WO2022269959A1 - Human life traceability device and improvement plan proposal support method - Google Patents

Abstract

The present invention addresses the problem of proposing an improvement plan such as for improving the health state, such as the living behavior, of a person to be analyzed. In the present invention, this human life traceability device 10 for analyzing the behavior of a person 2 to be analyzed includes: a storage unit 19 that stores living behavior information, which is time-series data indicating the living behavior of the person 2 to be analyzed, and activity amount information, which is time-series data indicating the activity amount of the person 2 to be analyzed, both types of information being based on sensor data for the person 2 to be analyzed; a detection unit 14 that detects an activity change satisfying a prescribed condition on the basis of the activity amount information; a comparison unit 15 that, when the activity change is detected, compares information relating to living behavior corresponding to the living behavior information in a period having a prescribed relationship with the timing at which the activity change occurred, and as a result of the comparison, identifies information relating to living behavior in which a living behavior change satisfying a prescribed condition has occurred; and an improvement plan estimation unit 17 that estimates an improvement plan for living behavior on the basis of the identified information relating to the living behavior.

Description

Human life traceability device and improvement proposal support method

The present invention relates to human life traceability that manages and analyzes human behavior and activities. In particular, it relates to techniques for managing human health conditions.

In recent years, communication technology and sensor technology have developed, making it possible to measure human behavior and activity levels in everyday life for the purpose of health management. For example, Patent Literature 1 discloses a technique for determining whether or not there is a change in lifestyle patterns based on walking data indicating the amount of activity.

Japanese Patent Application Laid-Open No. 2020-184168

Here, for health management, it is effective to detect changes in lifestyle behaviors, including lifestyle patterns, but unlike activity levels, it is difficult to directly detect them with sensors. For example, changes in the amount or timing of meals may not be due to health reasons, such as seasonal factors.

Also, in Patent Document 1, the presence or absence of changes in lifestyle patterns is determined, but the analysis of changes was not sufficiently done. In human life traceability, including health management, detailed analysis of human activities such as living behavior makes it possible to present improvement plans for improving health conditions.

Here, an object of the present invention is to provide a technology that can analyze human activities in more detail.

In order to solve the above problems, the present invention detects a change in the amount of activity of the person to be analyzed, corresponds to the detected change in the amount of activity, for example, identifies a change in living behavior in the near period, and identifies the identified Predict improvement plans for changes in living behavior. More specifically, in a human life traceability device that manages and analyzes the activity and behavior of an analysis subject, time-series data showing the life behavior of the analysis subject based on sensor data of the analysis subject from a sensor A storage unit that stores daily activity information and activity amount information that is time-series data indicating the activity amount of the person to be analyzed; a detection unit that detects a change in activity that satisfies a predetermined condition based on the activity amount information; When the activity change is detected, the information on the living activity corresponding to the living activity information in a period having a predetermined relationship with the timing at which the activity change occurred is compared with each other, and as a result of the comparison, the living activity change that satisfies the predetermined condition. A human life traceability device having a comparison unit that identifies information about the living behavior that caused the above-mentioned living behavior, and an improvement plan estimating unit that estimates an improvement plan for the living behavior based on the identified information about the living behavior. The invention also includes an improvement proposal support method using a human life traceability device. Furthermore, the present invention also includes a program for causing the human life traceability device to function as a computer and a storage medium for storing the program.

According to the present invention, it is possible to analyze the activities of the person to be analyzed in more detail, so that more appropriate health management is possible.

1 is a block diagram showing the configuration of a human life traceability system in one embodiment of the present invention; FIG. It is a hardware block diagram of the human life traceability apparatus in one Example of this invention. It is a figure which shows the living activity information used by one Example of this invention. It is a figure which shows the active mass information used by one Example of this invention. It is a figure which shows the living activity feature-value used by one Example of this invention. It is a figure which shows the activity feature-value used by one Example of this invention. It is a figure which shows the index value correspondence table|surface used by one Example of this invention. 4 is a flow chart showing a processing flow in one embodiment of the present invention; FIG. 4 is a diagram showing a display example in one embodiment of the present invention; It is a figure explaining the concept of the process of the modification in one Example of this invention. It is a flowchart which shows the processing flow of step S6 and S7 of the modification in one Example of this invention.

Hereinafter, one embodiment of the present invention will be described in detail with reference to the drawings, but the present invention is not limited to the following embodiments, and various modifications and applications within the technical concept of the present invention Also included in the scope.
<Configuration>
Prior to the configuration of this embodiment, application services and participants in this embodiment will be described.
In this embodiment, the activities and behaviors of the person to be analyzed 2 are detected regardless of whether they are inside the house 20 or outside the house, and the activities are analyzed based on the detection results. For this reason, in this embodiment, health-related services provided by insurance companies, day service providers, and the like are made available. The health-related service is a service for analyzing the activities and behaviors of the person to be analyzed 2, such as a health management service and a watching service by an applicant or a relative.

Then, analysis subject 2 lives mainly in home 20 such as his own home or workplace. Further, sensors are installed inside and outside the home 20 to detect the activities and actions of the person 2 to be analyzed.

In addition, it is desirable that the person to be analyzed 2 has a user terminal 23 that is used when receiving the services described above. Furthermore, it is desirable that related parties such as relatives of the person to be analyzed 2 use the user terminals 31 and 32 .

Furthermore, insurance companies and day service providers are assumed as service providers of this embodiment, and each of these will use a server for service provision. Furthermore, a human life traceability device 10 is provided for the services mentioned above.

The configuration of this embodiment will be described below. FIG. 1 is a block diagram showing the configuration of the human life traceability system in this embodiment. In this embodiment, the human life traceability device 10, the insurance company server 41 of the insurance company, and the day service provider server 51 of the day service provider are connected to each other via a network 60 such as the Internet. Sensors and user terminals (23, 31, 32) that cooperate with these various devices are also connected to the network 60. FIG.

First, the sensors are installed inside and outside the home 20 as described above. Here, the sensors 21 include activity amount sensors 21-1 and 21-2 for detecting the amount of activity of the person 2 to be analyzed, and living behavior sensors 22-1 and 22 for identifying the living behavior of the person 2 to be analyzed. -2 can be classified.

The activity sensors 21-1 and 21-2 can be realized by activity meters and cameras using millimeter wave radars and microwaves. In other words, the activity amount sensors 21-1 and 21-2 should be able to detect or specify the amount of activity such as the movement speed, acceleration, and energy consumption of the person to be analyzed 2 as sensor data (activity sensor data). Also, the living activity sensors 22-1 and 22-2 can be implemented by human sensors, sensors provided in home appliances, and the like. In other words, the living activity sensors 22-1 and 22-2 only need to detect sensor data (living activity sensor data) that can specify the living activities of the person 2 to be analyzed, such as sleeping, going out, eating, and doing housework.

It should be noted that the number of sensors is not limited to the number shown in the figure, and their types are not limited to those described above.

Also, the human life traceability device 10 is a device that executes the main processing of this embodiment. Here, FIG. 1 shows functional blocks of the human life traceability device 10 . This human life traceability device 10 includes, as functional blocks, an action estimation unit 11, an action feature amount calculation unit 12, an activity feature amount calculation unit 13, a detection unit 14, a comparison unit 15, a change factor estimation unit 16, an improvement plan estimation It has a unit 17 , an input/output unit 18 and a storage unit 19 .

First, the behavior estimation unit 11 creates living behavior information 101 indicating the living behavior of the person to be analyzed 2 based on the living behavior sensor data. Note that the living activity information 101 will be described in the item of "Information/Data". The activity estimator 11 may create the living activity information 101 from information obtained by subjecting the living activity sensor data to digital conversion or the like, or may create the living activity information 101 from the living activity sensor data.

In addition, the behavior feature quantity calculation unit 12 creates a living behavior feature quantity 103 that indicates the features of the living behavior information 101 . The living behavior feature quantity 103 will also be described in the item of "information/data".

Further, the activity feature amount calculation unit 13 creates an activity feature amount 104 indicating the feature of the activity amount information 102 indicating the activity amount of the person to be analyzed 2, which is the activity amount sensor data of the activity amount sensors 21-1 and 21-2. do. The activity amount information 102 and the activity feature amount 104 will also be explained in the item of "information/data". In addition, in the present embodiment, the activity amount information 102 has the same content as the activity amount sensor data. That is, the activity amount information 102 is obtained by converting the activity amount sensor data into a digital value or the like. However, the activity amount information 102 may use the activity amount sensor data itself.

In addition, the detection unit 14 detects whether a change (activity change) that satisfies a predetermined condition has occurred in the activity feature amount 104 or the activity amount information 102 . Here, it is preferable that the detection unit 14 compares the activity feature quantities 104 to detect whether a change has occurred. Here, the change in activity can be detected by comparing the activity feature amounts 104 for each predetermined period and determining whether the difference is equal to or greater than a threshold. It is also possible to assume that the amount of change per unit time is equal to or greater than a threshold value, that the value is greater than or less than a set value, or that the tendency of the amount of change is unique. Here, it is assumed that the tendency of the amount of change is unique, for example, being different from the average tendency of change. Furthermore, the activity feature amount 104 may be used by aggregating activity amounts for each of a plurality of categories. Then, the detection unit 14 identifies the time when the activity change occurred.

Further, the comparison unit 15 identifies the living behavior feature quantity 103 that has caused a change (living behavior change) that satisfies a predetermined condition, among the living behavior feature quantities 103 corresponding to the changes detected by the detection unit 14 .
For this reason, the comparison unit 15 identifies a period having a predetermined relationship with the time when the change detected by the detection unit 14 occurs. Next, the comparison unit 15 identifies the occurrence of lifestyle behavior change that satisfies a predetermined condition from the activity feature amount 104 or the activity amount information 102 in the identified period. It should be noted that the lifestyle behavior change can be assumed that the amount of change per unit time is greater than or equal to a threshold value, that the value is greater than or less than a set value, or that the tendency of the amount of change is unique. Here, it is assumed that the tendency of the amount of change is unique, for example, being different from the average tendency of change. Then, the comparison unit 15 identifies the living behavior feature quantity 103 in which the living behavior change occurs.

In addition, the change factor estimating unit 16 estimates the factor of the living behavior change of the person to be analyzed 2 according to the living behavior feature amount 103 specified by the comparing unit 15 . For this purpose, the change factor estimation unit 16 uses the index value correspondence table 105 . Note that the index value correspondence table 105 will be described in the item "Information/Data".

In addition, the improvement plan estimation unit 17 estimates improvement plans for living behavior based on the living behavior feature quantity 103 specified by the comparison unit 15 . For this reason, it is desirable that the improvement plan estimation unit 17 uses the living behavior change factors estimated by the change factor estimation unit 16 . At this time, the improvement plan estimation unit 17 also uses the index value correspondence table 105 .

The input/output unit 18 also has a function of connecting to the network 60 and the terminal devices 100-1 and 100-2. Here, the terminal devices 100-1 and 100-2 have input/output functions for the human life traceability device 10. FIG. As an input/output function, it outputs processing results and the like, receives instructions from an administrator, and notifies them to the human life traceability system 10 . Although two terminals 100-1 and 100-2 are shown in FIG. 1, the number of terminals is not limited. Also, the human life traceability device 10 and the terminal devices 100-1 and 100-2 may be connected via an intranet or may be directly connected via a cable or the like. Also, the terminal devices 100-1 and 100-2 may be omitted, or the human life traceability device 10 may be provided with these functions. The terminal devices 100-1 and 100-2 can be implemented by computers such as PCs, tablets, and smartphones.

In addition, the storage unit 19 stores the above-described various information, computer programs, and the like used in the processing of the human life traceability device 10 .

Here, the human life traceability device 10 can be realized by a computer. An example of this implementation is shown in FIG. FIG. 2 is a hardware configuration diagram of the human life traceability device 10 in this embodiment. In FIG. 2, the human life traceability device 10 includes a processing device 110, a network interface 181, an interface 182, a memory 191, and a storage device 192, which are interconnected via a communication path such as a bus.

Here, the processing device 110 can be realized by a processor such as a CPU. Then, according to various programs developed in the memory 191, the processing device 110, which is the functional blocks described above, includes the action estimation unit 11, the action feature amount calculation unit 12, the activity feature amount calculation unit 13, the detection unit 14, the comparison unit 15, The processing in the change factor estimation unit 16 and the improvement plan estimation unit 17 is executed. The details of this process will be described in the item "Processing flow". In the hardware configuration of FIG. 2, each functional block is realized by a software program, but may be realized by dedicated hardware or FPGA (Field-Programmable Gate Array).

Also, the network interface 181 and the interface 182 implement the input/output unit 18 in FIG. That is, the network interface 181 connects with the network 60 . Also, the interface 182 connects with the terminal devices 100-1 and 100-2.

A memory 191 and a storage device 192 correspond to the storage unit in FIG. In the memory 191, various information and programs used for processing in the processing device 110 are loaded.
FIG. 2 illustrates the program. That is, as programs, an action estimation program 111, an action feature amount calculation program 112, an activity feature amount calculation program 113, a detection program 114, a comparison program 115, a change factor estimation program 116, and an improvement plan estimation program 117 are developed in the memory 191. , according to which the processing unit 110 executes processing. Note that each of these programs is stored in a storage medium such as the storage device 192 and expanded to the memory 191 from here.

Note that the correspondence between each program and the function of each unit shown in FIG.
Behavioral Feature Amount Calculation Program 112: Behavioral Feature Amount Calculator 12
Activity feature quantity calculation program 113: activity feature quantity calculator 13
Detection program 114: detection unit 14
Comparison program 115: comparison unit 15
Change factor estimation program 116: change factor estimation unit 16
Improvement plan estimation program 117: Improvement plan estimation unit 17
In addition to the programs described above, the storage device 192 stores various types of information. These pieces of information include the living behavior information 101, the activity amount information 102, the living behavior feature quantity 103, the activity feature quantity 104, and the index value correspondence table 105 described above. Also, it can be realized by a storage such as a hard disk drive. Also, the storage device 192 may be realized by a separate housing from the human life traceability device 10 such as a file server. With this, the description of the human life traceability device 10 is finished, and returning to FIG. 1, the insurance company server 41 and the day service provider server 51 will be described.

The insurance company server 41 is operated by the insurance company and implemented by a computer. That is, it has a storage device 411 and a processing device 412, which are general configurations of a computer. Here, the insurance company server 41 executes processing for development of insurance products, customer management, and health management of policyholders related to this embodiment. The insurance company server 41 is also connected to a terminal device 42 similar to the terminal devices 100-1 and 100-2.

Also, in this embodiment, the day service provider server 51 is operated by the day service provider and implemented by a computer similar to the insurance company server 41 . That is, it has a storage device 511 and a processing device 512, which are general configurations of a computer. Here, the day service provider server 51 executes processing for facility operation management, user management, and user health management related to this embodiment. Also, the day service provider server 51 is connected to the terminal device 52 in the same manner as the insurance company server 41 .

Also, the user terminals 31 and 32 can be realized by computers used by related parties such as relatives of the person to be analyzed 2 . These can be realized by a PC, a smart phone, a tablet, etc., and the analysis result of the analysis target person 2 can be acquired from the human life traceability device 10 and displayed. In addition, it is desirable to be able to output an instruction for the analysis result to the human life traceability device 10 and the user terminal 23 of the person to be analyzed 2 . This completes the description of the configuration of this embodiment.
<Information/Data>
Next, various information and data used in this embodiment will be described. First, FIG. 3 is a diagram showing living activity information 101 used in this embodiment. As shown in FIG. 3, the living activity information 101 is time-series data in which living activity items indicating types of living activities and their times are associated with each other. The living activity information 101 is created by the activity estimation unit 11 based on living activity sensor data. Furthermore, the living activity information 101 is created for each person 2 to be analyzed and stored in the storage device 192 in FIG.

Next, FIG. 4 is a diagram showing the activity amount information 102 used in this embodiment. As shown in FIG. 4, the living activity information 101 is time-series data in which the instantaneous value of the amount of activity is associated with the time. Here, as the activity amount information 102, activity amount sensor data can be used. Also, as the instantaneous value of the amount of activity, the velocity and acceleration related to the motion of the person to be analyzed 2 at that time can be used. Furthermore, as the activity amount information 102, the energy consumption of the person to be analyzed 2 created from the activity amount sensor data may be used. Furthermore, the activity amount information 102 is managed for each person 2 to be analyzed. That is, it is stored in the storage device 192 in FIG.

Next, FIG. 5 is a diagram showing the living behavior feature quantity 103 used in this embodiment. As shown in FIG. 5, the living activity feature amount 103 is data in which a living activity item is associated with an index value of the time or number of times the activity was performed during the period. The living behavior feature quantity 103 is created from the living behavior information 101 by the behavior feature quantity calculation unit 12 . The index value is a total value related to the time or the number of executions, and indicates a representative value such as the average value of the time or number of times the person to be analyzed 2 executes the daily activity item in a period such as one week. One day or the like can be assumed as the unit period. Also, this index value serves as an index that indicates the characteristics of the living behavior feature quantity 103 . In other words, the living activity feature quantity 103 indicates an index value for each living activity item. In addition, the daily activity items are totaled for each index value item such as time and number of times.

It should be noted that the living behavior feature quantity 103 records the period when the corresponding sensor data was detected as a "period". In this example, a specific week such as "5/2-5/8" is described, but other units such as a specific day, day of the week, and month may be used. It should be noted that this “period” can be specified based on the time in the living behavior information 101 .

Furthermore, in this embodiment, it is possible to use a plurality of types of living behavior feature quantities 103 .
The contents of the first to third living activity feature quantities, which are examples of the plurality of living activity feature quantities 103, will be described below.

First, the first living activity feature amount is information indicating an integral value (index value) for each living activity item per day. Next, the second living activity feature amount is information indicating an integrated value (index value) for each living activity item per week. For this reason, the behavior feature amount calculation unit 12 calculates an average of integrated values for each living activity item per day from the first living activity feature amount for one week as the second living activity feature amount.

Next, the third living activity feature amount indicates the ratio of each cluster by living activity item represented by the integral value, which is created based on the first living activity feature amount or the second living activity feature amount. Information. For this reason, the behavior feature amount calculation unit 12 generates a cluster group including a plurality of clusters. Here, the cluster corresponds to the active state of the living activity item, which is affected by the day of the week, the weather, and the like. In other words, the clusters are clustered according to the living behavior of the person to be analyzed 2 and how active they are.

These first to third living behavior feature quantities are created by the behavior feature quantity calculator 12 and stored in the storage device 192 in FIG. It should be noted that how these are used will be described later in the item "processing flow".

Next, FIG. 6 is a diagram showing the activity feature quantity 104 used in this embodiment. As shown in FIG. 6, the activity feature amount 104 is data in which the amount of activity such as movement speed, acceleration, and energy is associated with the integral value of the time (m: minutes) at which the numerical value of the amount of activity was detected during a predetermined period. is. In particular, it is desirable to divide the integrated values into a plurality of categories and aggregate them. In FIG. 6, the movement speed is divided into categories such as 1-5 and 5-10. Also, the activity feature amount 104 is created from the activity amount information 102 by the activity feature amount calculator 13 . Although a plurality of activity amounts are used in this embodiment, the number is not limited. Here, the activity feature quantity 104 records the period when the sensor data was detected as a "period". In this example, "5/2-5/8", which is the same value as the living behavior feature quantity 103, is described. Therefore, other units such as specific days, days of the week, and months may be used. In addition, this “period” can be specified based on the time of the activity amount information 102 .

Furthermore, in this embodiment, it is possible to use a plurality of types of activity feature quantities 104 as well. The contents of the first to fourth activity feature quantities, which are examples of the plurality of activity feature quantities 104, will be described below.

First, the first activity feature amount is information indicating the integrated value of the amount of activity per day. Next, the second activity feature amount is information indicating the integrated value of the activity feature amount per week. For this purpose, the activity feature amount calculator 13 calculates an average of integrated values of the amount of activity per day from the first activity feature amount for one week as the second activity feature amount.

Next, the third activity feature amount is information indicating the respective ratios of the activity feature amount clusters represented by the integral values, which are created based on the first activity feature amount and the second activity feature amount. For this purpose, the activity feature quantity calculator 13 generates a cluster group including a plurality of clusters. Here, the cluster corresponds to the active state of the activity feature amount, which is affected by the day of the week, the weather, and the like. In other words, clusters are obtained by clustering the amount of activity of the person 2 to be analyzed according to how active they are.

Next, the fourth activity feature amount is information indicating the integral value of the activity feature amount per month. For this reason, the behavior feature amount calculation unit 12 uses the second activity feature amount for one month as the fourth activity feature amount, or the average of the integrated values of the activity amount per day from the second activity feature amount Calculate

These first to fourth activity feature quantities are also created by the activity feature quantity calculator 13 and stored in the storage device 192 in FIG. It should be noted that how these are used will also be described later in the item "processing flow".

The first to third living behavior feature amounts and the first to fourth activity feature amounts of this embodiment use weeks or months as a predetermined period, but these are just examples and are not limited to these. Also, although the average is used as the feature amount, other representative values may be used.

Finally, the index value correspondence table 105 will be explained. FIG. 7 is a diagram showing the index value correspondence table 105 used in this embodiment. The index value correspondence table 105 is used by the comparison unit 15 and the improvement plan estimating unit 17 to identify the living behavior information 101 in which a change in living behavior satisfying a predetermined condition has occurred and to estimate an improvement plan. Furthermore, it is desirable that the index value correspondence table 105 is used for estimating the change factor in the change factor estimation unit 16 .

Therefore, in the index value correspondence table 105, a change factor is set together with its index value (item) for each living activity item. This change factor indicates a reference in the index value (item) for specifying the living activity information 101 in the comparison unit 15 . For example, with regard to “going out behavior” in the “life activity item” of the first record in FIG. It is determined that a change has occurred.

Furthermore, the index value correspondence table 105 also associates an improvement plan with each living activity item.
Therefore, the improvement plan estimation unit 17 can estimate the improvement plan using the index value correspondence table 105 . For example, if the "variation factor" is "decrease in outing time (decrease by 10%)", it can be specified that the outing time is increased. In other words, it is possible to identify the message to the analysis target person 2, "The time spent at home is getting longer. Let's increase the opportunities to go out."

In this embodiment, the change factors include the criteria for index values (items), but these criteria may be provided as improvement proposals or other items. In this case, the comparison unit 15 uses the improvement plan or other items to determine the occurrence of lifestyle behavior change. Moreover, the index value correspondence table 105 is not limited to the index value correspondence table 105 shown in FIG.

The information and data used in this embodiment have been described above, and the processing flow of this embodiment will be described below.
<Processing flow>
Next, the processing flow of this embodiment will be described. FIG. 8 is a flow chart showing the processing flow in this embodiment. In the following, FIG. 8 will be described mainly with respect to each configuration (functional block) shown in FIG.

First, in step S1, the input/output unit 18 receives sensor data from sensors. Here, the input/output unit 18 converts the sensor data. As a result, the activity amount sensor data is converted into activity amount information 102 . Note that the input/output unit 18 preferably stores the converted sensor data in the storage unit 19 . Note that the activity amount information 102 has already been described with reference to FIG.

Next, in step S2, the input/output unit 18 sorts the received sensor data into activity amount sensor data and living activity sensor data. Note that steps S1 and S2 may be realized by receiving the activity amount sensor data and the living behavior sensor data by separate reception units, and step S2 may be executed by a configuration different from that of the input/output unit 18. good too.

　As a result of step S2, when the activity amount sensor data is received, the process proceeds to step S3. Moreover, when living activity sensor data is received, it changes to step S4.

Next, in step S3, the activity feature amount calculator 13 creates an activity feature amount 104 that indicates the feature from the activity amount information 102. In addition, it is desirable that the activity feature quantity calculation unit 13 stores the activity feature quantity 104 in the storage unit 19 . Note that the activity feature amount 104 has already been described with reference to FIG. Here, an example in which one type of activity feature quantity 104 is created will be described, and processing using the first to fourth activity feature quantities will be described later.

In addition, in step S4, the behavior feature amount calculator 12 creates living behavior information 101 from the converted living behavior sensor data. The contents of the living activity information 101 have already been explained with reference to FIG.

Next, in step S5, the behavioral feature quantity calculation unit 12 creates a living behavioral feature quantity 103 representing the characteristics from the living behavioral information 101. Moreover, it is desirable that the behavior feature amount calculation unit 12 stores the living behavior feature amount 103 in the storage unit 19 . Note that the living behavior feature amount 103 has already been described with reference to FIG. Also, here, an example in which one type of living activity feature quantity 103 is created will be described, and processing using the above-mentioned first to third living activity feature quantities will be described later.

Next, in step S6, the detection unit 14 detects the activity change of the activity feature quantity 104 and the time of occurrence thereof. For this reason, the detection unit 14 extracts the activity feature quantity 104 at predetermined intervals. Then, the detection unit 14 compares the extracted activity feature amounts 104 for each predetermined period, and determines whether the difference is equal to or greater than the threshold. As a result, when it is equal to or greater than the threshold, the detection unit 14 determines that an activity change has occurred, and detects the occurrence time.

For example, in step S6, when the predetermined period is "month", the following processing is performed.
First, the detection unit 14 extracts the activity feature amount 104 of the analysis target month. In the case of the activity feature amount 104 shown in FIG. 6, one week is created as one record, so the detection unit 14 extracts the activity feature amount 104 for one month=four weeks (four records).

In addition, the detection unit 14 extracts the past activity feature amount 104 from at least the previous month of the analysis target month as the comparison target activity feature amount. For example, when the analysis target month is May, the detection unit 14 extracts the activity feature amount 104 for April as the comparison target activity feature amount. At this time, the detection unit 14 may calculate a representative value such as an average value of the activity feature amounts 104 for a plurality of months, and use this as the comparison target activity feature amount.

Next, the detection unit 14 compares the activity feature amount 104 of the analysis target month with the comparison target activity feature amount. When the detection unit 14 detects a change in activity because the comparison result is equal to or greater than the threshold value, the detection unit 14 specifies the time of occurrence. As the occurrence time, "week", "day", "day of the week", "date and time", etc. can be assumed, but it is desirable that the predetermined period is a unit shorter than "one month". In particular, it is desirable to set the number of records of the activity feature amount 104=week. In addition, in this step, it is only necessary to be able to detect changes in activity based on the activity amount information 102, and using the activity feature amount 104 is an example.

As a result of step S6, if a change in activity is detected (YES), the process proceeds to step S7.
If no activity change is detected (NO), the process proceeds to step S1.

Next, in step S7, the comparison unit 15 compares the living activity feature quantities 103 corresponding to the activity changes detected in step S6 to detect the occurrence of living activity changes. For this purpose, first, the comparison unit 15 identifies a target period having a predetermined relationship with the occurrence time of the activity change detected in step S6. Next, the comparison unit 15 identifies a comparison period such as the period preceding the target period. Then, the comparison unit 15 compares the living activity feature amounts 103 of the target period and the comparison period for each living activity item. As a result, when the difference satisfies a predetermined condition, the comparison unit 15 determines that a change in living behavior has occurred and specifies the living behavior item.

Here, the target period is preferably a certain period including the time when the activity change occurs, especially the period before the time of occurrence. Here, the reason why it is before the time of occurrence is that the change in living behavior is caused by the change in the amount of activity. However, since these causal relationships are various, it may include after the time of occurrence or both, or may be shared with the time of occurrence (period). Thus, it is desirable that the target period and the comparison period are adjacent periods within a predetermined range.

Also, whether or not the difference satisfies a predetermined condition is determined using the index value correspondence table 105 shown in FIG. For example, when the "daily activity item" is "going out", the target period is compared with the comparison period, and if the period is "10% decrease", it is determined that the predetermined condition is satisfied.

In addition, in step S7, the comparison unit 15 may compare the living activity information 101 corresponding to the activity change detected in step S6 to detect the occurrence of the living activity change. In this way, in this step, it is sufficient to be able to detect a change in living activity based on the living activity information 101, and using the living activity feature quantity 103 is an example. That is, in this step, it suffices if information on living activities can be compared.

Next, in step S8, the comparison unit 15 determines whether or not the processing of step S7, that is, the comparison of the living activity feature quantities 103 has been completed for each living activity item. For this purpose, the comparison unit 15 uses the living activity items of the living activity feature quantity 103 shown in FIG. As a result, if the processing for each living activity item of the living activity characteristic quantity 103 has been completed (YES), the process proceeds to step S9. On the other hand, if it has not been completed yet (NO), the process proceeds to step S7, and the remaining living activity items are processed.

Next, in step S9, the change factor estimation unit 16 estimates the change factor for the living activity item specified in step S7. For this purpose, the change factor estimation unit 16 searches the index value correspondence table 105 for a change factor corresponding to the identified living activity item. When the above-mentioned "daily activity item" is "going out", it is estimated that the change factor is "decrease in going out time".

Next, in step S10, the improvement plan estimation unit 17 estimates an improvement plan according to the living action item identified in step S7 and/or the change factor estimated in step S9. For this purpose, the improvement plan estimation unit 17 uses the index value correspondence table 105 . For example, if the “daily activity item” is “going out” and/or the “change factor” is “decrease in going out,” the improvement plan estimating unit 17 selects “the time spent at home is increasing. Let's estimate. Thus, in this embodiment, information for improving living behavior is output as an improvement plan. In this embodiment, one index value correspondence table 105 is used in steps S7, S9 and S10, but this may be provided separately for each function.

Here, the improvement plan estimation by the improvement plan estimation unit 17 may be performed as follows.
(1) The improvement plan estimating unit 17 estimates an improvement plan according to the change factor based on data analysis using AI (artificial intelligence).
(2) The improvement plan estimating unit 17 estimates an improvement plan corresponding to the change factor, which is the condition, by a base approach such as the IF-THEN rule.
(3) The improvement plan estimating unit 17 selects from a plurality of improvement plans by performing filtering processing according to the amount of difference, which is the comparison result of the living behavior feature quantity 103 in step S7.

It should be noted that, in this embodiment, improvement proposals may be estimated by using the index value correspondence table 105 or by combining (1) to (3).

Then, the improvement plan estimation unit 17 outputs the improvement plan to the terminal devices 100-1 and 100-2 and the network 60 via the input/output unit 18. As a result, the user terminals (23, 31, 32), the insurance company server 41, the terminal device 42, the day service provider server 51, and the terminal device 52 receive the improvement plan. As a result, the terminal devices 100-1 and 100-2, the terminal device 42, and the terminal device 52 can display this improvement plan. Further, the improvement plan estimating unit 17 may also be configured to output the factor of change and display the factor of change on each device.

Here, FIG. 9 is a diagram showing a display example in this embodiment. This display may be common to the terminal devices 100-1, 100-2, user terminals (23, 31, 32), terminal device 42, and terminal device 52, or may be customized. In FIG. 9, the following information is displayed in the display screen 200 . These pieces of information may be created by the improvement plan estimating section 17, or may be created by using the information from the improvement plan estimating section 17 on the terminal side.

On the display screen 200, it is specified that the person to be analyzed is 2 (Mr. XX), and a change factor and improvement plan column 201, a period column 202, an activity amount display column 203, a living activity comparison result display column 204, and a graph display. A column 205 is included.

First, the change factors and improvement plans estimated in steps S9 and S10 are displayed. Also, the period column 202 indicates the period to be analyzed, that is, the period detected by the sensor, and the period compared by the comparison unit 15 is indicated by a triangle. In this figure, April (fourth week) and May (first week) are compared. Here, May (first week) is the target period, and April (fourth week) is the comparison period.

In addition, the activity amount display column 203 displays information obtained by graphing the activity amount information 102 . Here, the bar graph is displayed in a form different from that in FIG. 4, but other forms such as a line graph may be used.

In addition, the living activity comparison result display column 204 displays “living activity item”, “index value (item)”, “living activity change (variation amount)”, and living activity features of the target period and the comparison period with respect to the living activity feature amount 103 . A quantity 103 is displayed. Here, the “change in living behavior (amount of change)” indicates a difference that is the result of comparison by the comparison unit 15 between the target period and the comparison period.

Furthermore, the graph display column 205 is information that graphs the comparison result of the living activity feature quantity 103 and the like.

Note that the display screen 200 is merely an example, and can be customized in various ways.
In particular, health care providers may display personal information confidentially, omit the position column, or change the display position. In particular, the change factor and improvement plan column 201 may be configured to display either the change factor or the improvement plan.

As a result, the parties concerned can confirm the improvement plan for the analysis subject 2 and consider revising their living behavior. Furthermore, in health-related services, each server can be used to manage subscribers and propose new services.

This completes the description of the processing flow of the present embodiment, and a modified example of the feature amount will be described below.
<Modification>
As described above, in this embodiment, the first to third living action feature amounts can be used as the living action feature amount 103, and the first to fourth activity feature amounts can be used as the activity feature amount 104. FIG. Modified example processing (steps S3, S5, S6, and S7) relating to these will be described below.

FIG. 10 is a diagram for explaining the concept of processing in a modified example of this embodiment. FIG. 10(a) shows the concept of creating the first to fourth activity feature amounts for the activity amount information 102 (step S3). The activity feature amount calculator 13 acquires and stores the activity amount information 102 as information for each time. Then, the activity feature quantity calculator 13 creates a first activity feature quantity, which is an integrated value per day. That is, the first activity feature amount is generated in units of days.

As a result, the first activity feature amount for each day of the week is obtained as shown in the second row of FIG. 10(a). Then, the activity feature amount calculator 13 generates a second activity feature amount as an average value for each week. As a result, as shown in the third row of FIG. 10(a), the second activity feature amount for each week is obtained.

In addition, the activity feature amount calculation unit 13 creates a cluster group corresponding to the activity amount from the first activity feature amount for each day of the week. This results in the classification into "clusters with a large amount of activity" and "clusters with a small amount of activity" in FIG. 10(a). This allows for classification according to the day of the week, weather, season and impact. As a result of the above, the activity feature amount calculator 13 creates the third activity feature amount indicating the cluster ratio for each week.

Then, the activity feature amount calculator 13 generates a fourth activity feature amount as an average value of the second activity feature amounts for each week. In other words, the fourth activity feature amount is generated on a monthly basis.

Next, FIG. 10(b) shows the concept of creating the first to third living activity feature amounts as the living activity feature amount 103 (step S5). The behavior feature amount calculation unit 12 creates a first living activity feature amount, which is an integrated value for each activity (for each living activity item) per day, from the living activity information 101 for each time. In other words, the first living behavior feature quantity is generated in units of days.

As a result, as shown in the second row of FIG. 10(b), the first living activity feature amount for each day of the week is obtained. Then, the behavior feature amount calculation unit 12 generates a second living behavior feature amount as the average value for each week.

Also, the behavior feature amount calculation unit 12 creates a cluster group corresponding to the activity amount from the first activity feature amount for each day of the week. This results in classification into the "cluster with a lot of sleep", the "cluster with little sleep", the "cluster with a lot of housework", and the "cluster with a little housework" in FIG. 10(b). This allows for classification according to the day of the week, weather, season and impact. As a result, as shown in the third row of FIG. 10(b), the second living behavior feature quantity for each week is obtained.

Then, the behavior feature quantity calculation unit 12 creates a third activity feature quantity indicating the cluster ratio for each week and for each behavior from the second living behavior feature quantity. As described above, in steps S3 and S5, the first to third living activity feature amounts and the first to fourth living activity feature amounts are created. The first to third living behavior feature amounts and the first to fourth activity feature amounts described above are stored in the storage unit 19 .

Next, the processing in steps S6 and S7 will be described. FIG. 11 is a flow chart showing the processing flow of steps S6 and S7 in this modification. Here, steps S61 to S64 correspond to step S6, and steps S71 to S74 correspond to step S7.

First, in step S61, the detection unit 14 detects the activity change of the fourth activity feature amount and the time of occurrence thereof. This detection method is the same as that described in FIG. As a result, if a change in activity is detected (YES), the process proceeds to step S71. If no activity change is detected (NO), the process proceeds to step S62.

Next, in step S71, the comparison unit 15 compares the third living behavior feature amounts corresponding to the activity changes detected in step S61 to detect the occurrence of living behavior changes. This detection method is the same as that described in FIG. Then, this process is terminated, and the process proceeds to step S8.

Also, in step S62, the detection unit 14 detects the activity change of the third activity feature amount and the time of occurrence thereof. This detection method is the same as in step S61, and is the same in steps S63 and S64. As a result, if a change in activity is detected (YES), the process proceeds to step S72. If no activity change is detected (NO), the process proceeds to step S63.

Next, in step S72, the comparison unit 15 compares the second living behavior feature values corresponding to the activity changes detected in step S62 to detect the occurrence of living behavior changes. This detection method is the same as in step S71, and is the same in steps S63 and S64. Then, this process is terminated, and the process proceeds to step S8.

Also, in step S63, the detection unit 14 detects the activity change of the second activity feature amount and the time of occurrence thereof. As a result, if a change in activity is detected (YES), the process proceeds to step S73. If no activity change is detected (NO), the process proceeds to step S64.

Next, in step S73, the comparison unit 15 compares the first living activity feature amounts corresponding to the activity changes detected in step S63, and detects the occurrence of living activity changes. Then, this process is terminated, and the process proceeds to step S8.

Also, in step S64, the detection unit 14 detects the activity change of the first activity feature amount and the time of occurrence thereof. As a result, if a change in activity is detected (YES), the process proceeds to step S74. If no change in activity is detected (NO), this process is terminated. Processing after step 8 can be omitted.

Next, in step S74, the comparison unit 15 compares the living activity information 101 corresponding to the activity change detected in step S64, and detects the occurrence of the living activity change.
Then, this process is terminated, and the process proceeds to step S8.

As described above, in this embodiment, a plurality of feature amounts with different target periods are created, and the processes of steps S6 and S7 are executed in multiple stages in the order of the longest period. In particular, the comparison process is performed by the comparison section 15 according to the result of the detection section 14 . By performing such processing, it is possible to detect changes in the person to be analyzed from a long-term perspective. Note that the periods (days, weeks, months) of the first to third living activity feature amounts and the first to fourth living activity feature amounts are just examples and are not limited to these. Also, the number of these feature amounts is not limited to the number illustrated.

This completes the description of this embodiment, but the present invention is not limited to this embodiment, and various customizations are possible. For example, the human life traceability device 10 may be configured as a stand-alone system, so that a person concerned can confirm the processing results of the human life traceability device 10 . Furthermore, the activity amount information 102 and the activity feature amount 104 may be specified by combining a plurality of activity amounts. In this case, in addition to combining different amounts of activity such as speed and acceleration, different amounts of activity of the same type (velocities) may be combined. Furthermore, when an improvement proposal is output, it is desirable that the human life traceability device 10 periodically confirms changes in the living activity item for which the improvement proposal is output (changes in living behavior) before and after the output.

10: human life traceability device, 11: behavior estimation unit, 12: behavior feature amount calculation unit, 13: activity feature amount calculation unit, 14: detection unit, 15: comparison unit, 16: change factor estimation unit, 17: improvement Proposal estimation unit 18: input/output unit 19: storage unit 2: person to be analyzed 20: home 21-1, 21-2: activity amount sensor 22-1, 22-2: living activity sensor 23 , 31, 32: user terminal, 41: insurance company server, 411: storage device, 412: processing device, 42: terminal device, 51: day service provider server, 511: storage device, 512: processing device 52: terminal device , 60: network

Claims (10)

1. In the human life traceability device that manages and analyzes the activities and behavior of the person to be analyzed,
   Stores living behavior information, which is time-series data indicating the living behavior of the person to be analyzed, and activity amount information, which is time-series data indicating the amount of activity of the person to be analyzed, based on sensor data for the person to be analyzed from a sensor. a storage unit for
   a detection unit that detects a change in activity that satisfies a predetermined condition based on the activity amount information;
   when the change in activity is detected, comparing the information on the living activity according to the living activity information in a period having a predetermined relationship with the timing when the activity change occurs,
   a comparing unit that specifies information about living behaviors in which changes in living behaviors satisfying a predetermined condition have occurred as a result of the comparison;
   A human life traceability device having an improvement plan estimating unit for estimating improvement plans for living activities based on the identified information on the living activities.
2. The human life traceability device according to claim 1,
   further comprising a change factor estimating unit for estimating a factor of the lifestyle behavior change according to the identified information about the lifestyle behavior;
   The improvement plan estimation unit is a human life traceability device for estimating the improvement plan for the living behavior corresponding to the estimated factor of the change in living behavior.
3. In the human life traceability device according to claim 2,
   Furthermore, an activity feature amount calculation unit for calculating an activity feature amount indicating the feature of the activity amount information and a living activity feature amount calculation unit for calculating the living activity feature amount indicating the feature of the living activity information,
   The detection unit detects an activity change in the activity feature amount,
   The comparison unit is a human life traceability device that compares the living behavior feature quantities for the period.
4. In the human life traceability device according to claim 3,
   a behavior estimating unit for estimating a living activity item indicating a living activity and an index value thereof as the living activity information from the sensor data;
   The storage unit stores, as the living activity information, living activity items of the person to be analyzed and index values of the living activities,
   The living activity feature amount calculation unit calculates the living activity feature amount having the living activity item and the index value of the living activity,
   The comparison unit identifies a difference between a living activity item and its index value as the living activity change,
   The change factor estimating unit is a human life traceability device for estimating an improvement plan for the living activity corresponding to the difference between the specified living activity item and its index value.
5. In the human life traceability device according to claim 4,
   The activity feature quantity calculator,
   Calculate the daily activity integral value as the first activity feature value,
   calculating an average value of the amount of activity for each week from the first activity feature amount as a second activity feature amount, generating a cluster group from the first activity feature amount, and days and percentages belonging to the generated cluster group as the third activity feature,
   calculating a monthly integral value from the second activity feature amount as a fourth activity feature amount;
   The living behavior feature quantity calculation unit is
   Calculate the time integral value for each daily living activity item as a first living activity feature quantity,
   calculating a time integral value for each living activity item on a weekly basis from the first living activity feature amount as a second living activity feature amount;
   generating a cluster group for each living activity item from the first living activity feature amount, calculating the days and percentages belonging to the generated cluster group as a third living activity feature amount,
   The detection unit is
   Detecting whether or not the activity has changed by comparing the fourth activity feature amounts for the period,
   If the activity change is not detected, comparing the target third activity feature amounts to detect the presence or absence of the activity change,
   Furthermore, if the activity change is not detected, comparing the target second activity feature values to detect the presence or absence of the activity change,
   Furthermore, if the activity change is not detected, comparing the target first activity feature amounts to detect the presence or absence of the activity change,
   The comparison unit selects any one of the third activity feature amounts, the second activity feature amounts, and the first activity feature amounts, according to the activity change detection result of the detection unit. A human life traceability device that performs comparisons.
6. In an improvement plan proposal support method using a human life traceability device that manages and analyzes the activities and behavior of the person to be analyzed,
   Living behavior information, which is time-series data indicating the living behavior of the person to be analyzed, and activity, which is time-series data indicating the amount of activity of the person to be analyzed, based on sensor data for the person to be analyzed from a sensor in the storage unit store quantity information,
   A detection unit detects a change in activity that satisfies a predetermined condition based on the activity amount information;
   By the comparison part,
   when the change in activity is detected, comparing the information on the living activity according to the living activity information in a period having a predetermined relationship with the timing when the activity change occurs,
   As a result of the comparison, identifying information about living behavior that has caused a change in living behavior that satisfies a predetermined condition,
   An improvement plan proposal support method for estimating an improvement plan for a living activity based on the identified information on the living activity by an improvement plan estimating unit.
7. In the improvement plan proposal support method according to claim 6,
   Furthermore, a change factor estimating unit estimates a factor of the living behavior change according to the specified information about the living behavior,
   An improvement plan proposal support method for estimating, by the improvement plan estimating unit, an improvement plan for the living behavior corresponding to the estimated factor of the living behavior change.
8. In the improvement plan proposal support method according to claim 7,
   moreover,
   calculating an activity feature amount indicating the feature of the activity amount information by the activity feature amount calculating unit, calculating a living activity feature amount indicating the feature of the living activity information by the living activity feature amount calculating unit,
   The detection unit detects an activity change in the activity feature quantity,
   An improvement plan proposal support method in which the comparison unit compares the living activity feature amounts for the period.
9. In the improvement plan proposal support method according to claim 8,
   Furthermore, a behavior estimating unit estimates a living activity item indicating a living activity and an index value thereof as the living activity information from the sensor data,
   storing in the storage unit, as the living activity information, the living activity items of the person to be analyzed and index values of the living activities;
   calculating the living activity feature amount having the living activity item and the index value of the living activity by the living activity feature amount calculation unit;
   The comparison unit specifies a difference between a living activity item and its index value as the living activity change,
   An improvement proposal support method for estimating, by the change factor estimating unit, an improvement plan for the living behavior corresponding to the identified living behavior item and the difference between the index values thereof.
10. In the improvement plan proposal support method according to claim 9,
    By the activity feature quantity calculation unit,
    Calculate the daily activity integral value as the first activity feature value,
    calculating an average value of the amount of activity for each week from the first activity feature amount as a second activity feature amount, generating a cluster group from the first activity feature amount, and days and percentages belonging to the generated cluster group as the third activity feature,
    calculating a monthly integral value from the second activity feature amount as a fourth activity feature amount;
    By the living activity feature quantity calculation unit,
    Calculate the time integral value for each daily living activity item as a first living activity feature quantity,
    calculating a time integral value for each living activity item on a weekly basis from the first living activity feature amount as a second living activity feature amount;
    generating a cluster group for each living activity item from the first living activity feature amount, calculating the days and percentages belonging to the generated cluster group as a third living activity feature amount,
    By the detection unit,
    Detecting whether or not the activity has changed by comparing the fourth activity feature amounts for the period,
    If the activity change is not detected, comparing the target third activity feature amounts to detect the presence or absence of the activity change,
    Furthermore, if the activity change is not detected, comparing the target second activity feature values to detect the presence or absence of the activity change,
    Furthermore, if the activity change is not detected, comparing the target first activity feature amounts to detect the presence or absence of the activity change,
    According to the detection result of the activity change by the detection unit, the comparison unit selects any one of the third activity feature amounts, the second activity feature amounts, and the first activity feature amounts. Improvement proposal support method for performing comparison.

Images