JP7405352B2 - Cognitive function improvement solution estimation system, cognitive function improvement solution estimation method and program - Google Patents

Description

The present invention relates to a cognitive function improvement solution estimation system, a cognitive function improvement solution estimation method, and a program.

Today, reducing the anxiety and burden of elderly people with declining cognitive function and their families is recognized as a social issue. Therefore, there is increasing interest in solutions for cognitive decline. In the academic field, the effects of improving or improving cognitive function have been pointed out for four solutions: study, diet, exercise, and communication.

Hayato Nagatomo and 3 others, "Intervention effect of short-term intensive rehabilitation in improving cognitive function in elderly people with dementia", Aichi Physical Therapy Society, December 2011, Vol. 23, No. 2, pp. 51-55. Masaru Mimura, “Non-drug therapy for dementia with evidence,” Higher Brain Function Research, 2012, Vol. 32, No. 3, pp. 454-460 Ryuta Kawashima, “Aiming to prevent and improve dementia through learning therapy”, [online], July 23, 2009, [Retrieved October 11, 2021], <URL:http://homesweethome. jp/files/libs/126/20180420112850351.pdf＞

However, the four proposed solutions only mention general effects, and it is not known which of the four solutions is effective in improving or improving the cognitive function of a specific individual. In other words, there are individual differences between the four solutions and their effectiveness, and it is difficult to judge the degree of effectiveness.

An object of the present invention is to provide a service that estimates, for each user, a combination of one or more solutions that are expected to improve cognitive function.

The invention according to claim 1 proposes to the user a solution that combines one or more of a plurality of solutions including one or more of learning, eating, exercise, and communication for the purpose of improving cognitive function. a first acquisition unit that acquires first information representing an evaluation of the cognitive function of the subject before implementation of the solution that combines the one or more of the above, and a solution that combines the one or more of the above. a second acquisition unit that acquires second information representing an evaluation of the subject's cognitive function after implementation of the above, and a combination of the one or more of the above based on the first information and the second information a determination unit that determines a change in the evaluation of the subject's cognitive function that occurred before and after implementation of the solution; This is a cognitive function improvement solution estimation system that estimates a newly proposed solution or a combination of multiple solutions based on the following information.
The invention according to claim 2 is the cognitive function improvement solution estimation system according to claim 1, wherein the first information and the second information are provided as a result of analyzing measurement data measured in different periods. It is.
The invention according to claim 3 is the cognitive function improvement solution estimation system according to claim 1, wherein the first information and the second information are given as a result of a test regarding the cognitive function of the subject.
The invention according to claim 4 is the cognitive function improvement solution estimation system according to claim 1, wherein the proposal unit proposes one or a combination of solutions recommended for attribute data of the subject.
The invention according to claim 5 is the cognitive function improvement solution estimation system according to claim 4, wherein the proposal unit proposes one or a combination of solutions recommended for the classification to which the attribute data belongs. .
The invention according to claim 6 is the cognitive function improvement solution estimation according to claim 4, wherein the proposal unit proposes one or a combination of solutions recommended for another subject to which the attribute data belongs. It is a system.
The invention according to claim 7 is the invention according to claims 4 to 6, wherein the proposing unit estimates a newly proposed solution combining one or more based on the determination result of the determining unit and the attribute data. The cognitive function improvement solution estimation system according to any one of the above.
The invention according to claim 8 further includes a modification unit that modifies the attribute data based on an opinion regarding the solution combining one or more obtained from the subject after implementing the solution combining the one or more. A cognitive function improvement solution estimation system according to claim 7.
The invention according to claim 9 is characterized in that the proposal unit estimates, for each subject, a new solution combining one or more solutions that improves the evaluation of the subject's cognitive function using a Bayesian optimization method. This is the cognitive function improvement solution estimation system described in Item 1.
The invention according to claim 10 is a cognitive function improvement solution estimation method realized through the execution of a program by a computer, which comprises improving one or more of learning, eating, exercise, and communication for the purpose of improving cognitive function. A process of proposing for each subject a solution that combines one or more of the multiple solutions, and a first process that represents an evaluation of the subject's cognitive function before implementing the solution that combines one or more of the solutions. a process of acquiring information; a process of acquiring second information representing an evaluation of the cognitive function of the subject after implementation of a solution combining the one or more of the above; and the first information and the second information. A process of determining a change in the evaluation of the subject's cognitive function that occurred before and after the implementation of the one or more solutions combined based on the above, and a process of newly proposing one or more solutions based on the result of the determination. This is a method for estimating a cognitive function improvement solution, which includes a process of estimating a solution that combines a plurality of solutions.
The invention according to claim 11 provides a computer with a cognitive function of a subject before implementing a solution that combines one or more of a plurality of solutions including one or more of learning, eating, exercise, and communication. a function of acquiring first information representing an evaluation of the subject; a function of acquiring second information representing an evaluation of the subject's cognitive function after implementation of a solution combining one or more of the above; a function of determining, based on the information and the second information, a change in the evaluation of the cognitive function of the subject that occurred before and after implementing the one or more solutions combined, and based on the result of the determination, This is a program for realizing a function of estimating one or more newly proposed solutions.

According to the present invention, it is possible to provide a service that estimates for each user a combination of one or more solutions that are expected to improve cognitive function.

1 is a diagram illustrating a conceptual configuration example of a system that provides a cloud service assumed in the first embodiment; FIG. FIG. 2 is a diagram illustrating a configuration example of a cloud server used in the first embodiment. FIG. 2 is a diagram illustrating an example data structure of a DB stored in a storage device. (A) shows an example of the data structure of the attribute data DB, and (B) shows an example of the data structure of the measurement data DB. FIG. 7 is a diagram illustrating an example data structure of another DB stored in a storage device. (A) shows an example of the data structure of the execution history DB, and (B) shows an example of the data structure of the evaluation result DB. 3 is a flowchart illustrating an example of a processing operation executed by a cloud server assumed in the first embodiment. FIG. 3 is a diagram illustrating changes in cognitive function evaluation due to execution of the proposed solution. (A) is an example of a change in cognitive function when a solution is proposed based on the proposer's experience, (B) is an example of a change in cognitive function when a solution is proposed by this service, and (C) is an example of a change in cognitive function when a solution is proposed by this service. This is an example of changes in cognitive function when a solution is proposed taking preferences into consideration. FIG. 3 is a diagram illustrating a configuration example of a cloud server used in Embodiment 2. FIG. 12 is a flowchart illustrating an example of a processing operation executed by a cloud server assumed in Embodiment 2. FIG. FIG. 7 is a diagram illustrating a solution proposal method in Embodiment 4. (A) shows the distribution of attribute data and classifications G1 to G8 corresponding to each target person, and (B) shows an example of correspondence between classification and solution. FIG. 7 is a diagram illustrating a solution proposal method in Embodiment 5.

Embodiments of the present invention will be described below with reference to the drawings.
<Embodiment 1>
<Term>
In this embodiment, a service that provides a solution that is expected to be effective in improving the cognitive function of a target person will be described. In the following, the service provided in this embodiment will also be referred to as "this service".
"Expected" here is used to mean that there is a high probability or possibility that an improvement effect on cognitive function will be obtained.

"Target people" are not limited to those who have been diagnosed with dementia, but also include those who are suspected of having dementia, those who wish to prevent dementia, and those who are concerned about decline in cognitive function. Use with meaning. In this embodiment, all people who use this service for the purpose of improving cognitive function are included in the target audience.
On the other hand, users of this service include, for example, the target person, the target's family and relatives, the person in charge of the business that supports the target person's life, and doctors and staff at medical institutions.

<System configuration>
FIG. 1 is a diagram showing an example of a conceptual configuration of a system (hereinafter referred to as "cloud system") 1 that provides cloud services assumed in the first embodiment.
The cloud system 1 shown in FIG. 1 includes a network 10, an information terminal 20 used for uploading data recorded and measured about a subject, and a solution expected to improve cognitive function for each subject. It consists of a cloud server 30 for estimating, and an information terminal 40 that receives solutions from the cloud server 30.

The network 10 in this embodiment is, for example, the Internet, or a mobile communication system such as 4G or 5G.
However, depending on the form of provision of this service, the network 10 may be a LAN (=Local Area Network). Further, when data is transferred off-line, a semiconductor memory or an optical disk may be used instead of the network 10.

The information terminal 20 is, for example, an information terminal 20A installed in a private home where the target person resides or worn by the target person, or an information terminal installed in a nursing care facility where the target person resides or worn by the target person. It is 20B. Private homes and nursing care facilities are examples of the target person's living base.
The information terminals 20A and 20B are, for example, a notebook computer, a desktop computer, a tablet computer, a smartphone, or a wearable computer.

Various data necessary for providing this service is uploaded from the information terminals 20A and 20B to the cloud server 30 in real time or periodically. Data is automatically uploaded at predetermined intervals, such as once a week or once a month, for example. However, if the target person or the service user instructs uploading, the data will be uploaded at any time.
Various types of data necessary to provide services include, for example, information about the subject's own body and medical history, information about the medical history of family members and relatives, information about the subject's preferences, measurement data of the subject's sleep and behavior, and suggestions. Contains execution history data about the solutions created.

The physical information of the subject includes, for example, gender, age, height, weight, visual acuity, and physical disabilities.
Information regarding the medical history of the subject includes, for example, a diagnosis of dementia, a history of hospital visits, and nursing care certification.
Information regarding the target person's preferences includes, for example, whether he likes movies, likes fish dishes, likes mystery novels, likes karaoke, and likes talking.
Measured data includes, for example, sleep data and behavioral data.
Sleep data is biological data acquired during sleep, and includes data related to pulse rate, breathing rate, turning over in bed, and other body movements. Sleep data also includes sleep time and sleep cycles, which can be obtained from body movement data.

To acquire sleep data, for example, a mat-shaped sensor that can measure data on the subject's body movements without contact, a microwave Doppler radar, and an acceleration sensor built into a smartphone are used.
Sleep data does not need to cover the entire period during which the subject is asleep. For example, only biometric data from time periods in which sleep characteristics are likely to appear may be used. As the time period in which sleep characteristics are likely to appear, for example, the time period immediately after falling asleep, the time period during the first REM sleep and non-REM sleep, or the time period immediately before waking up may be used.

The behavioral data includes biometric data acquired while waking up, as well as output data from sensors that detect the operation of equipment and the behavior of the subject due to the subject's behavior.
The biological data acquired while waking up includes data related to heartbeat, pulse, breathing rate, blood pressure, blood flow, number of steps, and other body movements. For example, a sensor worn on the wrist can be used to acquire these biometric data.
Sensors that detect the operation of equipment or the behavior of the target person due to the target person's actions include, for example, a human sensor installed in the bathroom, a sensor that detects the opening and closing of the bathroom door, the operating status of the water heater, Sensors that detect water supply temperature, bathtub water level, and other data; controllers installed in bathrooms and kitchens; sensors that detect cooker operations; sensors that detect TV and video operations; sensors that detect air conditioner operations. There is.

The cloud server 30 is a server for the information terminal 20 and the information terminal 40, and for each target person, the type of solution proposed, the period in which the solution was implemented, the history of execution of the solution, and the measurement data collected within the same period. , record information representing the evaluation of cognitive function, etc. In this embodiment, the timing of proposing a new solution is used as a reference, and the information representing the evaluation of the subject's cognitive function before executing the solution is called "first information", and the information representing the evaluation of the subject's cognitive function before executing the solution is referred to as "first information". The information representing the evaluation of the subject's cognitive function in is called "second information".

Further, the cloud server 30 estimates a solution that is expected to be effective in improving cognitive function for each subject based on the subject's attribute data and collected information, and sends the estimated solution to the information terminal 40. .
The information terminal 40 is, for example, a notebook computer, a desktop computer, a tablet computer, a smartphone, or a wearable computer. The example in FIG. 1 shows a case where the information terminal 40 is a smartphone.
In the example of FIG. 1, it is assumed that the information terminal 40 is different from the information terminal 20, but the information terminal 40 may be the same terminal as the information terminal 20.

<Cloud server configuration>
FIG. 2 is a diagram illustrating a configuration example of the cloud server 30 used in the first embodiment. Note that the cloud server 30 is an example of a cognitive function improvement solution estimation system.
The cloud server 30 includes an arithmetic unit 301 that executes programs, a storage device 302 that stores various data, a communication unit 303 that is used for communication with the network 10 (see FIG. 1), and a bus that connects them. It is made up of wiring 304.

The arithmetic unit 301 includes a CPU (=Central Processing Unit), a ROM (=Read Only Memory) in which BIOS (=Basic Input Output System), etc. are stored, and a RAM (=Random Access Memory) used as a work area. have.
In the case of this embodiment, the arithmetic device 301 realizes the functions of the solution proposal section 311, the evaluation information acquisition sections 312 and 313, and the determination section 314 through the execution of a program by the CPU.

The solution proposal unit 311 in this embodiment is also provided with a sub-function for estimating an optimized solution for each target person.
The sub-function for estimating a solution is used at least when proposing a solution for the second and subsequent times. The sub-function here estimates the next solution to be proposed, depending on whether the cognitive function of the subject who executed the previously proposed solution improved, remained unchanged, or deteriorated as a result of executing the solution.

In the case of this embodiment, four types of solutions are proposed. Specifically, there are four types: diet, exercise, communication, and learning. However, these four types are representative examples of solutions, and other types may be proposed as solutions. Further, the number of types of solutions to be proposed is not limited to four, but may be two, three, five or more.
The solution proposal unit 311 in this embodiment proposes any one of these four types.

The solution proposal unit 311 in the present embodiment changes the method of estimating a solution between the first proposal of a solution and the second and subsequent proposals.
At the time of the first proposal, the solution proposal unit 311 proposes a solution individually determined by, for example, a doctor who examined the subject or an advisor in charge of the subject.
However, at the time of the first proposal, a predetermined solution may be proposed without specifying the target person. Note that, at the time of the first proposal, the solution proposal unit 311 may estimate the solution based on the target person's attribute data. For example, you may give priority to solutions related to things that the target audience likes.

At the time of the second and subsequent proposals, there is at least one solution executed by the subject, and there is also at least one piece of information regarding the quality of execution and the change in evaluation of cognitive function before and after execution. The quality of execution is given by, for example, the content and intensity of execution by the subject.
The solution proposal unit 311 in the present embodiment, when making a proposal for the second time or later, records the history of the solution proposed in the previous time and a plurality of solutions proposed in the previous time, and the cognitive function before and after the execution of the solution by the target person. A newly proposed solution is estimated based on the result of determining the change in the evaluation of .

For example, if the effect of the improvement is confirmed, the solution proposing unit 311 estimates the same solution as the previous solution as the newly proposed solution.
On the other hand, if the effect of the improvement is not confirmed, the solution proposal unit 311 estimates a type different from the previous one as a new solution.
The solution proposal unit 311 proposes one of the types that have not been proposed in the past as a solution until all four types of proposals have gone through one round. At this time, the solution proposal unit 311 refers to information on the target person's preferences and prioritizes solutions that the target person is likely to be interested in.

After one round of proposals, priority is given to types that are estimated to have a high possibility of improving the subject's cognitive function. For example, priority is given to types for which improvement effects have been recognized in the past.
However, even if the type of stimulation has been shown to have an improvement effect in the past, the improvement effect may not be achieved as expected due to continued stimulation of the same type, or the subject's health condition or level of cognitive function may change due to past stimulation. This may be different from the point at which improvement was observed.

In addition, the reason why the improvement effect is not recognized or is low may be found in the target person's implementation at the time. For example, if the quality of execution of the solution is poor.
For this reason, even solutions for which the improvement effect was not obtained or were low when proposed in the past are included in the proposals again in the second round and onwards.
On the other hand, the provision of this service may be once terminated after a predetermined number of proposals or after a predetermined number of visits (for example, two times).

The evaluation information acquisition unit 312 acquires information regarding the evaluation of the subject's cognitive function before executing the solution being executed. The evaluation information acquisition unit 312 is an example of a first acquisition unit in the claims.
In the case of this embodiment, the information regarding the evaluation of cognitive function represents, for example, the degree of normality of cognitive function. Information regarding the evaluation of cognitive function is given, for example, by several levels of indicators or numerical values.

Note that if the solution being executed is the first solution proposed for the subject, the results of a cognitive function test or diagnosis, for example, are used as information regarding the evaluation of the cognitive function.
Furthermore, if there is measurement data that can be used for the subject, the results of processing the measurement data using a known method may be obtained as information on the evaluation of cognitive function.

The evaluation information acquisition unit 313 acquires information regarding the evaluation of the subject's cognitive function after executing the solution currently being executed. The evaluation information acquisition unit 313 is an example of a second acquisition unit in the claims.
Since this is after the target person has executed the solution, the evaluation information acquisition unit 313 acquires the cognitive function evaluation based on the measurement data and execution history data acquired for the target person.

However, the evaluation information acquisition unit 313 may acquire the results of tests and diagnoses performed on the subject after the solution has been executed, as information on the evaluation of cognitive function. Basically, priority is given to the results of diagnosis by a doctor. However, in addition to the results of diagnosis by a doctor, measurement data during execution of the solution may also be included in the acquisition target.
Note that before and after execution of a solution is determined relative to a specific solution, so after execution of one solution corresponds to before execution of the next solution. Therefore, information representing the evaluation of cognitive function obtained after execution of a certain solution is used as information representing evaluation of cognitive function obtained before execution of the next solution.

The determination unit 314 compares the information acquired by the evaluation information acquisition unit 312 and the information acquired by the evaluation information acquisition unit 313, and determines the extent to which cognitive function is improved as a result of executing the solution.
The determination by the determination unit 314 may be a binary result of "cognitive function has improved" and "cognitive function has deteriorated", or a three-value determination result obtained by adding "no change in cognitive function" to these two values. may be used as the determination result.
Further, the determination by the determination unit 314 may be based on the amount of change in a numerical value representing the evaluation of cognitive function.

In this embodiment, the storage device 302 is, for example, a hard disk device. However, the storage device 302 may be a large-capacity semiconductor memory. Further, the storage device 302 may be a storage server connected to the network 10 (see FIG. 1) separately from the cloud server 30, instead of serving as an auxiliary storage device of the cloud server 30.

In the case of this embodiment, the storage device 302 stores an attribute data DB 321, a measurement data DB 322, an execution history DB 323, and an evaluation result DB 324.
The attribute data DB 321 stores attribute data regarding the target persons of this service.
The measurement data DB 322 stores measurement data regarding the target persons of this service.
The execution history DB 323 stores the types and contents of solutions executed by the target persons of this service.
The evaluation result DB 324 stores proposed solutions and the results of judgments regarding cognitive function evaluations after execution of the solutions.

<DB data structure>
FIG. 3 is a diagram illustrating an example of the data structure of a DB stored in the storage device 302 (see FIG. 2). (A) shows an example of the data structure of the attribute data DB 321, and (B) shows an example of the data structure of the measurement data DB 322.
In the case of FIG. 3A, the attribute data DB 321 is composed of gender, age, medical history, physical disability, preference, attribute classification, etc. for each subject. Items other than attribute classification are examples of attribute data.

Note that the preferences are obtained, for example, from responses to a questionnaire. Preferences may be content directly related to the solution or may be content not directly related to the solution. For example, dietary preferences are directly related to the solution. On the other hand, clothing preferences and hobbies are not directly related to the solution.
The attribute classification is information that identifies a set of subjects with similar attribute data. In this embodiment, a set used as a classification is given for one item or a combination of multiple items of attribute data.
In the case of FIG. 3B, the measurement data DB 322 is composed of sleep data, behavior data, etc. for each subject.

FIG. 4 is a diagram illustrating an example of the data structure of another DB stored in the storage device 302 (see FIG. 2). (A) shows an example of the data structure of the execution history DB 323, and (B) shows an example of the data structure of the evaluation result DB 324. The data structure examples shown in FIGS. 4(A) and 4(B) correspond to subject A.
In the case of FIG. 4A, the execution history DB 323 is configured with the date on which the solution was executed, the type of solution, the details of the execution, etc. for each target person.
The content of execution also includes information on the intensity at the time of execution. The intensity includes, for example, the number of times and execution time.

For example, for "meal", the content, number of items, calories, number of meals, etc. of the meal are stored. This information is entered by the subject himself or herself or by the user of the service. Note that the content and number of items of the meal may be registered by image recognition or the like from a photograph of the meal.
For example, for "exercise", the content of the exercise, the time spent exercising, the number of times the exercise was exercised, the intensity, etc. are stored.
For example, for "communication", the number of conversations, number of people, duration of conversations, etc. are stored.
For example, in the case of ``study'', the number of words written, the type and difficulty of the book read, the number of pages read, the number of characters read, the number and difficulty of questions calculated, the correct answer rate, etc. are stored.

In the case of FIG. 4B, the evaluation result DB 324 is configured with execution period, solution type, determination result, etc. for each subject.
In the example of FIG. 4(B), new solutions are proposed every two months. However, for solutions for which an improvement is recognized in the judgment results, the same solutions are proposed for the next period as well. In the example of FIG. 4(B), learning is proposed twice in a row.

<Example of processing operation>
FIG. 5 is a flowchart illustrating an example of processing operations executed by the cloud server 30 assumed in the first embodiment. Note that the symbol S shown in the figure means a step. The processing method shown in FIG. 5 is an example of a cognitive function improvement solution estimation method.
When providing the service, the cloud server 30 acquires information representing the evaluation of the subject's cognitive function (step 101). For example, the results of a cognitive function test or the results of a diagnosis by a doctor are used as the information representing the evaluation of the subject's cognitive function. The information here is an example of the first information in the claims.

Next, the cloud server 30 acquires attribute data of the target person (step 102).
Next, the cloud server 30 proposes a solution that is expected to improve the cognitive function of the subject (step 103). Note that the proposed solution may be estimated based on the target person's attribute data.
After proposing the solution, the cloud server 30 acquires the subject's execution history and measurement data for the proposed solution (step 104). These data are notified from the information terminal 20 (see FIG. 1) to the cloud server 30.

When the solution execution period ends, the cloud server 30 acquires information representing the evaluation of the subject's cognitive function after the solution execution (step 105). The information acquired in step 105 corresponds to an example of the second information in the claims regarding the latest solution executed. However, from the perspective of the new solution estimated in step 107, this information is an example of the first information in the claims.

Next, the cloud server 30 determines a change in the evaluation of cognitive function due to execution of the solution (step 106).
Thereafter, the cloud server 30 estimates and proposes a new solution based on the determination result (step 107).
Thereafter, the cloud server 30 uses steps 103 to 107 as one cycle to propose new solutions for the next time onwards, and determines whether the proposed solution improves the cognitive function.

FIG. 6 is a diagram illustrating changes in cognitive function evaluation due to execution of the proposed solution. (A) is an example of a change in cognitive function when a solution is proposed based on the proposer's experience, (B) is an example of a change in cognitive function when a solution is proposed by this service, and (C) is an example of a change in cognitive function when a solution is proposed by this service. This is an example of changes in cognitive function when a solution is proposed taking preferences into consideration.

Note that the vertical axis in FIGS. 6(A) to (C) is the score in the cognitive function test, and the horizontal axis is time. Higher scores indicate better cognitive function.
As shown in FIG. 6A, when a solution is proposed based on experience, it depends on the proposer's past experience, and the compatibility between the target person and the solution is neglected.
Furthermore, even when the proposed solution is determined programmatically, it is easy to simply apply solutions that are statistically more effective in order, and many proposals are required to arrive at a solution that is most effective for the target audience.

On the other hand, as shown in Figure 6 (B), when a solution is proposed using this service, the score rises and falls until the proposed solution goes through one round, but it is possible to find a solution that is effective for each target person. After that, it is possible to continuously propose highly effective solutions, so that improvements in cognitive function begin early.
Furthermore, as shown in FIG. 6(C), when a solution is proposed in consideration of the subject's preferences, further improvement in cognitive function can be expected. Solutions that everyone likes or is interested in are easier to continue and the quality of execution is easier to improve than solutions that everyone dislikes or is not interested in. This makes it easier to discover solutions that are expected to produce improvement effects from an early stage.

<Summary>
As mentioned above, by repeating the loop of proposing new solutions, collecting execution history and measurement data, and determining improvement in cognitive function, we can propose solutions tailored to the target audience, that is, personalize them, in a small number of times. Ru.
Furthermore, by preferentially proposing solutions that the subject likes or is interested in, it is possible to increase the possibility that cognitive function will improve.

In addition, the accuracy of the proposed solution is estimated by estimating the new solution in step 107 (see Figure 5), including the execution history and measurement data of the solution by the target person acquired in step 104 (see Figure 5). It is possible to increase it. For example, since the content and strength of the executed solution can be obtained from the execution history, it is possible to analyze the reasons why cognitive function did not improve or the reasons why it did improve, and it is possible to improve the accuracy of estimation of newly proposed solutions. become. For example, it is possible to include changes in the content or strength of the same solution in the proposal. Such proposals cannot be achieved by focusing only on solution types (major classifications).

Furthermore, in the description of this embodiment, the solution is executed using four types (major classifications) as units, but it is also possible to propose solutions using intermediate classifications or small classifications as units, which are lower layers of each type. For example, instead of meals, solutions may be proposed based on categories such as Japanese food and Western food, or solutions may be proposed based on categories such as meat dishes and fish dishes. Regarding exercise, communication, and learning, intermediate classifications and small classifications may be used as units of proposal.

When proposing solutions based on intermediate classifications or minor classifications, the number of solutions that can be proposed increases, and with round-robin proposals, it takes time to discover solutions that have an improvement effect on the target audience.
However, the method described in this embodiment not only compares information representing the evaluation of cognitive functions before and after execution of the solution, but also considers attribute data including the subject's preferences, the history of solution execution, and measurement data. This makes it possible to narrow down solutions efficiently.

<Embodiment 2>
In this embodiment, a case will be described in which, before estimating a new solution, a function is added to acquire the subject's impressions and answers regarding likes and dislikes regarding the executed solution and to modify attribute data. The opinions or answers regarding likes and dislikes here are examples of the subject's opinions in the scope of the claims.
Note that the configuration of a cloud system 1 (see FIG. 1) assumed in this embodiment is the same as that in the first embodiment.

FIG. 7 is a diagram illustrating a configuration example of the cloud server 30 used in the second embodiment. In FIG. 7, parts corresponding to those in FIG. 2 are shown with corresponding symbols.
The cloud server 30 shown in FIG. 7 differs from the cloud server 30 shown in FIG. 2 in that an attribute data correction unit 315 is added.
The attribute data correction unit 315 has a function of correcting the attribute data of the target person stored in the attribute data DB 321 when the results of a questionnaire etc. conducted on the target person who executed the solution are obtained from the information terminal 20. provide. For example, the attribute data modification unit 315 modifies the content of preferences in the attribute data DB 321 shown in FIG. 3(A).

In the preference column, in addition to likes and dislikes for the type of solution executed, likes and dislikes for medium-class and small-class solutions are added or modified.
Addition is performed in the case of unregistered information. Modification is performed on registered information.
For example, I wasn't interested when I ran it last time, but I find it interesting this time. In addition, your impression of the same event may change depending on your physical condition at the time of execution or the accumulation of experience. For example, if you see the same solution in succession, you may lose interest.
In this way, preferences are information that is relatively likely to change, and it is necessary to reflect the latest information of the subject when estimating a solution in order to increase the effect of improving cognitive function.

FIG. 8 is a flowchart illustrating an example of a processing operation executed by the cloud server 30 assumed in the second embodiment. In FIG. 8, parts corresponding to those in FIG. 5 are shown with corresponding symbols. The processing method shown in FIG. 8 is also an example of a cognitive function improvement solution estimation method.
In the case of FIG. 8, between step 104 and step 105, the cloud server 30 obtains the subject's impressions regarding the executed solution and corrects the attribute data (step 201). As described above, the modified attribute data is used when estimating a new solution in step 107.

Note that modification of attribute data is not limited to comments. For example, corrections may be made such as addition or reduction of limb incapacitated parts. Additionally, if the date of birth is not registered, the age will also be subject to correction.
In the case of this embodiment, an opportunity is ensured for information on impressions regarding the executed solution to be reflected in the target person's attribute data. Therefore, as a newly proposed solution, a solution that the subject likes or is interested in is more likely to be selected, and it becomes possible to increase the possibility that cognitive function will improve.

<Embodiment 3>
In the embodiment described above, a method was described in which one of four types of solutions prepared in advance or one of a plurality of solutions belonging to a lower hierarchy thereof is proposed for each target person.
However, the cloud server 30 may propose a combination of multiple solutions. For example, learning and exercise may be proposed as solutions to subject A, and exercise, communication, and diet may be proposed to subject B as solutions.

By implementing a combination of solutions that are of high interest to the subject, it is possible to increase the likelihood that cognitive function will improve compared to implementing only a single solution. Note that the combinations that produce improvement effects are not limited to combinations of solutions that individually have high improvement effects. For example, even if the improvement effect is low when executed alone, a high improvement effect may be obtained when executed in combination.

However, when combining a plurality of solutions, the number of solutions that can be proposed increases compared to the first embodiment. For example, even if there are four types of solutions prepared in advance, there will be a total of 15 solutions: any one, any two combinations, any three combinations, or all four.

Therefore, it is necessary to efficiently search for and propose combinations with high improvement effects in order to improve the cognitive function of the target person. However, in the embodiment described above, attribute data including the target person's preferences, Since the solution is estimated by taking into account the content and strength of the executed solution, as well as measurement data during execution, it is possible to combine solutions with high improvement effects more efficiently or with fewer trials than when relying on empirical rules. can be found. As a result, it becomes possible to enhance the effect of improving the cognitive function of the subject.

In this embodiment, Bayesian optimization aimed at optimizing a black box function is employed to estimate a combination of solutions with a high improvement effect. In the case of this embodiment, the inputs are the proposed solution, execution history, measurement data, target person's attribute data, information representing the evaluation of cognitive function before executing the proposed solution, and information after executing the proposed solution. The information represents the evaluation of cognitive function, and the output is a combination of newly proposed solutions. The attribute data also includes the target person's preferences.
Note that Bayesian optimization is an example of an estimation algorithm, and other methods can also be adopted.
Bayesian optimization can also be used to estimate solutions in the first and second embodiments.

<Embodiment 4>
In the first embodiment, the solution to be proposed to the target person is basically estimated by taking into account the proposal history of the four types of solutions that can be proposed, the target person's preferences, etc. The cloud server 30 preferentially proposes as a new solution a solution that has been proven to have an improvement effect on the classification to which the target person's attribute data belongs.

FIG. 9 is a diagram illustrating a solution proposal method in the fourth embodiment. (A) shows the distribution of attribute data and classifications G1 to G8 corresponding to each target person, and (B) shows an example of correspondence between classification and solution.
Note that the circles in FIG. 9(A) correspond to each subject. Further, the distribution of circles represents the positional relationship when attribute data is mapped onto the attribute space.
In this embodiment, recommended solutions are determined for each classification, as shown in FIG. 9(B).

For example, a combination of learning and exercise is recommended for subjects belonging to classification G1.
Furthermore, for subjects belonging to category G2, eating is recommended, and for subjects belonging to category G3, a combination of exercise, communication, and eating is recommended.
FIG. 9(B) shows classifications G2, G6, G7, and G8 to which only one solution is assigned, classifications G1, G4, and G5 to which a combination of two solutions is assigned, and classification G3 to which a combination of three solutions is assigned. It is shown.
By preferentially proposing a solution that has an improvement effect on other subjects with similar attribute data, it is possible to increase the possibility that the subject's cognitive function will improve by implementing the proposed solution.

<Embodiment 5>
In the first embodiment, a case has been described in which a solution that has a proven track record for the classification to which the target person's attribute data belongs is preferentially proposed as a new solution for the target person.
However, while proposals based on classification are expected to bring about statistical improvements, depending on the granularity of the classification, the difference from the target person's attribute data may become large.
Therefore, a method will be described in which the cloud server 30 (see FIG. 1) in this embodiment preferentially proposes solutions that have an improvement effect for other subjects whose attribute data are close to each other within the same classification. .

FIG. 10 is a diagram illustrating a solution proposal method in the fifth embodiment.
The circles in FIG. 10 correspond to each subject. Further, the distribution of circles represents the positional relationship when attribute data is mapped onto the attribute space. The closer the distance between the circles, the higher the degree of matching of the attribute data, and the farther the distance between the circles, the lower the degree of matching of the attribute data.
In the case of FIG. 10, the target of this service belongs to either classification G1 or classification G2. In the case of classification G1, subject A, subject B, and subject C belong to it. It can be seen that although the distance between the attribute data of subject A and subject B is close, the distance between the attribute data of subject A and subject C is long even within the same classification.

A solution using classification as a unit is a solution that has an improvement effect on the target person as a representative value using classification as a unit. Therefore, in the case of a subject who belongs to classification G1 but whose attribute data is mapped near the outer edge, the distance from the representative value becomes large.
Therefore, by adopting a method of proposing a solution to target person A that has an improvement effect on target person B, as in this embodiment, there is a high possibility that target person A's cognitive function will improve as well. Become.
Thereby, it is possible to increase the possibility that a solution or a combination of solutions with an improvement effect will be proposed with a small number of trials.

<Other embodiments>
(1) Although the embodiments of the present invention have been described above, the technical scope of the present invention is not limited to the range described in the above embodiments. It is clear from the claims that various changes or improvements made to the embodiments described above are also included within the technical scope of the present invention.

(2) In the first to fifth embodiments described above, when proposing a solution from the second time onwards, the attribute data of the target person is referred to and the solution to be applied preferentially is estimated. By referring to the target person's attribute data and determining the solution to be proposed when making a proposal, it becomes possible to propose a solution with a high improvement effect from the beginning.

(3) In the first to fifth embodiments described above, the solution proposed by the cloud server 30 is estimated and proposed to the service user, but the corresponding processing function is executed by the information terminal 20 or the information terminal 40. You may. That is, the solution may be estimated through the execution of a program installed on the information terminal 20 or the information terminal 40, rather than as a cloud service. The information terminal 20 and the information terminal 40 in this case are an example of a cognitive function improvement solution estimation system.

Further, the processing functions corresponding to the cloud server 30 may be executed by cooperation between the cloud server 30 and the information terminals 20 and 40. In other words, the corresponding processing functions may be distributed across multiple terminals. In this case, multiple terminals involved in realizing the processing function correspond to the cognitive function improvement solution estimation system.

1... Cloud system, 10... Network, 20, 20A, 20B, 40... Information terminal, 301... Arithmetic device, 311... Solution proposal section, 312, 313... Evaluation information acquisition section, 314... Judgment section, 321... Attribute data DB , 322...Measurement data DB, 323...Execution history DB, 324...Evaluation result DB

Claims (11)

a proposal unit that proposes to the user a solution that combines one or more of a plurality of solutions including one or more of learning, diet, exercise, and communication for the purpose of improving cognitive function;
a first acquisition unit that acquires first information representing an evaluation of the subject's cognitive function before implementation of the one or more solutions combined;
a second acquisition unit that acquires second information representing an evaluation of the subject's cognitive function after implementation of the one or more solutions combined;
a determination unit that determines, based on the first information and the second information, a change in the evaluation of the subject's cognitive function that occurred before and after the implementation of the one or more solutions combined;
has
The proposing unit estimates one or more solutions to be newly proposed based on the history of solutions proposed in the past and the result of the determination by the determining unit.
Cognitive function improvement solution estimation system. The first information and the second information are given as a result of analyzing measurement data measured in different periods,
The cognitive function improvement solution estimation system according to claim 1. The first information and the second information are given as a result of a test regarding the cognitive function of the subject,
The cognitive function improvement solution estimation system according to claim 1. The proposal unit proposes a solution that combines one or more of the recommended attributes of the target person.
The cognitive function improvement solution estimation system according to claim 1. The proposal unit proposes one or a combination of solutions recommended for the classification to which the attribute data belongs;
The cognitive function improvement solution estimation system according to claim 4. The proposal unit proposes a solution that combines one or more recommended solutions for other subjects to which the attribute data belongs;
The cognitive function improvement solution estimation system according to claim 4. The proposal unit estimates one or more newly proposed solutions based on the determination result of the determination unit and the attribute data.
The cognitive function improvement solution estimation system according to any one of claims 4 to 6. The cognitive function improvement according to claim 7, further comprising: a modification unit that corrects the attribute data based on an opinion regarding the solution combining one or more obtained from the subject after implementing the solution combining the one or more. Solution estimation system. The proposal unit estimates, for each subject, a new solution combining one or more solutions that improves the evaluation of the subject's cognitive function using a Bayesian optimization method.
The cognitive function improvement solution estimation system according to claim 1. A cognitive function improvement solution estimation method realized through the execution of a program by a computer,
A process of proposing for each subject a solution that combines one or more of a plurality of solutions, including one or more of learning, diet, exercise, and communication, for the purpose of improving cognitive function;
A process of acquiring first information representing an evaluation of the subject's cognitive function before implementation of the solution combining one or more of the above,
A process of acquiring second information representing an evaluation of the subject's cognitive function after implementation of the one or more solutions combined;
A process of determining, based on the first information and the second information, a change in the evaluation of the subject's cognitive function that occurred before and after the implementation of the one or more solutions combined;
Based on the result of the determination, a process of estimating one or more newly proposed solutions;
A method for estimating a solution for improving cognitive function. to the computer,
A function to obtain first information representing an evaluation of the subject's cognitive function before implementing a solution that combines one or more of a plurality of solutions including one or more of learning, diet, exercise, and communication. and,
a function of acquiring second information representing an evaluation of the subject's cognitive function after implementation of the solution combining one or more of the above;
A function of determining, based on the first information and the second information, a change in the evaluation of the subject's cognitive function that occurred before and after the implementation of the one or more solutions combined;
A function of estimating one or more newly proposed solutions based on the result of the determination;
A program to make this happen.