Prediction device, prediction method, and program for prediction device, and prediction system

Abstract

To provide a prediction device, a prediction method, a program for the prediction device, and a prediction system that improve prediction accuracy of reactivity of an object person regarding an intervention.SOLUTION: Data showing a type of the health supporting intervention received by an object person T, and object person reaction curve data measuring a reaction of the object person regarding the intervention in a specific measurement period from an intervention start time point are acquired (S1). Then, selected reaction curve data is selected from among reaction curve data of another person P excluding the object person, in accordance with the type of the intervention received by the object person, the object person basic data, and the object person reaction curve data (S2-S6). Further, a prediction reaction curve of the object person regarding the intervention is predicted (S7), for a period subsequent to the specific measurement period of the object person, based on the selected reaction curve data.SELECTED DRAWING: Figure 12

Description

[Technical field]

[0001] The present invention relates to the technical fields of a prediction device, a prediction method, a program for a prediction device, and a prediction system for predicting a subject's responsiveness to an intervention. [Background technology]

[0002] In the field of health care, a system for supporting health has been developed to improve health. For example, Patent Document 1 discloses an analysis system having a health transition prediction unit that acquires health information of a candidate who will undergo an intervention, acquires time-series health information acquired from the start of the intervention of an individual who has undergone the intervention from a storage unit as a health transition with intervention, predicts the time-series health information of the candidate when the candidate starts the intervention as a predicted transition with intervention based on the candidate's health information and the health transition with intervention, and stores the predicted predicted transition with intervention in the storage unit, a duration prediction unit that uses the predicted transition with intervention by a processor to predict a duration indicating a period during which the candidate's health will improve as an effect of the intervention, and stores the duration in the storage unit, and a support recipient selection unit that selects a recipient to whom the processor will perform the intervention from the candidates based on the duration. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2016-218966 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, in conventional technologies such as Patent Document 1, when an intervention is implemented, the subject may be affected and begin to undergo various changes, and prediction of the subject's reactivity based solely on health information such as basic personal attributes does not provide sufficient prediction accuracy.

[0005] Therefore, an example of the objective of the present invention is to provide a prediction device or the like that improves the prediction accuracy of a subject's responsiveness to an intervention. [Means for solving the problem]

[0006] In order to solve the above problem, the invention described in claim 1 is characterized by comprising a subject data acquisition means for acquiring basic data on the subject including the subject's attributes, data indicating the type of health-supporting intervention received by the subject, and subject response curve data measuring the subject's response to the intervention during a specific measurement period from the start of the intervention, a selective response curve data selection means for selecting selected response curve data from the response curve data of the other person in the storage means in accordance with the type of intervention received by the subject, the basic data of the subject, and the subject response curve data, by referring to a storage means that associates basic data on others other than the subject, the type of intervention, and the response curve data, and a response curve prediction means for predicting the subject's predicted response curve to the intervention after the specific measurement period of the subject based on the selected response curve data.

[0007] The invention described in claim 11 includes a subject data acquisition step in which a subject data acquisition means acquires basic data on the subject including attributes of the subject, data indicating a type of intervention to support health received by the subject, and subject response curve data measuring the response of the subject to the intervention during a specific measurement period from the start of the intervention; The method includes a selective response curve data selection step in which the selective response curve data selection means refers to a storage means which associates basic data, a type of intervention, and response curve data relating to others other than the subject, and selects selective response curve data from the response curve data of the other person in the storage means according to the type of intervention received by the subject, the basic data of the subject, and the subject's response curve data, and a response curve prediction step in which the response curve prediction means predicts the subject's predicted response curve to the intervention after the specific measurement period of the subject based on the selected response curve data.

[0008] The invention described in claim 12 is characterized in that the computer functions as a subject data acquisition means for acquiring basic data on the subject including the subject's attributes, data indicating the type of health-supporting intervention received by the subject, and subject response curve data measuring the subject's response to the intervention during a specific measurement period from the start of the intervention, a selective response curve data selection means for selecting selected response curve data from the response curve data of the other person in the storage means in accordance with the type of intervention received by the subject and the subject response curve data, by referring to a storage means that associates basic data on the subject, the type of intervention, and the response curve data, and a response curve prediction means for predicting a predicted response curve of the subject to the intervention after the specific measurement period of the subject based on the basic data of the subject and the selected response curve data.

[0009] The invention described in claim 13 provides a prediction system comprising a terminal device of a subject receiving an intervention to support health, and a prediction device which measures and predicts the subject's response to the intervention, wherein the prediction device comprises a subject data acquisition means which acquires from the terminal device basic data on the subject including attributes of the subject, data indicating the type of intervention received by the subject, and subject response curve data which measures the subject's response to the intervention during a specific measurement period from the start of the intervention, a selected response curve data selection means which refers to a storage means which associates basic data on others other than the subject, the type of intervention, and the response curve data, and selects selected response curve data from the response curve data of the other person in the storage means according to the type of intervention received by the subject, the basic data of the subject, and the subject response curve data, and a response curve prediction means which predicts the subject's predicted response curve to the intervention after the specific measurement period of the subject based on the selected response curve data. Effect of the Invention

[0010] According to the present invention, data indicating the type of health-supporting intervention received by a subject and subject response curve data measuring the subject's response to the intervention during a specific measurement period from the start of the intervention are obtained, and selective response curve data is selected from response curve data of persons other than the subject based on the type of intervention received by the subject, the subject's basic data, and the subject response curve data, and a predicted response curve of the subject to the intervention after the specific measurement period of the subject is predicted based on the selected response curve data, thereby improving the accuracy of prediction of the subject's responsiveness to the intervention. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a schematic diagram illustrating an example of a schematic configuration of a prediction system according to an embodiment. [Diagram 2] 2 is a block diagram showing an example of a schematic configuration of an information processing server device in FIG. 1. [Diagram 3] FIG. 3 is a diagram showing an example of data stored in the outcome database of FIG. 2. [Figure 4] FIG. 3 is a diagram showing an example of data stored in an intervention information database of FIG. 2. [Diagram 5] FIG. 1 is a schematic diagram showing an example of classification of types of intervention. [Figure 6] 3 is a diagram showing an example of data stored in a user basic database of FIG. 2. FIG. [Figure 7] FIG. 3 is a diagram showing an example of data stored in the reaction history database of FIG. 2. [Figure 8] FIG. 13 is a schematic diagram showing an example of a type of user intervention, basic data, and a response curve. [Figure 9] 2 is a block diagram showing an example of a schematic configuration of the mobile terminal device of FIG. 1. [Figure 10] 2 is a block diagram showing an example of a schematic configuration of the wearable terminal device of FIG. 1. [Figure 11] FIG. 13 is a schematic diagram showing an example of a screen displayed on a mobile terminal device. [Figure 12]13 is a flowchart showing an example of the operation of the information processing server device. [Figure 13] 10 is a schematic diagram showing an example of the type of intervention, basic data, and response curve of a target user. FIG. [Figure 14] FIG. 13 is a schematic diagram illustrating an example of a predicted response curve of a target user. [Figure 15A] FIG. 13 is a schematic diagram showing an example of a comparison of prediction accuracy. [Figure 15B] FIG. 13 is a schematic diagram showing an example of a comparison of prediction accuracy. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Note that the embodiment described below is an embodiment in which the present invention is applied to a prediction system.

[0013] [1. Overview of the prediction system configuration and functions] First, the configuration of a prediction system 1 according to this embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram showing an example of a schematic configuration of the prediction system 1 according to this embodiment.

[0014] As shown in FIG. 1, the prediction system 1 includes an information processing server device 10 (an example of a prediction device) that acquires data measuring the response of a target user T (an example of a subject) to an intervention to support health and predicts the response, a mobile terminal device 20 that transmits data of the target user T to the information processing server device 10, a wearable terminal device 30 worn by the target user T, and a supporter terminal device 40 used by a support user S that provides health advice to the target user T.

[0015] Here, the intervention is a digital approach using an app or a human approach such as an expert, with the intention of causing a change in health-related behavior or lifestyle. For example, digital approaches include displaying the number of steps and weight on a smartphone or web app, supporting goal setting, planning an exercise plan, and evaluating the degree of achievement. The goal setting support is to have the user decide the number of steps to achieve each day on the app. Planning an exercise plan is to have the user input a plan such as "walk more on weekends." The degree of achievement is evaluated by displaying a review of the week on the display unit of the mobile terminal device 20 or the like. Examples of human approaches include health guidance during health checkups, personal training at gyms, and health support workshops at companies and local governments. Examples of workshops include frailty prevention and healthy eating courses.

[0016] These actions cause changes in the behavior and lifestyle of the target user T. The changes in behavior and lifestyle are observed by measuring outcomes, that is, variables that are desired to be changed by intervention. Outcome variables are variables such as physical activity amount, nutritional state amount, physical state amount, sleep quality, and psychological state amount. More specifically, for physical activity amount, the number of steps and the number of stairs climbed; for nutritional state amount, the number of calories, number of meals, meal time, and number of chews; for physical state amount, the weight value, heart rate, and blood pressure value; for sleep quality, the level indicating the depth of sleep and its duration; and for psychological state amount, the stress level, etc.

[0017] Interventions are designed to induce or reduce a particular behavior, and a change in that behavior will result in a quantified change in the outcome.

[0018] The information processing server device 10, the mobile terminal device 20, the supporter terminal device 40, etc. are capable of transmitting and receiving data to and from each other via a network N, for example, using a communication protocol such as TCP / IP. The network N is constructed, for example, by the Internet.

[0019] The network N may be constructed using a dedicated communication line, a mobile communication network, a gateway, etc. The network N may also have an access point Ap. The mobile terminal device 20, the wearable terminal device 30, etc. may be connected to the network N via the access point Ap.

[0020] The information processing server device 10 has computer functions. The information processing server device 10 acquires basic data on each target user T, data indicating the type of intervention for supporting health that each target user T received, and subject response curve data measuring the response of each target user T to the intervention. The information processing server device 10 calculates and outputs a predicted response curve of each target user T to the intervention based on a database constructed in advance.

[0021] The initial database of the information processing server device 10 is constructed based on basic data from the results of a questionnaire survey of multiple users P (an example of others other than the subject) and data measuring the values ​​of each variable of the outcome of multiple users P in response to the intervention.

[0022] Here, the basic data is data quantified from answers to questions about the user's personal attributes, the user's environmental factors, the user's psychological characteristics, the user's mental and physical condition, etc. Personal attributes include, for example, age and gender. Environmental factors include, for example, social support, the physical environment in which exercise is performed, the home environment such as the family's understanding of exercise, and the environment of a place near the home where exercise can be performed. Psychological characteristics include, for example, personality and cognitive function. Mental and physical conditions include, for example, lifestyle habits, behavioral tendencies, health status, etc. Depending on the questionnaire item, the questions are set so that answers can be scored on a 5- or 7-point scale, etc.

[0023] The response curve data is data for each individual that measures the history of response to an intervention. Specifically, when the measurement is performed using the wearable terminal device 30, the response curve data is time series data such as the user's number of steps, number of stairs climbed, body temperature, heart rate, level of sleep depth calculated from heart rate fluctuations, and stress level. When the measurement is performed using a weight scale, the response curve data is time series data of weight values ​​recorded every day. In the case of diet, the response curve data is time series data such as the number of calories eaten, the number of meals, and meal times recorded each time. The response curve data may also be time series data of test results such as blood tests.

[0024] The mobile terminal device 20, which is an example of a terminal device, has computer functions. The mobile terminal device 20 is, for example, a smartphone or a tablet terminal. The target user T uses the mobile terminal device 20 to input answers to questionnaires, etc. The mobile terminal device 20 transmits data measured by the wearable terminal device 30 and data input by the target user T to the information processing server device 10.

[0025] The wearable terminal device 30, which is an example of a terminal device, has computer functions. The wearable terminal device 30 is, for example, a wristband-type wearable computer. The wearable terminal device 30 has various sensors and measures the number of steps, heart rate, body temperature, etc. The mobile terminal device 20 and the wearable terminal device 30 can communicate with each other via wireless communication.

[0026] The supporter terminal device 40, which is an example of a terminal device, has computer functions. The supporter terminal device 40 is, for example, a personal computer, a laptop computer, a tablet terminal, or the like. The support user S uses the supporter terminal device 40 to input basic data of the target user T, data such as the results of a medical interview with the target user T, and the like. For example, the support user S is a public health nurse, a health counselor (in a health insurance association, etc.), or a gym trainer. Note that, in the case where the target user T inputs data or plans to change his / her behavior independently without the support of the support user S, the prediction system 1 does not need to include the supporter terminal device 40.

[0027] [2. Configuration and functions of the information processing server device and each terminal device] (2.1 Configuration and Functions of Information Processing Server Device 10) Next, the configuration and functions of the information processing server device 10 will be described with reference to FIGS.

[0028] FIG. 2 is a block diagram showing an example of a schematic configuration of the information processing server device 10. FIG. 3 is a diagram showing an example of data stored in the outcome database 12a. FIG. 4 is a diagram showing an example of data stored in the intervention information database 12b. FIG. 5 is a schematic diagram showing an example of classification of types of intervention. FIG. 6 is a diagram showing an example of data stored in the user basic database 12c. FIG. 7 is a diagram showing an example of data stored in the response history database 12d. FIG. 8 is a schematic diagram showing an example of the type of user intervention, basic data, and response curve.

[0029] 2, the information processing server device 10 includes a communication unit 11, a storage unit 12, an output unit 13, an input unit 14, an input / output interface unit 15, and a control unit 16. The control unit 16 and the input / output interface unit 15 are electrically connected via a system bus 17. The information processing server device 10 also has a clock function.

[0030] The communication unit 11 is electrically or electromagnetically connected to the network N and controls the state of communication with the mobile terminal device 20 and the like.

[0031] The storage unit 12 is configured, for example, with a hard disk drive, a solid state drive, etc. The storage unit 12 stores the classification of the intervention, data on the contents of the outcome which is a variable to be changed by the intervention, basic data such as the user's personal attributes, environmental factors, psychological characteristics, etc. The storage unit 12 also stores various programs such as an operating system and a server program, various files, etc. The various programs, etc. may be acquired, for example, from another server device, etc. via the network N, or may be recorded on a recording medium and read via a drive device.

[0032] The memory unit 12 also has an outcome database 12a (hereinafter referred to as "outcome DB12a"), an intervention information database 12b (hereinafter referred to as "intervention information DB12b"), a user basic information database 12c (hereinafter referred to as "user basic information DB12c"), a response history database 12d (hereinafter referred to as "response history DB12d"), etc. The memory unit 12 is an example of a storage means that associates basic data on people other than the subject, types of intervention, and response curve data.

[0033] As shown in FIG. 3, the outcome DB 12a stores an outcome type ID, outcome content, etc. in association with an outcome ID indicating each outcome variable. When the outcome ID of an outcome variable indicates "number of steps", the outcome type is "amount of physical activity" and the outcome content is "number of steps". When the outcome ID of an outcome variable indicates "weight", the outcome type is "amount of physical activity" and the outcome content is "number of steps". The outcome type ID may be entered in the upper digit of the outcome ID of the outcome variable.

[0034] The intervention information DB12b, which is an example of an intervention classification storage means, stores information on the classification of interventions, etc. For example, as shown in FIG. 4, the intervention information DB12b stores an outcome type ID, an intervention classification ID, an intervention ID, etc. in association with an intervention ID indicating each intervention. As shown in FIG. 5, the interventions are divided into a plurality of hierarchies, such as major classifications and minor classifications. The major classifications of the interventions are, for example, "self-monitoring" in which the target user T monitors the outcome himself, "goal setting" in which the target user T sets an outcome goal, and "reward and threat" in which the target user T is given an incentive or a penalty depending on the outcome.

[0035] Furthermore, the "self-monitoring" in the major category A is classified into "behavior monitoring" for monitoring outcomes that appear when the target user T is acting, and "result monitoring" for monitoring outcomes that appear as a result of the target user T's action. The "goal setting" in the major category B is classified into "goal setting" for setting outcome goals for the target user T, and "action plan" for setting a plan for achieving outcome goals for the target user T. The "rewards and threats" in the major category C are classified into "physical rewards" for giving points or the like when the outcome goal is achieved, and "social rewards" for sending messages or the like when the outcome goal is achieved. Note that the intervention category ID may be classified into the part indicating the major category at the top and the part indicating the minor category at the bottom. Also, it may be classified separately, such as the major category ID and the minor category ID.

[0036] Furthermore, at a higher level than this classification, there is the type of outcome, which is indicated by the outcome type ID. Note that the classification system for interventions does not depend on the type of outcome, so each type of outcome is classified into major categories of interventions, such as A, B, C, etc., and major category A is further subdivided into A1, A2, etc., and major category B is subdivided into B1, B2, etc.

[0037] More specifically, if the outcome type is "physical activity," the intervention content for A1 "behavior monitoring" is "record the number of steps taken each day," "record the number of stairs climbed," etc. The intervention content for A2 "result monitoring" is "record weight (as an exercise result)," the intervention content for B1 "goal setting" is "achieve 10,000 steps," the intervention content for B2 "action plan" is "make a training plan," the intervention content for C1 "physical reward" is "reward points for every 1,000 steps walked," and the intervention content for C2 "social reward" is "a message of appreciation when the step goal is achieved."

[0038] For the outcome type "nutritional status," an example of the intervention for subcategory A1 would be "recording daily meals," for subcategory A2 would be "recording skin condition (result of dietary improvement)," for subcategory B1 would be "achieving a low-carb diet for one week," for subcategory B2 would be "deciding a daily menu," for subcategory C1 would be "rewarding points for each meal record entered," and for subcategory C2 would be "sending a congratulatory message each time fat is reduced."

[0039] As shown in FIG. 6, the user basic information DB12c stores basic data such as personal attributes, user environmental factors, user psychological characteristics, and user mental and physical conditions in association with the user ID of the user P. More specifically, from the answers to the questionnaire, personal attributes such as “age”, “gender”, “social support” score calculated from the answers to each question in the questionnaire, “physical environment for performing exercise”, “personality” score, “cognitive function” score, “lifestyle” score, “behavioral tendency” score, and “health condition” score are stored in the user basic information DB12c in association with the user ID. For the environmental factors, psychological characteristics, and mental and physical conditions, the total score calculated from each item is stored in the user basic information DB12c. The total score of the environmental factors is calculated from the “social support” score, the “physical environment for performing exercise” score, and the like. In addition, basic data on the target user T is also added to the user basic information DB12c. When added in this way, this target user T also becomes an example of a person other than the target person for other target users T.

[0040] As shown in FIG. 7, the reaction history DB12d stores the intervention ID of the applied intervention, the intervention classification ID indicating the type of intervention, the start time of the intervention, the latest time of the intervention, the value of the outcome variable, and the like, in association with the user ID of the user P. As the value of the outcome variable, values ​​such as the number of steps, the heart rate, the body temperature, the stress level, and the weight are stored in the reaction history DB12d together with the measurement time. Note that the measurement time is, for example, a date, and in the case of the number of steps, it is a cumulative value for one day, in the case of the heart rate, it is an average heart rate, in the case of the body temperature, it is the body temperature at the time of waking up, and in the case of the stress level, it is an average stress level. The reaction curve data of each outcome is obtained by plotting the value of each outcome variable against the measurement time (day). Note that the intervention classification ID may not be necessary, and the intervention classification ID may be identified from the intervention ID and the intervention information DB12b.

[0041] FIG. 8 shows a schematic representation of the relationship between the intervention applied to user P, the basic data of user P, and the measured response curve data. For convenience, the response curve data is divided into "Period 1" of the initial response and "Period 2" of the latter response (after Period 1). In reality, it is a series of time-series data from the start of the intervention to the latest intervention. Here, as shown in FIG. 8, the type of intervention (Interv.), age (Age), personal attributes such as gender (M / F), total score of environmental factors (Env.), total score of psychological characteristics (Psych), and response curve data for "Period 1" and "Period 2" are shown in association with the user number (or user ID). The total score is a standardized value such that the sum of the highest scores of each element is "7".

[0042] When the output unit 13 outputs a video image, the output unit 13 has, for example, a liquid crystal display element, an EL (Electro Luminescence) element, etc. When the output unit 13 outputs a sound, the output unit 13 has a speaker.

[0043] The input unit 14 includes, for example, a keyboard and a mouse.

[0044] The input / output interface section 15 performs interface processing between the communication section 11, the storage section 12, etc. and the control section 16.

[0045] The control unit 16 includes a central processing unit (CPU) 16a, a read only memory (ROM) 16b, a random access memory (RAM) 16c, etc. The control unit 16 calculates a predicted response curve of each target user T by causing the CPU 16a to read and execute codes of various programs stored in the ROM 16b and the storage unit 12.

[0046] (2.2 Configuration and Functions of the Mobile Terminal Device 20) Next, the configuration and functions of the mobile terminal device 20 will be described with reference to FIG.

[0047] FIG. 9 is a block diagram showing an example of a schematic configuration of the mobile terminal device 20. As shown in FIG.

[0048] 9, mobile terminal device 20 has output unit 21, storage unit 22, communication unit 23, input unit 24, sensor unit 25, input / output interface unit 26, and control unit 27. Control unit 27 and input / output interface unit 26 are electrically connected via system bus 28. A mobile terminal ID is assigned to each mobile terminal device 20.

[0049] The output unit 21 has, for example, a liquid crystal display element or an EL element as a display function. The output unit 32 has a speaker that outputs sound.

[0050] The storage unit 22 is configured, for example, with a hard disk drive, a solid state drive, etc. The storage unit 22 stores various programs, such as an operating system and an application for the mobile terminal device 20. The various programs may be acquired, for example, from another server device or the like via the network N, or may be recorded on a recording medium and read via a drive device. The storage unit 22 may also have information of a database, such as the storage unit 12 of the information processing server device 10.

[0051] The communication unit 23 is electrically or electromagnetically connected to the network N to control the state of communication with the information processing server device 10, etc. The communication unit 23 also has a wireless communication function for communicating with the wearable terminal device 30 by radio waves or infrared rays.

[0052] The input unit 24 has, for example, a display panel of a touch switch type such as a touch panel. The input unit 24 acquires position information of the output unit 21 where the user's finger touches or is in proximity. The input unit 24 has a microphone for inputting voice.

[0053] The sensor unit 25 has various sensors such as a GPS (Global Positioning System) sensor, a direction sensor, an acceleration sensor, a gyro sensor, an air pressure sensor, a temperature sensor, and a humidity sensor. The sensor unit 25 has an imaging element such as a CCD (Charge Coupled Device) image sensor of a digital camera or a CMOS (Complementary Metal Oxide Semiconductor) image sensor. The mobile terminal device 20 acquires current location information of the mobile terminal device 20 by the GPS sensor. Each sensor is assigned a unique sensor ID.

[0054] The input / output interface section 26 performs interface processing between the output section 21, the storage section 22, etc. and the control section 27.

[0055] The control unit 27 is configured with a CPU 27a, a ROM 27b, a RAM 27c, etc. In the control unit 27, the CPU 27a reads out various programs stored in the ROM 27b and the storage unit 22 and executes them.

[0056] The supporter terminal device 40 has the same configuration and functions as the information processing server device 10 or the mobile terminal device 20 .

[0057] If the supporter terminal device 40 is a personal computer, it has a similar configuration and functions to the information processing server device 10. If the supporter terminal device 40 is a tablet terminal, it has almost the same configuration and functions as the mobile terminal device 20. In addition, each supporter terminal device 40 is assigned a terminal ID.

[0058] (2.3 Configuration and Function of the Wearable Terminal Device 30) Next, the configuration and functions of the wearable terminal device 30 will be described with reference to FIG.

[0059] FIG. 10 is a block diagram showing an example of a schematic configuration of the wearable terminal device 30. As shown in FIG.

[0060] 10, the wearable terminal device 30 includes an output unit 31, a storage unit 32, a communication unit 33, an input unit 34, a sensor unit 35, an input / output interface unit 36, and a control unit 37. The control unit 37 and the input / output interface unit 36 ​​are electrically connected via a system bus 38. A terminal ID is assigned to each wearable terminal device 30. The wearable terminal device 30 has a clock function.

[0061] The output unit 31 has, for example, a liquid crystal display element or an EL element as a display function, a speaker for outputting sounds such as notifications, and the like.

[0062] The storage unit 32 is configured, for example, with a solid state drive, etc. The storage unit 32 stores various programs, such as an operating system and an application for the wearable terminal device 30. Note that the various programs may be acquired, for example, from another server device via the network N through the connected mobile terminal device 20.

[0063] The communication unit 33 is configured to control communication with the mobile terminal device 20 through wireless communication. The communication unit 33 may be electrically or electromagnetically connected to the network N to control the state of communication with the information processing server device 10 and the like.

[0064] The input unit 34 has, for example, a display panel of a touch switch type such as a touch panel. The input unit 34 acquires position information of the output unit 31 where the user's finger touches or is in proximity. The input unit 34 has a microphone for inputting voice.

[0065] The sensor unit 35 includes an acceleration sensor, a gyro sensor, a temperature sensor, a pressure sensor, an ultrasonic sensor, an optical sensor, an electric sensor, a magnetic sensor, an image sensor, etc. Each sensor is assigned a unique sensor ID.

[0066] The acceleration sensor measures the acceleration of the wearable terminal device 30. From the measurement data of the acceleration sensor, the arm movement of the target user T and the up and down movement of the target user T are measured. The gyro sensor measures the angular acceleration of the wearable terminal device 30. The wearable terminal device 30 measures the number of steps, posture during sleep, number of turns, etc. of the target user T using the acceleration sensor and gyro sensor.

[0067] The body temperature sensor measures the temperature of the part that is in contact. The pressure sensor measures, for example, a pulse wave. The optical sensor detects a response to electromagnetic waves being irradiated onto the skin, etc., i.e., at least one of reflected waves and transmitted waves. The optical sensor measures blood flow speed, blood oxygen concentration, and other blood components. The ultrasonic sensor detects a response to ultrasound being irradiated onto the skin, i.e., at least one of reflected waves and transmitted waves.

[0068] Electrical sensors measure voltage, current, impedance, etc. Electrical sensors measure electric fields generated by muscle activity, blood flow, nerve excitation, etc. Electrical sensors can also be combined with electrodes to detect components of sweat, etc., and function as chemical sensors, pH sensors, etc.

[0069] The magnetic sensor measures magnetic fields generated by muscle activity, blood flow, nerve excitation, etc.

[0070] The image sensor detects skin color, surface temperature, surface movement, blood flow, sweat patterns, and the like.

[0071] The sensor unit also includes a GPS sensor, a direction sensor, an air pressure sensor, etc. The wearable terminal device 30 may measure the distance traveled, the amount of exercise, and the like using these sensors.

[0072] In addition, the microphone of the input section may capture the snoring and breathing sounds of the subject T while he or she is sleeping.

[0073] The input / output interface unit 36 ​​performs interface processing between the output unit 31, the storage unit 32, etc., and the control unit 37.

[0074] The control unit 37 is configured with a CPU 37a, a ROM 37b, a RAM 37c, etc. In the control unit 37, the CPU 37a reads out various programs stored in the ROM 37b and the storage unit 32 and executes them.

[0075] The type of the wearable terminal device 30 may be a wristband type, a glasses type, a finger ring type, a shoe type, a pocket type, a necklace type, a clothing type, etc. The wearable terminal device 30 may be a combination of a wristband type wearable terminal device and a wearable terminal device other than a wristband type, or a plurality of the same type of wearable terminal devices may be worn.

[0076] [3. Example of operation of Prediction System 1] An example of the operation of the prediction system 1 will be described with reference to the drawings.

[0077] (3.1 Collection of basic data on target user T and initiation of intervention) The operation of collecting basic data of a new target user T and starting intervention will be described with reference to Fig. 11. Fig. 11 is a schematic diagram showing an example of a screen displayed on the mobile terminal device 20.

[0078] The target user T, who inputs the basic data of the target user T, starts a dedicated application. As shown in FIG. 11 , questions asking about personal attributes, environmental factors, psychological characteristics, etc. are displayed on the display of the output unit 21 of the mobile terminal device 20 of the target user T. The target user T answers the questions according to the questions. Note that the support user S who provides health advice may input the basic data of the target user T using the supporter terminal device 40 while asking questions to the target user T.

[0079] The mobile terminal device 20 transmits the input basic data to the information processing server device 10 together with the user ID of the target user T. When the support user S inputs the basic data to the supporter terminal device 40, the supporter terminal device 40 transmits the input basic data to the information processing server device 10 together with the user ID of the target user T. The information processing server device 10 stores the basic data together with the received user ID of the target user T in the user basic information DB 12c.

[0080] Next, the target user T selects the intervention to be received according to the display of the dedicated app. For example, in the case of improving health by walking, the intervention of "recording the number of steps every day" is selected. The mobile terminal device 20 transmits the outcome ID of the selected intervention together with the user ID of the target user T to the information processing server device 10. The information processing server device 10 stores the intervention ID and the date of the start of the intervention together with the received user ID of the target user T in the response history DB 12d.

[0081] Next, the mobile terminal device 20 connects to the wearable terminal device 30 at a predetermined interval and acquires information such as the number of steps and the heart rate from the wearable terminal device 30. The mobile terminal device 20 may calculate the number of steps from information such as acceleration of an acceleration sensor of the wearable terminal device 30.

[0082] The target user T receives the intervention by looking at the number of steps taken per day displayed on the mobile terminal device 20 for an initial measurement period of one week, and inputting the number of steps taken per day into an app or recording it on a record sheet. When the intervention is to "record the number of steps taken every day", it is preferable to have the target user T consciously do the work. Note that the mobile terminal device 20 may notify the number of steps taken yesterday and the current day at a predetermined time at least once a day. Note that daily weight may also be input. Note that the intervention is not limited to once a day, and the current number of steps may be notified or input at predetermined intervals such as every hour, every two hours, or every three hours.

[0083] The mobile terminal device 20 transmits information such as the cumulative number of steps and average heart rate for one day together with the user ID of the target user T to the information processing server device 10. The information processing server device 10 stores the received information such as the number of steps and heart rate together with the user ID of the target user T as subject response curve data together with the date in the response history DB 12d. The data at the latest intervention point in the response history DB 12d is updated.

[0084] (3.2 Example of reaction curve prediction) Next, an example of an operation for predicting a response curve for a target user T will be described with reference to FIGS.

[0085] Fig. 12 is a flowchart showing an example of the operation of the information processing server device 10. Fig. 13 is a schematic diagram showing an example of the type of intervention, basic data, and reaction curve of the target user T. Fig. 14 is a schematic diagram showing an example of a predicted reaction curve of the target user T.

[0086] First, the information processing server device 10 refers to the reaction history DB 12d based on the user ID of the target user T who is a new user, and determines whether or not a specific measurement period (for example, one week) has passed since the start of the intervention. If the specific measurement period has passed, the information processing server device 10 starts the process of predicting the reaction curve.

[0087] As shown in Fig. 12, the prediction system 1 acquires subject data (step S1). Specifically, the information processing server device 10 reads out basic data from the user basic information DB 12c based on the user ID of the target user T. The information processing server device 10 acquires an intervention ID and time-series data of outcome variables in a specific measurement period from the start of the intervention, i.e., an initial response curve of the target user T, from the response history DB 12d based on the user ID of the target user T. Furthermore, the information processing server device 10 acquires an intervention classification ID by referring to the intervention information DB 12b based on the acquired intervention ID.

[0088] More specifically, as shown in Fig. 13, the information processing server device 10 identifies the intervention subcategory "A1" from the intervention classification ID, and identifies the gender, age, environmental factor total score "1", and psychological characteristic total score "5" from the read basic data. The information processing server device 10 also reads out a response curve of the outcome variable "number of steps" in the specific measurement period "period 1". Multiple outcome variables may be read out.

[0089] In this way, the information processing server device 10 functions as an example of a subject data acquisition means for acquiring basic data regarding the subject including the subject's attributes, data indicating the type of health-supporting intervention received by the subject, and subject response curve data measuring the subject's response to the intervention during a specific measurement period from the start of the intervention.

[0090] Next, the prediction system 1 identifies interventions with similar types of interventions (step S2). Specifically, the information processing server device 10 identifies similar interventions by referring to the intervention information DB 12b based on the intervention classification ID. More specifically, the information processing server device 10 identifies interventions or interventions of a subclassification that belongs to the same major classification as the major classification to which the intervention belongs. When the information processing server device 10 is the intervention subclassification "A1", it identifies the intervention classification IDs of the intervention subclassifications "A1" and "A2", or the intervention IDs that belong to the major classification "A". Note that as interventions with similar types of interventions, intervention IDs that belong to the subclassification "A1" of the same intervention may be extracted. In this case, the range of similarity is narrowed. The intervention similarity between the type of intervention received by the subject and the type of intervention received by another person is highest when the intervention contents match, second highest when the subclassifications match, and third highest when the major classifications match.

[0091] In this way, the information processing server device 10 functions as an example of an intervention similarity calculation means that refers to an intervention classification storage means in which the types of intervention are classified, and calculates the intervention similarity between the type of intervention received by the subject and the type of intervention received by the other person in accordance with the classification.

[0092] Next, the prediction system 1 identifies other users with a similar type of intervention (step S3). Specifically, the information processing server device 10 refers to the response history DB 12d to identify the user IDs of other users that correspond to the identified intervention ID or intervention category ID. As shown in FIG. 14, other users, User No. 1 in subcategory "A1" and User No. 3 in subcategory "A2", are selected. The other users are users who have outcome data longer than the specific measurement period, for example, User P.

[0093] Next, the prediction system 1 calculates the similarity of the basic data of the specified other user (step S4). Specifically, the information processing server device 10 reads out the basic data of the specified other user P from the user basic information DB 12c based on the user ID of the specified other user P. The information processing server device 10 calculates the similarity from the read out basic data of the specified other user P and the basic data of the target user T. When a plurality of other users are specified, the similarity of each user is calculated.

[0094] More specifically, as shown in FIG. 13, the information processing server device 10 standardizes each element of the basic data 50 of the target user T, namely, gender "M", age "25", environmental factor total score "1", and psychological characteristic total score "5", and converts them into a four-dimensional vector. As shown in FIG. 14, the information processing server device 10 also converts the basic data of the specified other users "User No. 1" and "User No. 3" into a standardized four-dimensional vector. The information processing server device 10 calculates the cosine similarity between the vector of the basic data 50 of the target user T and the vector of the basic data of each other user P. Note that the dimension of the vector is not limited to four dimensions, and may be five dimensions including the total score of the mental and physical condition, and each element such as "social support" and "physical environment for performing exercise" may be used as a component of the vector instead of the total score. The similarity is not limited to the cosine similarity, and may be a similarity based on a correlation coefficient.

[0095] In this way, the information processing server device 10 functions as an example of a basic data similarity calculation means that calculates the basic data similarity in accordance with the attributes, the psychological characteristics, and the environmental factors.

[0096] Next, the prediction system 1 calculates the reaction curve similarity for the reaction curve of the identified other person (step S5). Specifically, the information processing server device 10 reads out the reaction curve of the identified other user P from the reaction history DB 12c based on the user ID of the identified other user P. The information processing server device 10 calculates the similarity between the reaction curve of the identified other user P that has been read out and the reaction curve of the target user T. When multiple other users are identified, the similarity for each is calculated.

[0097] More specifically, the information processing server device 10 calculates the reaction curve similarity between the reaction curve 51 of the target user T as shown in Fig. 13 and the reaction curves of "User No. 1" and "User No. 3" of the specified other user P in period 1 as shown in Fig. 14. The reaction curve similarity is calculated, for example, by dynamic time warping, which determines the similarity by pattern matching between time series. The reaction curve similarity may be calculated using a correlation coefficient between curves, Euclidean distance, etc.

[0098] The period for calculating the similarity is preferably the same period from the start of the intervention. However, it does not have to be exactly the same period, since the degree of agreement of the initial response curves from the start of the intervention is being examined. It is sufficient that the amount of data required to calculate the similarity is observed in approximately the same period from the start of the intervention. In addition, if measurements are taken several times a day, the response curve may be calculated from the cumulative value for that day or a smoothed value.

[0099] In addition, when calculating the response curve similarity, the response curve of the same outcome variable is preferable. If the response curve of the target user T is the response curve of "step count", the response curve of the other user P is also preferable to be the "step count". It does not have to be the same outcome variable, but it may be a response curve of an outcome variable belonging to the same physical activity amount (for example, the number of stairs climbed). Furthermore, in some cases, it may be a response curve of an outcome variable in the case of a physical condition amount other than the physical activity amount (for example, heart rate, which is correlated with the number of steps). Here, if the outcome variable has a sufficient correlation in the database or is known to have a sufficient correlation from experience, it may be the subject of the calculation of the response curve similarity.

[0100] Furthermore, when calculating the response curve similarity, the response curve patterns may be classified into similarity patterns based on the classification of trend patterns. For example, response curve patterns are broadly classified into an upward trend type, a downward trend type, and a non-increasing and declining type. The upward trend type is further classified into a proportional type, an exponential type, a saturated type, etc. The downward trend type is further classified into a proportional type, an exponential type, a steep drop in the latter half, etc. The non-increasing and declining type is further classified into a constant type, a peak type, a valley type, etc. The information processing server device 10 classifies the response curve trend patterns, and can determine that the similarity is high when the minor classifications match, and the similarity is the next highest when the major classifications match.

[0101] In this way, the information processing server device 10 functions as an example of a response curve similarity calculation means that calculates the response curve similarity by pattern matching of the response curves.

[0102] Next, the prediction system 1 selects reaction curve data based on each similarity (step S6). Specifically, the information processing server device 10 selects reaction curve data with a basic data similarity of a predetermined value or more (an example of selected reaction curve data) and reaction curve data with a reaction curve similarity of a predetermined value or more (an example of selected reaction curve data). More specifically, as shown in FIG. 14, the similarity of the basic data 60 of the user No. 1 to the basic data 50 of the target user T is determined to be a predetermined value or more, and the reaction curve data of the user No. 1 is selected. In addition, the reaction curve similarity of the reaction curve 61 of the user No. 3 to the reaction curve 51 of the target user T is determined to be a predetermined value or more, and the reaction curve data of the user No. 3 is selected. Note that reaction curve data with a basic data similarity of a predetermined value or more and a reaction curve similarity of a predetermined value or more may be selected.

[0103] In this way, the information processing server device 10 functions as an example of a selective response curve data selection means that refers to a storage means in which basic data, a type of intervention, and response curve data related to another person other than the subject are associated with each other, and selects selected response curve data from the response curve data of the other person in the storage means according to the type of intervention received by the subject, the basic data of the subject, and the subject's response curve data. The information processing server device 10 also functions as an example of a selective response curve data selection means that selects the selected response curve data from the response curve data of the other person who has received an intervention similar to the type of intervention received by the subject, according to the basic data similarity to the basic data of the subject and the response curve similarity to the subject's response curve data in the specific measurement period. The information processing server device 10 also functions as an example of a selective response curve data selection means that selects the selected response curve data from the response curve data of the other person who has received an intervention similar to the type of intervention received by the subject, according to the basic data similarity to the basic data of the subject and the response curve similarity to the subject's response curve data in the specific measurement period. The information processing server device 10 also functions as an example of a selected reaction curve data selection means for selecting a plurality of the selected reaction curve data. The information processing server device 10 also functions as an example of a selected reaction curve data selection means for selecting a plurality of the selected reaction curve data. The information processing server device 10 also functions as an example of a selected reaction curve data selection means for selecting the plurality of the selected reaction curve data, including selected reaction curve data based only on the reaction curve similarity and selected reaction curve data based only on basic data similarity with the subject's basic data.

[0104] Next, the prediction system 1 calculates the predicted response curve of the subject (step S7). Specifically, the information processing server device 10 extracts response curve data from the specific measurement period of the target user T to the end from the selected response curve data. If the specific measurement period of the target user T is 30 days, data from the 31st day to the end (the most recent intervention time) is extracted. More specifically, as shown in FIG. 14, for example, in "period 2", a predicted response curve 62 of user No. 1 and a predicted response curve 63 of user No. 3 are calculated. If the prediction period presented to the target user T is fixed, the length of the response curve data may be up to the prediction period.

[0105] The information processing server device 10 may calculate a predicted response curve from a plurality of response curves. Specifically, the selected response curves are averaged over the prediction period to calculate the predicted response curve of the target user T. If the data length from the specific measurement period to the end is not the same, the response curves are averaged according to the response curve data with the shorter data length. For example, as shown in FIG. 13, a predicted response curve 52 may be calculated by averaging the predicted response curve 62 and the predicted response curve 63 in "period 2". The values ​​of the predicted response curve 62 and the predicted response curve 63 at the same time are averaged.

[0106] The multiple response curves may be a response curve of a user P whose basic data similarity is greater than or equal to a predetermined value and whose response curve similarity is greater than or equal to a predetermined value, a response curve of a user P whose basic data similarity is greater than or equal to a predetermined value or whose response curve similarity is greater than or equal to a predetermined value, a response curve of a user P whose basic data similarity is greater than or equal to a predetermined value, a response curve of a user P whose response curve similarity is greater than or equal to a predetermined value, or a response curve of a user P with a similar type of intervention.

[0107] The average of the multiple response curves may be a simple average or a weighted average. In the case of a weighted average, the weighting is determined based on at least one of the similarity of the basic data and the similarity of the response curves. For example, the weighting of each selected response curve is set high when the similarity of the basic data, the similarity of the response curve, etc. is high.

[0108] In this way, the information processing server device 10 functions as an example of a response curve prediction means that predicts the subject's predicted response curve to the intervention after the specific measurement period of the subject based on the selected response curve data. The information processing server device 10 also functions as an example of a response curve prediction means that predicts the predicted response curve from the multiple selected response curve data. The information processing server device 10 also functions as an example of a response curve prediction means that predicts the predicted response curve from the multiple selected response curve data according to weighting by at least one of the response curve similarity and the basic data similarity.

[0109] Next, the prediction system 1 outputs the predicted response curve (step S8). Specifically, the information processing server device 10 transmits the calculated predicted response curve data to the mobile terminal device 20 of the target user T. The mobile terminal device 20 displays the predicted response curve on the display of the output unit 21.

[0110] The intervals for plotting the predicted response curve displayed on the output unit 21 can be freely set to 1 hour, 2 hours, 3 hours, 1 day, 2 days, 3 days, 1 week, 2 weeks, 1 month, etc. If the smallest unit of data is "1 minute", the intervals for plotting can be set in minutes.

[0111] The period of the predicted response curve displayed by the output unit 21 can be set arbitrarily from after the specific measurement period of the target user T to the latest point of intervention in the response curve data of the user P, that is, to the end of the data.

[0112] The predicted response curves displayed on the display of the output unit 21 may be a predicted response curve 62 calculated based on similarity of basic data, a predicted response curve 63 calculated based on similarity of response curves, a predicted response curve based on similarity of basic data and similarity of response curves, or a response curve calculated by averaging a plurality of response curves. A plurality of response curves may be displayed, or the target user T may select which one to display.

[0113] In this way, the information processing server device 10 functions as an example of an output means that outputs the predicted response curve at a predetermined time interval. Also, the information processing server device 10 functions as an example of an output means that outputs the predicted response curve for a predetermined period within the range of the selected response curve data.

[0114] The information processing server device 10 may transmit the predicted response curve data to the supporter terminal device 40, and the support user S may give advice to the target user T while looking at the predicted response curve.

[0115] (Example) Next, an embodiment will be described with reference to FIG. 15A and FIG. 15B. 15A and 15B are schematic diagrams showing an example of a comparison of prediction accuracy.

[0116] The wearable terminal device 30 was lent to 70 adults who had no exercise habits as subjects, and the number of steps, which was an outcome variable, was observed eight times a day for one month. The 70 people were divided into three groups according to the intervention. The three groups were a self-monitoring group, a goal-setting group, and a reward-threat group. The number of people in the self-monitoring group was 23, the number of people in the goal-setting group was 23, and the number of people in the reward-threat group was 24. Specifically, the intervention for the self-monitoring group was to display the results on the screen of the wearable terminal device 30 or the mobile terminal device 20. The intervention for the goal-setting group was to have them set a daily step count goal, and to notify the wearable terminal device 30 or the mobile terminal device 20 when the goal was achieved. The input for the reward-threat group was to check the number of steps of a friend every day as a social comparison. The friends were included in the 70 people.

[0117] To obtain basic data, all subjects were asked to enter information regarding their age, sex, whether they were employed, their education level, personality, etc.

[0118] After obtaining the data on the number of steps taken over a one-month period, the data on the number of steps taken over the first week after the start of the intervention was used as training data, and the data on the last three weeks was used as test data to evaluate the prediction accuracy of the response curve. The basic data and number of steps taken by the other 69 subjects were set as other-person data.

[0119] Figure 15A shows the experimental results for all 70 people, regardless of the type of intervention. Figure 15B shows the experimental results for the self-monitoring group only. The vertical axis shows the prediction error in units of the squared number of steps. The horizontal axis shows the type of predicted response curve calculated and the conventional method. The types of predicted response curves are the "response curve similarity" response curve calculated from the response curve similarity, the "basic data similarity" response curve calculated from the basic data similarity, and the "ensemble" which is the average of the response curve of the response curve similarity and the response curve of the basic data similarity. The conventional methods are the ARIMA model and the LLM (Linear Mixed Model). The graph shows the average value and standard error.

[0120] As shown in FIG. 15A, “ensemble” had the least error, followed by “response curve similarity.”

[0121] In addition, as shown in Figure 15B, the results were almost the same regardless of the type of intervention in the self-monitoring group. This also means that the prediction of the predicted response curve is robust to the type of intervention.

[0122] As described above, according to this embodiment, data indicating the type of health-supporting intervention received by the target user T and subject response curve data measuring the target user T's response to the intervention during a specific measurement period from the start of the intervention are obtained, and selected response curve data is selected from the response curve data of other users P other than the target user T according to the type of intervention received by the target user T, the subject's basic data, and the subject response curve data, and a predicted response curve of the target user T to the intervention after the specific measurement period of the target user T is predicted based on the selected response curve data, thereby improving the prediction accuracy of the target user T's responsiveness to the intervention.

[0123] When selecting response curve data from response curve data of other users P who received an intervention similar to the type of intervention received by the target user T, based on the response curve similarity with the subject's response curve data in a specific measurement period, the response curve data is narrowed down by type of intervention, and then response curve data with similar response curves at the beginning of the specific measurement period is selected based on the response curve similarity, thereby further improving the prediction accuracy of the target user T's responsiveness to the intervention.

[0124] When the selected response curve data is selected from the response curve data of another person who has received an intervention similar to the type of intervention received by the target user T, based on the basic data similarity with the basic data of the subject and the response curve similarity with the subject's response curve data in the specific measurement period, a response curve whose basic data similarity is greater than or equal to a predetermined value and whose response curve similarity is greater than or equal to a predetermined value can be selected, thereby further improving the accuracy of predicting the response of the target user T to the intervention.

[0125] When multiple pieces of selected response curve data are selected and a predicted response curve is predicted from the multiple selected response curve data, the prediction accuracy of the target user T's responsiveness to an intervention can be further improved, for example, by obtaining a predicted response curve by averaging these multiple selected response curves.

[0126] When selecting multiple selective response curve data including selective response curve data based only on basic data similarity and selective response curve data based only on response curve similarity, the accuracy of predicting the responsiveness of the target user T to an intervention can be further improved by, for example, obtaining a predicted response curve by averaging the selective response curve based only on basic data similarity and the selective response curve based only on response curve similarity.

[0127] When predicting a predicted response curve from multiple selected response curve data based on weighting based on at least one of response curve similarity and basic data similarity, the accuracy of predicting the responsiveness of the target user T to an intervention can be further improved, for example, by obtaining a predicted response curve by taking a weighted average of multiple selected response curves.

[0128] When calculating the intervention similarity between the type of intervention received by the target user T and the type of intervention received by another user P according to the classification by referring to an intervention classification storage means in which types of intervention are classified, the prediction accuracy of the target user T's responsiveness to the intervention can be further improved by narrowing down the intervention similarity to response curve data whose intervention similarity is below a predetermined value.

[0129] When the basic data includes data on the attributes of the target user T, data on the psychological characteristics of the subject, and data on the environmental factors of the subject, and the basic data similarity is calculated based on the attributes, psychological characteristics, and environmental factors, the accuracy of determining the similarity of the basic data is improved, and the accuracy of predicting the responsiveness of the target user T to an intervention can be further improved.

[0130] When the response curve similarity is calculated by pattern matching of the response curves, the accuracy of determining the response curve similarity is improved, and the accuracy of predicting the responsiveness of the target user T to an intervention can be further improved.

[0131] When a predicted response curve is output at a predetermined time interval, a display that is desired or easy to view can be provided to the target user T. When the type of intervention is related to the display of outcome variables, it can be useful for supporting the health of the target user T.

[0132] When the predicted response curve for a predetermined period is output within the range of selected response curve data, a display that is desired or easy to view can be provided for the target user T. When the type of intervention is related to the display of outcome variables, it can be useful for supporting the health of the target user T.

[0133] Furthermore, the present invention is not limited to the above-mentioned embodiments. The above-mentioned embodiments are merely examples, and anything that has substantially the same configuration as the technical idea described in the claims of the present invention and exhibits similar effects is included in the technical scope of the present invention. [Explanation of symbols]

[0134] 1: Prediction system 10: Information processing server device (prediction device) 12: Storage unit (storage means) 20: Portable terminal device (terminal device) 30: Wearable terminal device (terminal device) 40: Supporter terminal device (terminal device) T: Target user (target person) P: Other users (others)

Claims

1. A subject data acquisition means for acquiring basic data on the subject including attributes of the subject, data indicating the type of health support intervention received by the subject, and subject response curve data measuring the response of the subject to the intervention during a specific measurement period from the start of the intervention; a selective response curve data selection means for selecting selective response curve data from the response curve data of the other person stored in the storage means in accordance with the type of intervention received by the subject, the basic data of the subject, and the subject's response curve data, by referring to a storage means in which basic data, a type of intervention, and response curve data relating to the other person other than the subject are associated with each other; A response curve prediction means for predicting a predicted response curve of the subject to the intervention after the specific measurement period of the subject based on the selected response curve data; A prediction device comprising:

2. 2. The prediction device according to claim 1, A prediction device characterized in that the selected response curve data selection means selects the selected response curve data from the response curve data of others who have received an intervention similar to the type of intervention received by the subject, based on the basic data similarity with the basic data of the subject and the response curve similarity with the subject's response curve data during the specific measurement period.

3. 3. The prediction device according to claim 2, The selective reaction curve data selection means selects a plurality of the selective reaction curve data, A prediction device, characterized in that the reaction curve prediction means predicts the predicted reaction curve from the plurality of selected reaction curve data.

4. 4. The prediction device according to claim 3, A prediction device characterized in that the selective response curve data selection means selects the multiple selective response curve data including selective response curve data based only on the basic data similarity and selective response curve data based only on the response curve similarity.

5. 5. The prediction device according to claim 3, A prediction device characterized in that the response curve prediction means predicts the predicted response curve from the multiple selected response curve data in accordance with weighting based on at least one of the response curve similarity and the basic data similarity.

6. The prediction device according to any one of claims 2 to 4, A prediction device characterized by further comprising an intervention similarity calculation means for calculating an intervention similarity between the type of intervention received by the subject and the type of intervention received by the other person in accordance with the classification by referring to an intervention classification storage means in which the types of intervention are classified.

7. The prediction device according to any one of claims 2 to 4, The basic data includes data on the attributes of the subject, data on psychological characteristics of the subject, and data on environmental factors of the subject; The prediction device further comprises a basic data similarity calculation means for calculating the basic data similarity in accordance with the attribute, the psychological characteristic, and the environmental factor.

8. The prediction device according to any one of claims 2 to 4, The prediction device further comprises a reaction curve similarity calculation means for calculating the reaction curve similarity by pattern matching of reaction curves.

9. The prediction device according to any one of claims 2 to 4, further comprising an output means for outputting the predicted reaction curve; A prediction device, characterized in that the output means outputs the predicted response curve at predetermined time intervals.

10. The prediction device according to any one of claims 2 to 4, further comprising an output means for outputting the predicted reaction curve; A prediction device, characterized in that the output means outputs the predicted response curve for a predetermined period within the range of the selected response curve data.

11. a subject data acquisition step in which a subject data acquisition means acquires basic data on the subject including attributes of the subject, data indicating the type of health support intervention received by the subject, and subject response curve data measuring the subject's response to the intervention during a specific measurement period from the start of the intervention; a selective response curve data selection step in which a selective response curve data selection means refers to a storage means in which basic data on other people other than the subject, a type of intervention, and response curve data are associated with each other, and selects selective response curve data from the response curve data of the other people in the storage means according to the type of intervention received by the subject, the basic data on the subject, and the subject's response curve data; A response curve prediction step in which a response curve prediction means predicts a predicted response curve of the subject to the intervention after the specific measurement period of the subject based on the selected response curve data; A prediction method comprising:

12. Computer, a subject data acquisition means for acquiring basic data on the subject including attributes of the subject, data indicating the type of health support intervention received by the subject, and subject response curve data measuring the response of the subject to the intervention during a specific measurement period from the start of the intervention; a selective response curve data selection means for selecting selective response curve data from the response curve data of the other person stored in the storage means in accordance with the type of intervention received by the subject, the basic data on the subject, and the subject response curve data by referring to a storage means in which basic data on other people other than the subject, the type of intervention, and the response curve data are associated with each other; and A program for a prediction device, characterized by functioning as a response curve prediction means for predicting a predicted response curve of the subject to the intervention after the specific measurement period of the subject based on the selected response curve data.

13. A prediction system including a terminal device of a subject receiving an intervention to support health, and a prediction device that measures and predicts a response of the subject to the intervention, The prediction device, A subject data acquisition means for acquiring from the terminal device basic data on the subject including attributes of the subject, data indicating the type of intervention received by the subject, and subject response curve data measuring the response of the subject to the intervention during a specific measurement period from the start of the intervention; a selective response curve data selection means for selecting selective response curve data from the response curve data of the other person stored in the storage means in accordance with the type of intervention received by the subject, the basic data of the subject, and the subject's response curve data, by referring to a storage means in which basic data, a type of intervention, and response curve data relating to the other person other than the subject are associated with each other; A response curve prediction means for predicting a predicted response curve of the subject to the intervention after the specific measurement period of the subject based on the selected response curve data; A prediction system comprising:

Images