data acquisition unit 132, a regression line derivation unit 233, a cardiac disease symptom analysis unit 134, and a parameter storage unit 135, and the configuration other than the regression line derivation unit 233 is the same as in embodiment 1. The operation of the regression line derivation unit 233 will be described below with reference to Fig. 6.

[0034] <Regression line derivation unit 233> The regression line derivation unit 233 derives a regression line of the point cloud with the horizontal or vertical axis being the torso angle (θ), which is determined based on the acceleration data and has the user's lying position on a horizontal plane as 0° and the user's sitting position, edge sitting position, or standing position as 90°, or the value of the sine function (sinθ) of the torso angle (θ), and the vertical or horizontal axis being the heart rate data (S233).

[0035] For example, the regression line derivation unit 233 may calculate the angle between the vector of the acceleration data and the horizontal plane (ground) as the trunk angle (θ). The method of analyzing heart disease symptoms using the slope of the regression line is the same as in the first embodiment.

[0036] <Relationship between the slope of the regression line and heart disease symptoms> Even when a user is hospitalized and mostly bedridden, there are many situations in which the angle of the user's torso changes, such as when the user changes the angle of the bed or sits up in bed. Since the energy required when lying down and when sitting up differs, the user's heart rate responds in the same way as in embodiment 1. Therefore, if the slope of the regression line in embodiment 2 falls within a predetermined range, it can be evaluated that the symptoms of heart disease have improved or are in a good condition, but if it is outside the predetermined range, it can be evaluated that the symptoms of heart disease have worsened or are in a worse condition.

[0037] According to the data analysis system 2 of this embodiment, the trunk angle (θ) or its sine function (sinθ) is calculated as a feature based on acceleration data, and by referring to changes in heart rate response, it is possible to analyze heart disease symptoms in users with low physical activity levels, such as those who are hospitalized. [Embodiment 3]

[0038] The data analysis system 2 of embodiment 2 may not be able to accurately analyze cardiac disease symptoms for users whose resting heart rate varies greatly or whose resting heart rate is high, since there is little room for the heart rate to increase.

[0039] Therefore, in embodiment 3, we disclose a data analysis system that focuses on the y-intercept (x-intercept) of the regression line and can continuously analyze the heart disease symptoms of users whose resting heart rates vary greatly and users whose resting heart rates are high.

[0040] Although not shown, the data analysis system 3 of this embodiment includes the same acceleration sensor 11 and heart rate sensor 12 as in the first embodiment, and an analysis device 33 different from that in the first embodiment.

[0041] The functional configuration of the analysis device 33 of this embodiment will be described with reference to Fig. 7. As shown in the figure, the analysis device 33 of this embodiment includes an acceleration data acquisition unit 131, a heart rate data acquisition unit 132, a regression line derivation unit 133, a cardiac disease symptom analysis unit 134, a parameter storage unit 135, and a resting index derivation unit 336, and the configuration other than the resting index derivation unit 336 is the same as in the first embodiment.

[0042] Hereinafter, the operation of the resting index derivation unit 336 will be described with reference to FIG.

[0043] <Resting state index derivation unit 336> The resting index derivation unit 336 outputs the y-intercept value of the regression line of the point cloud, with the acceleration data or a value based on the acceleration data as the horizontal axis and the heart rate data as the vertical axis, as the resting index of the cardiac disease symptom (S336). Note that the resting index derivation unit 336 may also output the x-intercept value of the regression line of the point cloud, with the acceleration data as the vertical axis and the heart rate data as the horizontal axis, as the resting index of the cardiac disease symptom.

[0044] [Variations] In the above-described analysis device 33, a part of the analysis device 13 of embodiment 1 is modified, but a similar modification may be made to the analysis device 33 of embodiment 2. As shown in Fig. 9, an analysis device 33A of a modification of embodiment 3 includes an acceleration data acquisition unit 131, a heart rate data acquisition unit 132, a regression line derivation unit 233, a cardiac disease symptom analysis unit 134, a parameter storage unit 135, and a resting index derivation unit 336A, and the configuration other than the resting index derivation unit 336A is the same as that of embodiment 2.

[0045] Hereinafter, the operation of the resting index derivation unit 336A will be described with reference to FIG.

[0046] <Resting state index derivation unit 336A> The resting index derivation unit 336A outputs the y-intercept value of the regression line of the point cloud, with the trunk angle or the value of the sine function of the trunk angle as the horizontal axis and the heart rate data as the vertical axis, as the resting index of the cardiac disease symptom (S336A). Note that the resting index derivation unit 336A may also output the x-intercept value of the regression line of the point cloud, with the trunk angle or the value of the sine function of the trunk angle as the vertical axis and the heart rate data as the horizontal axis, as the resting index of the cardiac disease symptom. [Embodiment 4]

[0047] In the data analysis systems 2 and 3 of the second and third embodiments, if the user is active for a long time, the analysis accuracy is likely to deteriorate. This is because the change in heart rate due to activity or exercise is much larger than the heart rate response due to changes in posture, making it impossible to calculate the regression line as intended.

[0048] Therefore, in the fourth embodiment, a data analysis system 4 (not shown) is disclosed that can continuously analyze the user's heart disease symptoms based only on the resting heart rate by excluding acceleration data that exceeds a predetermined value.

[0049] The analysis device 43 of this embodiment is characterized by including a regression line derivation unit 433 (not shown) instead of the regression line derivation unit 133 of embodiment 3 or the regression line derivation unit 233 of the modified example of embodiment 3. If the acceleration (or the trunk angle, or the value of the sine function of the trunk angle) exceeds a predetermined threshold, the regression line derivation unit 433 derives the regression line after previously adding processing to exclude points (data sets) consisting of the heart rates corresponding to this acceleration from the derivation of the regression line (S433). As in embodiment 3, the value of the y-intercept or x-intercept of the regression line is output as a resting index of a cardiac disease symptom.

[0050] According to the data analysis system 4 of this embodiment, it is possible to continuously analyze the user's cardiac disease symptoms based solely on the resting heart rate, regardless of the level of activity of the user. Generally, as cardiac disease symptoms worsen, the resting cardiac disease symptom index based on the value of the y-intercept or x-intercept tends to increase. [Embodiment 5]

[0051] In the data analysis systems 2 and 3 of the second and third embodiments, if there is a large bias in the distribution of the body angles, the weighting of values ​​at locations with a large number of data points will be increased in the process of deriving the regression line, which may result in a deterioration in accuracy.

[0052] Therefore, in embodiment 5, a data analysis system 5 (not shown) is disclosed that reduces bias in the distribution of torso angles and improves the accuracy of analysis by using statistical values ​​of torso angles and heart rates within a predetermined range of torso angles as representative values ​​for that range.

[0053] Analysis device 53 of this embodiment is characterized by including regression line derivation unit 533 (not shown) instead of regression line derivation unit 133 of embodiment 3 or regression line derivation unit 233 of the modified example of embodiment 3. Regression line derivation unit 533 derives a regression line for a predetermined range of trunk angles using a point cloud in which statistical values ​​(for example, average values ​​or median values) of trunk angles and heart rates within that range are used as representative values ​​(S533).

[0054] According to the data analysis system 5 of this embodiment, the bias in the distribution of trunk angles can be reduced, thereby improving the accuracy of analyzing heart disease symptoms.

[0055] <Evaluation experiment of new index (resting index)> The increase and decrease of the new index (resting index) disclosed in the above embodiment was observed in the data of patients who were actually hospitalized with heart disease.

[0056] The original data for the regression line for the new index is one day's worth of data for the patient in question, and the horizontal axis of the data points and regression line is the sine function value (sinθ) of the torso angle θ, where 0° is when the patient is lying on a horizontal plane and 90° is when the patient is in a sitting position, edge sitting position, or standing position, and the vertical axis is the heart rate of the patient in question.

[0057] To take into account the duration of immobility, the acceleration used to derive the trunk angle θ was smoothed over 6 minutes. Due to the influence of recent movement, data where acceleration was above a certain value and where sinθ = 1 was near was excluded from the point cloud used to derive the regression line. To equalize the number of data points for trunk angle θ, the median heart rate for each window of a fixed width for trunk angle θ was obtained to determine the regression line.

[0058] Figure 11 shows the results of plotting the y-intercept of the regression line for each date (normalized to 1 as the value on the final day) on the vertical axis and each date on the horizontal axis. Figure 12 also shows the results of plotting the proBNP values ​​for the same patient on each date (normalized to 1 as the value on the final day) on the vertical axis and each date on the horizontal axis. It can be seen that the new index using the y-intercept (x-intercept) of the regression line shows a similar trend to the proBNP values. Furthermore, it was confirmed that the new index can quantitatively estimate the increase or decrease in proBNP levels at the time of admission or discharge with an accuracy of approximately 79%.

[0059] [Example] An example that further embodies the processing of the above-described embodiment will be described below. In this example, the acceleration sensor 11 and heart rate sensor 12 are combined into a single device, and triaxial acceleration is acquired at 25 Hz and heart rate is acquired at 1 Hz. Note that data acquisition and data processing may be performed on the same device or on separate devices.

[0060] For example, an edge device can acquire acceleration and heart rate, and transmit the data to a smartphone or the cloud via wireless communication such as Bluetooth (registered trademark), where the data can be processed.

[0061] The obtained triaxial acceleration and heart rate may contain missing values ​​depending on the observation environment, so it is preferable to perform missing value processing as necessary. For example, processing can be performed by interpolation or linear interpolation using the immediately preceding or following values, or their average values. To display the exercise load during activity, processing using a moving window is performed from then on. The width of the moving window is, for example, about 10 seconds or 1 minute. The step width does not need to match the width of the moving window; it can be, for example, 5 seconds or 10 seconds, and the exercise load for that period is estimated for each moving window.

[0062] First, we will discuss preprocessing of the acceleration data. After processing the acceleration data for each dimension, it is integrated. First, to remove the effects of gravitational acceleration, the average value within a moving window is calculated and then this average is divided by the total. This average value can be considered the gravity vector. Next, a high-pass filter is applied to remove noise and estimate exercise load based on human activity. For example, the cutoff frequency is set to 10 Hz. The torso angle is calculated based on the vector after applying the filter and the gravity vector. After that, the average absolute value for the relevant period is calculated. This average value is then used as the component for each dimension to calculate the norm of the three-axis acceleration vector.

[0063] When performing an analysis based on the first embodiment, for example, the time interval is set to one minute and the fixed period is set to one day. If data could be acquired correctly for all time periods, 1,440 plots could be drawn to calculate one index. Linear regression is performed on this data to obtain a regression line. The slope of this regression line is an index representing the state of cardiac disease.

[0064] This indicator can be displayed on the patient's smartphone, for example, and used for cardiac rehabilitation, remote medical consultations by the patient's doctor, and remote monitoring.

[0065] When performing the analysis based on the second embodiment, the horizontal axis in the first embodiment is replaced with sin(θ).

[0066] When performing the analysis according to the third embodiment, the y-intercept (x-intercept) of the regression line in the second embodiment is used as a resting index of cardiac disease symptoms.

[0067] When performing an analysis based on the fourth embodiment, only stationary states can be extracted by plotting data points after applying a filter to the acceleration data, such as acc_fil<0.1 or acc_fil<0.01 in the third embodiment. Because it is important to be stationary for a certain period of time, not just immediately before, the acceleration window width used for filtering does not necessarily have to be one minute; it can also be longer, such as six minutes. Furthermore, to remove standing postures, the plot range can be set to, for example, sin(θ)<0.75, and a regression line can be found using only the data in that range.

[0068] When performing an analysis based on the fifth embodiment, before calculating the regression line in the fourth embodiment, windows are set on the horizontal axis, for example, at intervals of 0.1 or 0.05, and the median of the data within each range is calculated and replaced with one representative point for each window. A regression line is then drawn based on these representative points, and the index is calculated in the same manner.

[0069] <Additional Notes> The functions performed by the components described herein may be implemented in circuitry or processing circuitry, including general-purpose processors, application-specific processors, integrated circuits, ASICs (Application Specific Integrated Circuits), a CPU (a Central Processing Unit), conventional circuits, and / or combinations thereof, programmed to perform the described functions. A processor includes transistors and other circuits and is considered to be circuitry or processing circuitry. A processor may also be a programmed processor that executes programs stored in memory.

[0070] In this specification, a circuitry, unit, or means is hardware that is programmed to realize or performs the described functions, which may be any hardware disclosed herein or any hardware known to be programmed to realize or perform the described functions.

[0071] If the hardware is a processor considered to be a type of circuitry, the circuitry, means, or unit is a combination of the hardware and software used to configure the hardware and / or processor.

[0072] The various processes described above can be implemented by loading a program that executes each step of the above method into the recording unit 10020 of the computer shown in Figure 13 and operating the control unit 10010, input unit 10030, output unit 10040, etc.

[0073] The program describing the processing contents can be recorded on a computer-readable recording medium, which may be, for example, a magnetic recording device, an optical disk, a magneto-optical recording medium, a semiconductor memory, or any other suitable recording medium.

[0074] The program may be distributed, for example, by selling, transferring, lending, etc. a portable recording medium such as a DVD or CD-ROM on which the program is recorded. Furthermore, the program may be stored in a storage device of a server computer, and then transferred from the server computer to another computer via a network, thereby distributing the program.

[0075] A computer that executes such a program may first temporarily store the program recorded on a portable recording medium or transferred from a server computer in its own storage device. Then, when executing a process, the computer reads the program stored on its own recording medium and executes the process in accordance with the read program. Alternatively, the computer may read the program directly from a portable recording medium and execute the process in accordance with the program. Furthermore, the computer may execute the process in accordance with the program each time a program is transferred from a server computer to the computer. The server computer may not transfer the program to the computer, but may instead execute the process through a so-called ASP (Application Service Provider) service, which realizes the processing function by issuing an execution instruction and obtaining the results. Furthermore, the server computer may execute the process on a terminal using a so-called SaaS (Software as a Service) service, which allows users to use part of the server computer along with the program. In this embodiment, the program includes information used for computer processing that is equivalent to a program (such as data that is not a direct instruction to the computer but has properties that define computer processing).

[0076] Furthermore, in this embodiment, the device is configured by executing a predetermined program on a computer, but at least a part of the processing contents may be realized by hardware.

Claims

1. an acceleration sensor that acquires acceleration data of the user's torso; a heart rate sensor for acquiring heart rate data of the user; The analysis device analyzes that the heart disease symptoms have improved or are in a good condition when the gradient of the regression line of the point cloud, which has the acceleration data or a value based on the acceleration data as the horizontal or vertical axis and the heart rate data as the vertical or horizontal axis, is within a predetermined range. Data analysis system.

2. 10. The data analysis system of claim 1, The analysis device The analysis is performed based on the trunk angle, which is determined based on the acceleration data, with the user's lying position on a horizontal plane being 0° and their sitting position, edge sitting position, or standing position being 90°, or based on the slope of the regression line of the point cloud, with the value of the sine function of the trunk angle as the horizontal or vertical axis and the heart rate data as the vertical or horizontal axis. Data analysis system.

3. 3. The data analysis system of claim 2, The analysis device The value of the y-intercept of the regression line of the point group with the acceleration data or a value based on the acceleration data as the horizontal axis and the heart rate data as the vertical axis is output as the index of the heart disease symptom. Data analysis system.

4. 3. The data analysis system of claim 2, The analysis device The value of the x-intercept of the regression line of the point group with the acceleration data or a value based on the acceleration data as the vertical axis and the heart rate data as the horizontal axis is output as the index of the heart disease symptom. Data analysis system.

5. 3. The data analysis system of claim 2, The analysis device The value of the y-intercept of the regression line of the point group with the trunk angle or the value of the sine function of the trunk angle as the horizontal axis and the heart rate data as the vertical axis is output as the index of the heart disease symptom. Data analysis system.

6. 3. The data analysis system of claim 2, The analysis device The value of the x-intercept of the regression line of the point group with the trunk angle or the value of the sine function of the trunk angle as the vertical axis and the heart rate data as the horizontal axis is output as the index of the heart disease symptom. Data analysis system.

7. A data analysis method performed by an acceleration sensor, a heart rate sensor, and an analysis device, comprising: the acceleration sensor acquires acceleration data of the user's trunk; the heart rate sensor acquires heart rate data of the user; The analysis device analyzes that the heart disease symptoms have improved or are in a good condition when the gradient of the regression line of the point cloud, with the acceleration data or a value based on the acceleration data as the horizontal or vertical axis and the heart rate data as the vertical or horizontal axis, is within a predetermined range. Data analysis methods.

8. A program that causes a computer to function as the analysis device of the data analysis system according to any one of claims 1 to 6.

Images