Physical condition detection method, physical condition detection system and program

Abstract

A physical condition detection method performed by a computer, wherein: body motion data is acquired for a subject during a current period, which is a fixed period including the present (S20); a constructed mathematical model is used to generate a current state model expressing the current state of the physical condition of the subject from the body motion data for the subject during the current period (S21); and the current state model is outputted in order to detect changes in the physical condition of the subject using the difference obtained by comparing the current state model and a normal state model expressing a normal state of the physical condition of the subject, the normal state model being generated from the body motion data of the subject during past periods using the constructed mathematical model (S22).

Description

[Technical Field]

[0001] The present disclosure relates to a physical condition detection method. etc. Regarding. [Background technology]

[0002] For example, Patent Document 1 proposes a care system that enables a close relative in a remote location to be present at the time of death of a close relative and provide care. Patent Document 1 discloses that an estimated time of death of a subject from the present is obtained based on information showing a correlation between information on changes in electrocardiograms, heart rate, etc. obtained from many past death cases and information on how much time has passed since each of these changes until the subject's death. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2017-33502 Summary of the Invention [Problem to be solved by the invention]

[0004] However, it is difficult to collect a large amount of data on past deaths. Even if it were possible to collect a large amount of data on past deaths, because changes in electrocardiograms and heart rates vary greatly from person to person, the estimated time from the present to the time of death that can be obtained based on the information showing the correlations described above is expected to be only a few hours.

[0005] On the other hand, it is relatively easy to collect data (called body movement data) on the heart rate, breathing, and body movements of the relatives themselves, i.e., the subject themselves. It is also desirable to detect early changes in the subject's physical condition, including changes in physical condition toward death, during terminal care.

[0006] The present disclosure has been made in consideration of the above circumstances, and provides a physical condition detection method capable of detecting a change in the physical condition of a subject. etc. The purpose is to provide. [Means for solving the problem]

[0007] In order to achieve the above object, a physical condition detection method according to one embodiment of the present disclosure is a physical condition detection method performed by a computer, which acquires body movement data of a subject at a current time, which is a fixed time including the present, generates a current state model representing the current state of the subject's physical condition from the body movement data of the subject at the current time using a mathematical model constructed using the body movement data of the subject in a past period when the physical condition of the subject was steady, and outputs the current state model to detect a change in the physical condition of the subject based on a difference obtained by comparing the current state model with a steady state model representing the steady state of the physical condition of the subject generated using the mathematical model from the body movement data of the subject in the past period. The steady-state model is compared with the current-state model to obtain the difference, and an alert is sent based on the difference. do.

[0008] Some of these specific aspects may be realized using a system, a method, an integrated circuit, a computer program, or a computer-readable recording medium such as a CD-ROM, or may be realized using any combination of a system, a method, an integrated circuit, a computer program, and a recording medium. [Effects of the Invention]

[0009] According to the physical condition detection method etc. of the present disclosure, it is possible to detect changes in the physical condition of a subject. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram illustrating an example of a configuration of a physical condition detection system according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of the configuration of the physical condition detection device according to the embodiment. [Figure 3] FIG. 3 is a diagram illustrating an example of a subject from whom body movement data according to the embodiment is acquired. [Figure 4A]FIG. 4A is a flowchart showing the operation of the physical condition detection device according to the embodiment. [Figure 4B] FIG. 4B is a flowchart showing the operation of the physical condition detection device according to the embodiment. [Figure 5A] FIG. 5A is a diagram illustrating a steady state model and a current state model according to Example 1 of the embodiment. [Figure 5B] FIG. 5B is a diagram illustrating a steady state model and a current state model according to Example 1 of the embodiment. [Figure 5C] FIG. 5C is a diagram illustrating a steady state model and a current state model according to Example 1 of the embodiment. [Figure 6A] FIG. 6A is a diagram illustrating a steady state model and a current state model according to Example 2 of the embodiment. [Figure 6B] FIG. 6B is a diagram illustrating a steady state model and a current state model according to Example 2 of the embodiment. [Figure 6C] FIG. 6C is a diagram illustrating a steady state model and a current state model according to Example 2 of the embodiment. [Figure 6D] FIG. 6D is a diagram illustrating a steady state model and a current state model according to Example 2 of the embodiment. [Figure 6E] FIG. 6E is a diagram illustrating a steady state model and a current state model according to Example 2 of the embodiment. [Figure 6F] FIG. 6F is a diagram illustrating a steady state model and a current state model according to Example 2 of the embodiment. [Figure 6G] FIG. 6G is a diagram illustrating a steady state model and a current state model according to Example 2 of the embodiment. [Figure 6H] FIG. 6H is a diagram illustrating a steady state model and a current state model according to Example 2 of the embodiment. [Figure 6I] FIG. 6I is a diagram illustrating a steady state model and a current state model according to Example 2 of the embodiment. [Figure 7] FIG. 7 is a diagram illustrating an example of the configuration of a physical condition change detection device according to the embodiment. [Figure 8]FIG. 8 is a diagram showing the results of performance verification when a mathematical model is constructed using seven patterns of feature amounts. DETAILED DESCRIPTION OF THE INVENTION

[0011] A physical condition detection method according to one aspect of the present disclosure is a physical condition detection method performed by a computer, which acquires body movement data of a subject at a current time, which is a fixed period of time including the present, and generates a current state model representing the current state of the subject's physical condition from the body movement data of the subject at the current time using a mathematical model constructed using the body movement data of the subject during a past period when the subject's physical condition was steady, and outputs the current state model to detect changes in the subject's physical condition based on the difference obtained by comparing the current state model with a steady state model representing the steady state of the subject's physical condition generated using the mathematical model from the body movement data of the subject during the past period.

[0012] This makes it possible to output a current state model for detecting changes in the subject's physical condition using only the subject's body movement data. In other words, it is possible to output a current state model that can detect changes in the subject's physical condition by comparing the steady state model with the current state model.

[0013] In this way, a method for detecting physical condition can be realized that can detect changes in the physical condition of a subject.

[0014] Here, for example, the steady state model and the current state model are further compared to obtain the difference, and an alert is sent based on the difference.

[0015] This allows users to receive alerts and be informed of any changes in the subject's physical condition.

[0016] Furthermore, for example, when notifying the alert, the alert may be notified if a change in the physical condition of the subject is detected based on the difference.

[0017] This allows the user to receive an alert and be informed that a change in the subject's physical condition has been detected.

[0018] Here, for example, the difference is the distance between the steady state model and the current state model and the direction of movement from the area where the steady state model is located to the area where the current state model is located, and changes in the subject's physical condition are detected based on the distance and the direction.

[0019] Furthermore, for example, before generating the current state model, the steady state model is generated from the body movement data of the subject in the past period using the mathematical model.

[0020] Furthermore, for example, if there is no difference between the steady-state model and the current-state model, a steady-state model indicating the steady state of the subject may be regenerated from the body movement data of the subject in a new past period including the current time.

[0021] This allows the period during which the subject's physical condition was in a steady state, used to generate the steady state model, to be closer to the present, thereby making it possible to detect changes in the subject's physical condition with greater accuracy.

[0022] Furthermore, for example, the mathematical model may be constructed using features obtained by calculating statistics based on interactions from the body movement data of the subject in the past period.

[0023] This makes it possible to construct a mathematical model that can generate a steady-state model and a current-state model using only the body movement data of the subject himself / herself.

[0024] A health condition detection device according to one aspect of the present disclosure includes an acquisition unit that acquires body movement data of a subject at a current time, which is a fixed time period including the present; a processing unit that generates a current state model that represents the current state of the subject's health condition from the body movement data of the subject at the current time acquired by the acquisition unit using a mathematical model constructed using the body movement data of the subject during a past period when the subject's health condition was steady; and an output unit that outputs the current state model to detect changes in the subject's health condition based on the difference obtained by comparing the steady state model that represents the steady state of the subject's health condition generated using the mathematical model from the body movement data of the subject during the past period with the current state model.

[0025] Some of these specific aspects may be realized using a system, a method, an integrated circuit, a computer program, or a computer-readable recording medium such as a CD-ROM, or may be realized using any combination of a system, a method, an integrated circuit, a computer program, or a recording medium.

[0026] A physical condition detection method according to one aspect of the present disclosure will be described in detail below with reference to the drawings. Note that each of the embodiments described below represents a specific example of the present invention. The numerical values, shapes, materials, components, and component placement positions shown in the following embodiments are merely examples and are not intended to limit the present invention. Furthermore, among the components in the following embodiments, components that are not described in the independent claims that represent the highest concept will be described as optional components. Furthermore, the contents of each of the embodiments can be combined.

[0027] (Embodiment) First, a physical condition detection system used to realize the physical condition detection method will be described.

[0028] [1. Health condition detection system] FIG. 1 is a diagram showing an example of the configuration of a physical condition detection system according to the present embodiment.

[0029] The physical condition detection system according to this embodiment can detect changes in the physical condition of a subject by including at least a physical condition detection device 10. In this embodiment, the physical condition detection system includes a physical condition detection device 10, a physical condition change detection device 20, and a room sensor 30, which are connected by a network 40, as shown in FIG.

[0030] The room sensor 30 is installed in a room where a subject is present and acquires the subject's biometric information. The physical condition detection device 10 acquires the subject's biometric information from the room sensor 30 and generates and outputs information used to detect the subject's physical condition. The physical condition change detection device 20 detects and notifies changes in the subject's physical condition using the information output by the physical condition detection device 10. Note that the room sensor 30 is just one example, and any sensor that can acquire the subject's biometric information may be used. The room sensor 30 may be, for example, a sensor worn by the subject, or a sheet-type sensor that is placed under a bed sheet or mattress.

[0031] Each device will be described below.

[0032] [2. Physical condition detection device 10] FIG. 2 is a diagram showing an example of the configuration of the physical condition detection device 10 according to the present embodiment.

[0033] The physical condition detection device 10 outputs a state model that represents the physical condition of the subject and is generated from the subject's own body movement data as information for detecting changes in the subject's physical condition. In this embodiment, the physical condition detection device 10 includes an acquisition unit 11, a processing unit 12, and an output unit 13, as shown in Fig. 2. Each component will be described in detail below.

[0034] [2.1 Acquisition part 11] The acquisition unit 11 acquires the subject's body movement data from the room sensor 30 via the network 40. More specifically, the acquisition unit 11 acquires the subject's body movement data at a current time, which is a certain time including the present. The acquisition unit 11 may also acquire the subject's body movement data from a past period.

[0035] Here, the body movement data includes, for example, heart rate information, breathing information, and body movement information of the subject, but is not limited to these, and may be any biological information of the subject. The heart rate information includes time-series heart rate data of the subject linked to time, the breathing information includes time-series breathing data of the subject linked to time, and the body movement information includes time-series body movement data of the subject linked to time. Furthermore, body movement refers to bodily movement, and in medical settings it often refers to unconscious bodily movement, such as when sleeping.

[0036] FIG. 3 is a diagram showing an example of a subject from whom body movement data according to this embodiment is acquired.

[0037] The subject 50 shown in Fig. 3 is, for example, a patient receiving terminal care in a hospice facility. Three types of biological information, namely, heart rate information, respiratory information, and body movement information, are acquired as body movement data from a room sensor 30 installed in the hospice facility. In this embodiment, the physical condition detection device 10 outputs information for detecting a change in the subject's physical condition using the body movement data obtained when the subject 50 is sleeping in bed 60. In this case, the change in physical condition means that the physical condition of the subject 50 has changed toward death during terminal care.

[0038] The subject of the physical condition detection is not limited to the above-mentioned example of a patient in a hospice facility, but may be a patient with myocardial infarction or sleep apnea syndrome. Also, the subject of the physical condition detection does not have to be a patient, but may be anyone whose physical condition is to be detected using body movement data while sleeping.

[0039] [2.2 Processing section 12] The processing unit 12 uses a mathematical model 121 constructed using the subject's body movement data from a past period when the subject's physical condition was steady, to generate a current state model that represents the current state of the subject's physical condition from the subject's body movement data at the current time.

[0040] Furthermore, the processing unit 12 uses the mathematical model 121 to generate a steady-state model that represents the steady state of the subject's physical condition, generated from the subject's body movement data from a past period. Note that if there is no difference between the steady-state model and the current state model, the processing unit 12 may regenerate the steady-state model that represents the subject's steady state from the subject's body movement data from a new past period including the current time. The regenerated steady-state model is then used to detect the subject's physical condition.

[0041] Here, the mathematical model 121 is constructed for each subject by evaluating a model generated (which can also be referred to as trained) by constructing a neural network from a steady-state data set, i.e., body movement data of the subject from a past period when the subject's physical condition was steady. In this embodiment, the mathematical model 121 is constructed using features obtained by calculating statistics based on interactions from the subject's body movement data from a past period. Note that the features may include, for example, one or more of the moving average values (speeds) of the heart rate component and the respiratory component, the moving skewness (skewness of the distribution), the variance (disturbance) of the respiratory component, and the moving outliers (presence or absence of sudden changes) of the heart rate component.

[0042] [2.3 Output section 13] The output unit 13 outputs the state model generated by the processing unit 12. More specifically, the output unit 13 outputs the steady state model generated by the processing unit 12. Furthermore, the output unit 13 outputs the current state model generated by the processing unit 12 in order to detect a change in the physical condition of the subject from a difference obtained by comparing the steady state model with the current state model.

[0043] 2.4 Operation of the physical condition detection device 10 Next, the operation of the physical condition detection device 10 configured as above will be described.

[0044] 4A and 4B are flowcharts showing the operation of the physical condition detection device 10 according to this embodiment. Fig. 4A shows the operation from when a mathematical model is constructed until the physical condition detection device 10 generates a steady state model. Fig. 4B shows the operation of the physical condition detection device 10 to generate a current state model.

[0045] 4A, first, a mathematical model 121 is constructed using body movement data of the subject during a past period in which the subject's physical condition was steady (S10). In this embodiment, the mathematical model 121 is constructed using feature quantities obtained by calculating statistics based on interactions from the body movement data of the subject during a past period in which the subject's physical condition was steady.

[0046] Next, the computer of the physical condition detection device 10 uses the mathematical model 121 constructed in step S10 to generate a steady state model of the subject from the subject's body movement data during a past period when the subject's physical condition was steady (S11).

[0047] Next, the computer of the physical condition detection device 10 outputs the steady state model generated in step S11 (S12).

[0048] As shown in FIG. 4B, first, the computer of the physical condition detection device 10 acquires the subject's body movement data at the current time, which is a certain time including the present (S20).

[0049] Next, the computer of the physical condition detection device 10 uses the mathematical model 121 constructed in advance to generate a current state model of the subject from the body movement data of the subject at the current time acquired in step S20 (S21).

[0050] Next, the computer of the physical condition detection device 10 outputs the current state model generated in step S12 in order to detect changes in the physical condition of the subject (S22).

[0051] Then, by comparing the output steady state model with the current state model by displaying them on a display or the like, if a difference between the steady state model and the current state model is obtained, it is possible to detect a change in the subject's physical condition.

[0052] In the following, in Examples 1 and 2, it will be explained that a change in physical condition can be detected by comparing a steady state model with a current state model.

[0053] Example 1 5A to 5C are diagrams showing a steady state model and a current state model according to Example 1 of the present embodiment.

[0054] In this embodiment, the subject whose physical condition is to be detected is a patient in a hospice facility, and body movement data is acquired using a room sensor 30 as shown in FIG. 3. The steady-state model and current-state model shown in FIGS. 5A to 5C were generated using body movement data obtained when the subject 50 was asleep in bed 60. FIG. 5A shows the current-state model when the current time is "12:00-24:00 on May 17th." FIG. 5B shows the current-state model when the current time is "0:00-12:00 on May 18th." FIG. 5C shows the current-state model when the current time is "12:00-24:00 on May 18th." Each of FIGS. 5A to 5C also shows a steady-state model for a past period in which the physical condition of the subject 50 was steady. The subject 50 died around "24:00 on May 20th."

[0055] As can be seen from Figures 5A to 5C, the current state model gradually shifts downward compared to the steady state model from "0:00-12:00 on 5 / 18" shown in Figure 5B, i.e., the morning of 5 / 18. Then, from "12:00-24:00 on 5 / 18" shown in Figure 5C, i.e., the afternoon of 5 / 18, onwards, it can be seen that the current state model clearly shifts downward compared to the steady state model.

[0056] This makes it possible to detect changes in the subject's 50 physical condition not just in units of hours, but also in units of days, i.e., several days in the past.

[0057] Example 2 6A to 6I are diagrams showing a steady state model and a current state model according to Example 2 of the present embodiment.

[0058] In this example, the subject is a patient in a hospice facility, different from the subject in Example 1, and body movement data is acquired by a room sensor 30 as shown in Fig. 3. The steady-state model and current-state model shown in Figs. 6A to 6I are generated using body movement data when subject 50 is sleeping in bed 60.

[0059] Figure 6A shows a current state model when the current time is "12:00-24:00 on May 16th." Figure 6B shows a current state model when the current time is "0:00-12:00 on May 17th." Figure 6C shows a current state model when the current time is "12:00-24:00 on May 17th."

[0060] Also, FIG. 6D shows a current state model when the current time is "0:00-12:00 on 5 / 18." FIG. 6E shows a current state model when the current time is "12:00-24:00 on 5 / 18." FIG. 6F shows a current state model when the current time is "0:00-12:00 on 5 / 19." FIG. 6G shows a current state model when the current time is "12:00-24:00 on 5 / 19." FIG. 6H shows a current state model when the current time is "0:00-12:00 on 5 / 20." FIG. 6I shows a current state model when the current time is "12:00-24:00 on 5 / 20."

[0061] 6A to 6I also show a steady state model for a past period in which the physical condition of subject 50 was in a steady state. Note that subject 50 died around "24:00 on 5 / 20."

[0062] As can be seen from Figures 6A to 6H, starting from "0:00-12:00 on 5 / 18" (shown in Figure 6D), i.e., the morning of 5 / 18, the current state model gradually deviates from the steady state model (becoming an outlying distribution). Also, from "0:00-12:00 on 5 / 19" (shown in Figure 6F), i.e., the morning of 5 / 19, the current state model clearly deviates from the steady state model. And from "12:00-24:00 on 5 / 20" (shown in Figure 6I), i.e., the afternoon of 5 / 20, onwards, the current state model moves to a completely different position compared to the steady state model.

[0063] As such, it can be seen that subject 50's physical condition had changed towards death two to three days prior.

[0064] Therefore, if a patient's relatives (target persons 50) wish to be present at the end of their life, staff at the hospice facility can detect any changes in the patient's physical condition on a day-by-day basis and notify the relatives, ensuring that the relatives have time to rush over to be present at the end of their life.

[0065] The physical condition detection device 10 may detect a change in the subject's physical condition by having the physical condition change detection device 20 compare the steady state model and the current state model output by the physical condition detection device 10. The physical condition change detection device 20 will be described below.

[0066] [3. Physical condition change detection device 20] FIG. 7 is a diagram showing an example of the configuration of the physical condition change detection device 20 according to this embodiment.

[0067] The physical condition change detection device 20 can detect a change in the subject's physical condition by comparing the steady state model output by the physical condition detection device 10 with the current state model. In this embodiment, the physical condition change detection device 20 includes an acquisition unit 201, a storage unit 202, a physical condition change detection unit 203, and a notification unit 204, as shown in Fig. 7. Each component will be described in detail below.

[0068] [3.1 Acquisition part 201] The acquisition unit 11 acquires in advance the steady state model output by the physical condition detection device 10 via the network 40 and stores it in the storage unit 202. The acquisition unit 11 also acquires the current state model output by the physical condition detection device 10 via the network 40. The acquisition unit 11 acquires the current state model output by the physical condition detection device 10 at predetermined time intervals.

[0069] [3.2 Storage section 202] The storage unit 202 has a non-volatile storage area and stores information used for various processes performed by the physical condition change detection device 20. The storage unit 202 is, for example, a ROM (Read Only Memory), a flash memory, or an HDD (Hard Disk Drive). In this embodiment, the storage unit 202 stores the steady state model output by the physical condition detection device 10. The storage unit 202 may also temporarily store the current state model output by the physical condition detection device 10.

[0070] [3.3 Physical condition change detection unit 203] The physical condition change detection unit 203 compares the steady state model with the current state model to obtain a difference, if any. Furthermore, the physical condition change detection unit 203 detects a change in the subject's physical condition based on the difference obtained by comparing the steady state model with the current state model.

[0071] Here, the difference is the distance between the steady-state model and the current-state model, and the direction of movement from the part where the steady-state model is located to the part where the current-state model is located.

[0072] Here, for example, using the steady-state model and current-state model shown in FIG. 6F , the physical condition change detection unit 203 places the steady-state model and the current-state model in a space defined by the same coordinate axes and compares, for example, the center of gravity of the steady-state model distribution with the center of gravity of the current-state model distribution. If the comparison reveals a discrepancy between the centers of gravity, the physical condition change detection unit 203 obtains the discrepancy as a difference. Note that the center of gravity is an example of the distance between the steady-state model and the current-state model, and is not limited to this. The physical condition change detection unit 203 may also place the steady-state model and the current-state model in a space defined by the same coordinate axes and compare whether or not there is a discrepancy between the portion of the steady-state model distributed in the space and the portion of the current-state model distributed in the space. In this case, the physical condition change detection unit 203 obtains, as a difference, the distance between the portions distributed in the space and the direction of movement from the portion of the steady-state model distributed in the space to the portion of the current-state model distributed in the space.

[0073] If there is no difference between the steady state model and the current state model when comparing them, the physical condition change detection unit 203 may notify the physical condition detection device 10 of this fact via the notification unit 204. This allows the physical condition change detection unit 203 to trigger the physical condition detection device 10 to regenerate the steady state model. Therefore, the physical condition change detection unit 203 can update the steady state model stored in the storage unit 202 to a fresh steady state model. Furthermore, by using a fresh steady state model, the physical condition change detection unit 203 can more accurately detect changes in the subject's physical condition. In other words, the period during which the subject's physical condition was steady, which is used to generate the steady state model, can be made closer to the present, allowing the physical condition detection device 10 to more accurately detect changes in the subject's physical condition.

[0074] [3.4 Notification section 204] The notification unit 204 issues an alert based on the difference obtained by the physical condition change detection unit 203. More specifically, the notification unit 204 issues an alert when the physical condition change detection unit 203 detects a change in the subject's physical condition based on the difference it has obtained. In this way, by receiving the alert, the subject can know that a change in the subject's physical condition has occurred or that a change in the subject's physical condition has been detected.

[0075] The notification unit 204 may also send an alert to a mobile device such as a smartphone connected via the network 40. As a result, if the subject whose change in physical condition is detected is a patient at a hospice facility and a relative of the subject 50 wishes to be present at the time of death, the subject's relative or a staff member at the hospice facility will receive an alert on their mobile device when the subject's physical condition changes toward death. This allows the subject's relative to be informed of changes in the subject's physical condition on a day-by-day basis, such as several days in advance, via a staff member at the hospice facility or directly, thereby ensuring time to rush to the scene to be present at the time of death.

[0076] [4. Effects, etc.] As described above, the physical condition detection device 10 according to this embodiment can generate and output a steady state model that represents the steady state of the subject's physical condition using a mathematical model constructed from body movement data when the subject's physical condition was in a steady state. Furthermore, the physical condition detection device 10 according to this embodiment can generate and output a current state model that represents the current state of the subject's physical condition from the subject's body movement data at the current time using the constructed mathematical model. Then, by comparing the steady state model output by the physical condition detection device 10 with the current state model, it is possible to detect changes in the subject's physical condition.

[0077] In this way, according to the physical condition detection device 10 of this embodiment, it is possible to output a current state model for detecting changes in the physical condition of the subject, thereby making it possible to detect changes in the physical condition of the subject.

[0078] [Verification of mathematical models] In the above embodiment, it has been explained that the mathematical model 121 is constructed for each subject using features obtained by calculating statistics based on interactions from a steady-state data set, i.e., the subject's body movement data from a past period when the subject's physical condition was steady.

[0079] We verified the performance of a mathematical model constructed using features obtained by calculating various statistics, rather than using a steady-state data set as is, and will explain this below.

[0080] Figure 8 shows the results of performance verification when a mathematical model is constructed using seven patterns of feature quantities. The body movement data used to construct the mathematical model were time-series body movement data, heart rate data, and respiration data from the sample subjects while they were asleep in a steady state. The sample subjects were 22 patients at a hospice facility. Mathematical models were constructed for 22 individuals using each of the seven patterns of feature quantities. The overlap between the area occupied by the individual mathematical model constructed using each pattern of feature quantity and the area plotted with the daily and individual body movement data from 1 to 3 days before the date of death was determined, and the number of samples where a discrepancy was determined to exist was defined as the number of separable samples.

[0081] As shown in Figure 8, in patterns 1 to 3, only feature values calculated as moving averages from body movement data were used to suppress fluctuations in the time-series data. Patterns 1 to 3 differ only in the time of the moving average. On the other hand, patterns 4 to 7 use feature values obtained by calculating statistics based on interactions and moving averages from the same body movement data. The statistics include the proportions and differences of time-series body movement data, heart rate data, and respiratory data included in the body movement data. Patterns 4 to 7 differ only in the time of the moving average.

[0082] As can be seen from Figure 8, the best average number of days to detection was 2.75 days in advance. Furthermore, as in patterns 4 to 7, it was found that by using feature quantities obtained by calculating statistics based on interactions, it is possible to detect at least 2.4 days in advance. In other words, by constructing a mathematical model using feature quantities obtained by calculating statistics based on interactions, it is possible to detect changes in the subject's body movement data, i.e., changes in the subject's physical condition, two days before the date of death.

[0083] The above describes the physical condition detection device 10 and the like according to one or more aspects of the present disclosure based on embodiments and modifications, but the present disclosure is not limited to these embodiments. Various modifications conceivable by a person skilled in the art to the present embodiment, and configurations constructed by combining components of different embodiments, may also be included within the scope of one or more aspects of the present disclosure, as long as they do not deviate from the spirit of the present disclosure. For example, the following cases are also included in the present disclosure.

[0084] (1) The above-described physical condition detection device 10 and physical condition change detection device 20 may be used to detect changes in physical condition due to pathology, including signs of myocardial infarction and the onset of sleep apnea syndrome. In this case, the body movement data may include appropriate biological information, such as weight information, required to detect changes in physical condition.

[0085] (2) Some or all of the components constituting the above-mentioned physical condition detection device 10 and physical condition change detection device 20 may be a computer system consisting of a microprocessor, ROM, RAM, hard disk unit, display unit, keyboard, mouse, etc. A computer program is stored in the RAM or hard disk unit. Each device achieves its function when the microprocessor operates in accordance with the computer program. Here, the computer program is composed of a combination of multiple instruction codes that indicate commands to a computer to achieve a predetermined function.

[0086] (3) Some or all of the components constituting the above-described physical condition detection device 10 and physical condition change detection device 20 may be configured as a single system LSI (Large Scale Integration). A system LSI is an ultra-multifunctional LSI manufactured by integrating multiple components on a single chip, and specifically, is a computer system configured including a microprocessor, ROM, RAM, etc. A computer program is stored in the RAM. The system LSI achieves its functions when the microprocessor operates in accordance with the computer program.

[0087] (4) Some or all of the components constituting the above-mentioned physical condition detection device 10 and physical condition change detection device 20 may be configured as an IC card or a standalone module that can be attached to or detached from each device. The IC card or module is a computer system composed of a microprocessor, ROM, RAM, etc. The IC card or module may include the above-mentioned ultra-multifunctional LSI. The IC card or module achieves its functions when the microprocessor operates according to a computer program. The IC card or module may be tamper-resistant. [Industrial Applicability]

[0088] The present disclosure can be used in a physical condition detection method, physical condition detection device, and program used to detect not only changes in physical condition leading to death in terminal care, but also changes in physical condition due to pathology, including signs of myocardial infarction and the onset of sleep apnea syndrome. [Explanation of symbols]

[0089] 10. Physical condition detection device 11, 201 Acquisition Department 12 Processing section 13 Output section 20. Physical condition change detection device 30 Room Sensor 40 Network 50 Target Audience 60 beds 121 Mathematical Model 202 Storage section 203 Physical condition change detection unit 204 Notification Department

Claims

1. A physical condition detection method performed by a computer, comprising: Acquire body movement data of the subject at a current time, which is a certain time including the present; generating a current state model representing the current state of the physical condition of the subject from the body movement data of the subject at the current time using a mathematical model constructed using the body movement data of the subject during a past period when the physical condition of the subject was steady; outputting the current state model in order to detect a change in the physical condition of the subject based on a difference obtained by comparing the current state model with a steady state model that represents a steady state of the physical condition of the subject, which is generated using the mathematical model from the body movement data of the subject in the past period; comparing the steady-state model with the current-state model to obtain the difference; issuing an alert based on the difference; Physical condition detection method.

2. When notifying the alert, notifying the alert when a change in the physical condition of the subject is detected based on the difference; The physical condition detection method according to claim 1 .

3. the difference is a distance between the steady-state model and the current-state model and a direction of movement from a portion where the steady-state model is located to a portion where the current-state model is located; detecting a change in the physical condition of the subject based on the distance and the direction; The physical condition detection method according to claim 2 .

4. Before generating the current state model, generating the steady-state model from the body movement data of the subject in the past period using the mathematical model; The physical condition detection method according to any one of claims 1 to 3.

5. and further, if there is no difference between the steady-state model and the current-state model, regenerating a steady-state model representing the steady state of the subject from the body movement data of the subject in a new past period including the current time. The physical condition detection method according to any one of claims 1 to 4.

6. The mathematical model is constructed using feature quantities obtained by calculating statistics based on interactions from body movement data of the subject in the past period. The physical condition detection method according to any one of claims 1 to 5.

7. Acquire body movement data of the subject at a current time, which is a certain time including the present; generating a current state model representing the current state of the physical condition of the subject from the body movement data of the subject at the current time using a mathematical model constructed using the body movement data of the subject during a past period when the physical condition of the subject was steady; outputting the current state model in order to detect a change in the physical condition of the subject based on a difference obtained by comparing the current state model with a steady state model that represents a steady state of the physical condition of the subject, which is generated using the mathematical model from the body movement data of the subject in the past period; comparing the steady-state model with the current-state model to obtain the difference; Notifying an alert based on the difference; A program that a computer runs.

8. an acquisition unit that acquires body movement data of a subject at a current time, which is a certain time including the present; a processing unit that generates a current state model that represents the current state of the physical condition of the subject from the body movement data of the subject at the current time acquired by the acquisition unit, using a mathematical model constructed using body movement data of the subject during a past period when the physical condition of the subject was steady; an output unit that outputs the current state model in order to detect a change in the physical condition of the subject based on a difference obtained by comparing a steady state model that represents a steady state of the physical condition of the subject, generated using the mathematical model from the body movement data of the subject in the past period, with the current state model; a physical condition change detection unit that obtains the difference by comparing the steady state model with the current state model; a notification unit that notifies an alert based on the difference. Health condition detection system.

Images