Program, information processing method, and information processing device

Abstract

A program according to an embodiment of the present disclosure acquires subject data including an answer to a question regarding a meal of a subject, a meal photograph, and glucose data, outputs a plurality of action targets regarding lifestyle improvement on the basis of the subject data, and stores an action target selected by the subject, from among the plurality of action targets, in association with identification information of the subject.

Description

Program, Information Processing Method, and Information Processing Apparatus

[0001] The present technology relates to a program, an information processing method, and an information processing apparatus.

[0002] Conventionally, a system for promoting improvement of lifestyle has been proposed. For example, the lifestyle analysis system described in Patent Document 1 quantifies a lifestyle from body measurement data and meal record data of a subject, and extracts habits that cause lifestyle diseases such as metabolic syndrome and diabetes from each parameter of the quantified lifestyle analysis items, and provides a specific improvement method by teaching the corresponding foods in the subject's own meal record.

[0003] Japanese Patent Application Publication No. 2013 - 097403

[0004] However, the system described in Patent Document 1 only provides a method for improving lifestyle, and does not consider at all a method for making the target person continue the actions to be improved.

[0005] The present disclosure has been made in view of such circumstances, and an object thereof is to provide a program or the like that can make a target person continue actions to improve lifestyle.

[0006] The program according to an embodiment of the present disclosure acquires subject data including answers to questions regarding the subject's diet, diet photographs, and glucose data, outputs a plurality of action goals related to lifestyle improvement based on the subject data, and stores the action goal selected by the subject from the plurality of action goals in association with the identification information of the subject.

[0007] In the program according to an embodiment of the present disclosure, it is possible to make a target person continue actions to improve lifestyle.

[0008] This is an explanatory diagram showing an overview of the information processing system. This is a block diagram showing an example of the configuration of the information processing device. This is a block diagram showing an example of the configuration of the terminal device. This is an explanatory diagram showing an example of the record layout of the target DB. This is an explanatory diagram showing an example of the record layout of the glucose DB. This is an explanatory diagram showing an example of the record layout of the question and answer DB. This is an explanatory diagram showing an example of the record layout of the behavioral goal DB. This is an explanatory diagram showing an example of the glucose data transmission screen. This is an explanatory diagram showing an example of the question and answer screen. This is an explanatory diagram showing an example of inputting reference information to the language model related to behavioral goal output processing. This is an explanatory diagram showing an example of inputting prompts to the language model related to behavioral goal output processing. This is an explanatory diagram showing an example of the output of the language model related to behavioral goal output processing. This is an explanatory diagram showing an example of the behavioral goal selection screen. This is an explanatory diagram showing an example of the implementation status input screen. This is a flowchart showing an example of behavioral goal output and selection processing. This is an explanatory diagram showing an example of implementation status input registration processing. This is an explanatory diagram showing an example of inputting reference information to the language model related to behavioral goal output processing according to Embodiment 2. This is an explanatory diagram showing an example of inputting prompts to the language model related to behavioral goal output processing according to Embodiment 2. This is an explanatory diagram showing an example of the output of the language model related to behavioral goal output processing according to Embodiment 2. This is an explanatory diagram showing an overview of the information processing system according to Embodiment 3. This is a block diagram showing an example of the configuration of a medical terminal. This is an explanatory diagram showing an example of an approval information input screen. This is an explanatory diagram showing an example of a chat screen. This is a flowchart showing an example of behavioral goal output and selection processing according to Embodiment 3. This is an explanatory diagram showing an example of input to a language model related to advice output processing. This is an explanatory diagram showing an example of output from a language model related to advice output processing. This is an explanatory diagram showing an example of a chat screen according to Embodiment 4.

[0009] (Embodiment 1) The information processing device 1 according to Embodiment 1 outputs comments and behavioral goals related to lifestyle improvement by inputting the subject's answers to questions about their health status, lifestyle, or diet, meal photos, and subject data into a language model M. Embodiment 1 will be described below with reference to the drawings.

[0010] Figure 1 is an explanatory diagram showing an overview of the information processing system S. The information processing system S comprises an information processing device 1, a terminal device 2, and a language model server 3. The information processing device 1 outputs questions about the subject's health status, lifestyle, or diet to the terminal device 2 and obtains answers to the questions from the terminal device 2. The information processing device 1 transmits the answers to the questions about health status, lifestyle, or diet to the language model server 3 and inputs them into the language model M of the language model server 3. In the following, the transmission of information to the language model server 3 and the input of information to the language model M are considered synonymous. The information processing device 1 obtains comments related to the answers output by the language model M that received the answers to the questions about health status, lifestyle, or diet, and recommended behavioral goals for the subject. The information processing device 1 stores the obtained comments and behavioral goals and transmits them to the terminal device 2 upon request from the terminal device 2. The terminal device 2 is, for example, a terminal device owned by a patient (subject) with a lifestyle-related disease such as obesity, diabetes, hyperlipidemia, or hypertension.

[0011] Figure 2 is a block diagram showing an example configuration of the information processing device 1. The information processing device 1 is, for example, a server computer and comprises a processing unit 11, a storage unit 12, and a communication unit 13. The processing unit 11 is composed of a CPU (Central Processing Unit), MPU (Micro Processing Unit), GPU (Graphical Processing Unit), or quantum processor, and performs various control processing, arithmetic processing, etc., by reading and executing a program P (program product) and a database pre-stored in the storage unit 12. A database server or the like may be provided outside the information processing device 1, and the database may be read from the database server or the like. Furthermore, the information processing device 1 may have its functions realized by multiple server devices or computers. Also, the information processing device 1 may correspond to a node on a blockchain. Furthermore, the processing unit 11 of the information processing device 1 may execute some or all of the processing performed by the terminal device 2.

[0012] The storage unit 12 of the information processing device 1 is, for example, a volatile memory and a non-volatile memory. The storage unit 12 stores a program P, a target database 121, a glucose database 122, a question and answer database 123, and a behavioral target database 124. The program P may be provided to the information processing device 1 using a storage medium 12a that is readable by a computer. The storage medium 12a is, for example, a portable memory. Examples of portable memory include CD-ROMs, USB (Universal Serial Bus) memory, SD cards, microSD cards, or CompactFlash memory (registered trademark). If the storage medium 12a is a portable memory, the processing element of the processing unit 11 may read the program P from the storage medium 12a using a reading device (not shown). The read program P is written to the storage unit 12. Furthermore, the program P may be provided to the information processing device 1 by the communication unit 13 communicating with an external device. Details of the target database (DB) 121, glucose database 122, questionnaire response database 123, and behavioral goal database 124 will be described later.

[0013] The communication unit 13 is a communication module or communication interface for communicating with the terminal device 2 and the language model server 3 by wire or wireless connection, and is, for example, a wide-area wireless communication module such as LTE®, 4G, or 5G. The processing unit 11 communicates with the terminal device 2 and the language model server 3 via the communication unit 13, for example, through an external network N such as the Internet.

[0014] The language model server 3 stores the language model M. Details of the language model M will be described later. Note that the language model M may also be stored in the information processing device 1 or the terminal device 2.

[0015] Figure 3 is a block diagram showing an example configuration of the terminal device 2. The terminal device 2 is, for example, a smartphone, a tablet terminal, or a personal computer. In this embodiment, the example of the terminal device 2 being a smartphone will be described. The terminal device 2 comprises a terminal processing unit 21, a storage unit 22, a communication unit 23, an input unit 24, a display unit 25, and a shooting unit 26. The terminal processing unit 21 is composed of a CPU or MPU, and performs various control processing, calculation processing, etc. The terminal device 2 may perform all or part of the processing performed by the information processing device 1 or the language model server 3.

[0016] The storage unit 22 stores an application program Pa that displays questions about health status, lifestyle, or diet acquired from the information processing device 1 and accepts input of answers, and displays multiple behavioral goals acquired from the information processing device 1 and accepts selection of behavioral goals. The application program Pa is provided to the terminal device 2, for example, using a storage medium 22a. The terminal processing unit 21 of the terminal device 2 may acquire the application program Pa using the internet and store it in the storage unit 22.

[0017] The input unit 24 receives input from the subject, such as answers to questions about their health status, lifestyle, or diet, and selection of behavioral goals. The display unit 25 displays the questions and multiple behavioral goals obtained from the information processing device 1. If the terminal device 2 is a smartphone, the input unit 24 and the display unit 25 are integrated by a touch panel.

[0018] The shooting unit 26 takes a photograph of the subject's meal. If the terminal device 2 is a smartphone, the shooting unit 26 is composed of the smartphone's camera function.

[0019] Figure 4 is an explanatory diagram showing an example of the record layout of the subject DB 121. The subject DB stores the results of the subject's answers to questions and the results of their examinations. The management items (fields) of the subject DB 121 include the subject ID field, glucose data field, question answer field, taste test result field, IPAQ field, exercise record field, Athens Insomnia Scale field, cooking ability field, DTR-QOL field, SF-36 field, MEQ field, and SESD field. For any field in the subject DB 121 that does not have data to store, an empty value is stored.

[0020] The Subject ID field stores the ID assigned to the subject (Subject ID). The Glucose Data field stores the ID of the glucose data database where the subject's glucose data is stored (Glucose Data ID). The Question Answer field stores the answers to questions about the subject's health status, lifestyle, or diet. The Taste Test Results field stores the results of the subject's taste test.

[0021] The IPAQ field stores the subjects' responses to the International Physical Activity Questionnaire (IPAQ). The IPAQ is a questionnaire designed to investigate the physical activity of adults. The IPAQ field stores data files of the subjects' responses to each item of the IPAQ, for example. Note that the subject DB121 may also include fields where each question of the IPAQ is expanded into individual items.

[0022] The exercise record field stores data related to the subject's exercise. The exercise record data stored in the exercise record field includes, for example, the number of steps, walking distance, exercise intensity, or exercise time obtained from a smartwatch worn by the subject, or the details of the exercise entered by the subject in terminal device 2.

[0023] The Athens Insomnia Scale field stores the self-assessment responses of the participant to the Athens Insomnia Scale. The Athens Insomnia Scale is a questionnaire used to assess the severity of insomnia. The Athens Insomnia Scale field stores, for example, a data file containing the participant's responses to each item of the Athens Insomnia Scale. Note that participant DB121 may also include fields where each question of the Athens Insomnia Scale is expanded into individual items.

[0024] The Cooking Ability field stores the answers to questions about the subject's cooking ability. The data stored in the Cooking Ability field (Cooking Ability Data) includes, for example, the subject's cooking frequency or their specialty dishes.

[0025] The DTR-QOL field stores the subjects' responses to the Diabetes Therapy-Related Quality of Life (DTR-QOL) questionnaire. For example, the DTR-QOL field stores data files containing the subjects' responses to each item of the DTR-QOL questionnaire. The subject DB121 may also include fields where each question of the DTR-QOL questionnaire is broken down into individual items.

[0026] The SF-36 field stores the subjects' responses to a questionnaire (SF-36® registered trademark) used to measure health-related quality of life. The SF-36 field stores, for example, data files of the subjects' responses to each item of the SF-36. The subject DB121 may also include fields in which each question of the SF-36 is expanded item by item.

[0027] The MEQ field stores the subjects' responses to the Morningness-Evening Questionnaire (MEQ). The MEQ field contains data files, for example, the subjects' responses to each item of the MEQ, or scores calculated based on those responses. The subject DB121 may also include fields where each question of the MEQ is broken down into individual items.

[0028] The SESD field stores the subjects' responses to the Self-Efficacy Scale for Diabetes self-care (SESD) questionnaire. For example, the SESD field stores data files containing the subjects' responses to each item of the SESD. Note that subject DB121 may also include fields where each question of the SESD is expanded into individual items.

[0029] The information stored in the IPAQ field, Athens Insomnia Scale field, Cooking Ability field, DTR-QOL field, SF-36 field, MEQ field, and SESD field includes data obtained from responses to questionnaires given to subjects in advance or from interviews conducted by healthcare professionals.

[0030] Figure 5 is an explanatory diagram showing an example of the record layout of the glucose DB 122. The glucose DB 122 stores photographs of meals consumed by the subject and the changes in blood glucose concentration (glucose data) within a certain period of time after meal consumption. The management items (fields) of the glucose DB 122 include a date field, breakfast time field, breakfast meal photo field, lunch time field, lunch meal photo field, dinner time field, dinner meal photo field, and multiple glucose value fields. The glucose DB 122 also includes attribute information such as the subject ID and glucose data ID. In other words, the storage unit 12 stores glucose DB 122 for multiple subjects, and multiple glucose DB 122 can be identified by the subject ID and glucose data ID.

[0031] The date field stores the date the glucose data was collected. The breakfast time field stores the time the subject ate breakfast. The breakfast meal photo field stores a photo of the breakfast the subject ate. The lunch time field stores the time the subject ate lunch. The lunch meal photo field stores a photo of the lunch the subject ate. The dinner time field stores the time the subject ate dinner. The dinner meal photo field stores a photo of the dinner the subject ate. The breakfast time field, lunch time field, and dinner time field store the time entered, for example, on terminal device 2. The breakfast meal photo field, lunch meal photo field, and dinner meal photo field store the meal photos taken on terminal device 2, for example, in file format. The glucose DB 122 may also include a snack time field and a snack meal photo field depending on the snacks the subject ate.

[0032] The glucose DB 122 includes multiple glucose value fields that store glucose values, for example, every 15 minutes. Each glucose value field stores the glucose value of the subject measured at the corresponding time. The glucose value is measured, for example, by a blood glucose meter attached to the subject, and is transmitted from the terminal device 2, which acquired the glucose value from the blood glucose meter, to the information processing device 1.

[0033] Figure 6 is an explanatory diagram showing an example of the record layout of the question-answer database 123. The question-answer database 123 stores the subject's answers to questions about diet, including questions about eating behavior and questions about food preferences, as well as questions about health status or lifestyle. The management items (fields) of the question-answer database 123 include a question number field, a question content field, and an answer field. The question-answer database 123 also includes attribute information such as the subject ID, question-answer ID, and answer recording date. In other words, each time a subject answers a question about diet, the question-answer database 123 is stored in the storage unit 12. Multiple question-answer databases 123 can be identified by the subject ID and question-answer ID.

[0034] The Question Number field stores the number assigned to the question regarding the subject's health status, lifestyle, or diet. The Question Content field stores the content of the question regarding health status, lifestyle, or diet. The Question Answer field stores the subject's answers to the questions regarding health status, lifestyle, or diet. Details of the questions regarding health status, lifestyle, or diet will be described later.

[0035] Figure 7 is an explanatory diagram showing an example of the record layout of the Behavioral Goal DB 124. The Behavioral Goal DB 124 stores the behavioral goals selected by the subject and the status of the subject's implementation of those goals. The management items (fields) of the Behavioral Goal DB 124 include a selection date field, a behavioral goal field, and implementation status fields related to multiple dates. The Behavioral Goal DB 124 also includes the subject ID and behavioral goal ID as attribute information. In other words, the storage unit 12 stores multiple Behavioral Goal DBs 124 for each subject, and multiple Behavioral Goal DBs 124 can be identified by the subject ID and behavioral goal ID.

[0036] The "Selected Date" field stores the date on which the subject selected an action goal from among multiple action goals output by the language model M on terminal device 2. The "Action Goal" field stores the selected action goal. Each of the "Implementation Status" fields for multiple dates stores the implementation status of the action goal for each day, as entered by the subject on terminal device 2. The "Implementation Status" field stores either a flag indicating that the action goal was implemented (implemented) or a flag indicating that the action goal was not implemented (not implemented).

[0037] Figure 8 is an explanatory diagram showing an example of a glucose data transmission screen. The glucose data transmission screen is displayed on the display unit 25 of the terminal device 2 and includes a date input field, a glucose data display field, and a meal photo input field. The date on which the glucose data was collected is entered in the date input field. In the glucose data display field, glucose data measured every 15 minutes by, for example, a blood glucose meter attached to the subject is displayed in graph format. The graph is displayed with time on the horizontal axis and glucose value (unit: mg / dL) on the vertical axis, and the glucose value for the subject measured every 15 minutes is plotted. The information processing device 1 may also acquire glucose data of the subject measured at a medical institution, or glucose data read from a blood glucose meter brought to a medical institution. Alternatively, the information processing device 1 may sequentially acquire and store the subject's glucose data by communicating with the blood glucose meter.

[0038] The meal photo input field includes fields for breakfast time, breakfast meal photo, lunch time, lunch meal photo, dinner time, and dinner meal photo. The breakfast time, lunch time, and dinner time input fields are used to input the times when the subject ate breakfast, lunch, and dinner, respectively. The breakfast meal photo input field, lunch meal photo input field, and dinner meal photo input field are used to input the meals taken by the camera unit 26 and stored in the memory unit 22, respectively.

[0039] Furthermore, the glucose data transmission screen displays a function button (send button) for sending glucose data, meal time, and a photo of the meal. When the send button is pressed, the terminal processing unit 21 of the terminal device 2 transmits the glucose data, meal time, and photo of the meal related to the entered date to the information processing device 1. The processing unit 11 of the information processing device 1, having acquired the glucose data, meal time, and photo of the meal, stores the acquired glucose data, meal time, and photo of the meal in the glucose DB 122.

[0040] Figure 9 is an explanatory diagram showing an example of a question-answering screen. The question-answering screen is displayed on the display unit 25 of the terminal device 2. The question-answering screen displays multiple questions related to meals, including questions about eating behavior, questions about food preferences, or questions about seasoning, as well as questions about the subject's health status or lifestyle, and answers to each question are entered.

[0041] In the example shown in Figure 9, questions about eating behavior include questions about the subject's eating speed, eating methods, feelings towards food, or meal times (question numbers: 1-10). Questions about food preferences include questions about the subject's evaluation of how much they want to eat specific foods (question numbers: 11-20). Questions about seasoning include questions about the subject's preferred seasoning for food (question numbers: 21-24). Questions about the subject's health status include questions about their history of lifestyle-related diseases (question numbers: 25-28). Questions about the subject's lifestyle include, for example, whether they work night shifts (question number: 29) and whether they smoke (question number: 30). In the example shown in Figure 9, the answer input field for each question may contain, for example, an evaluation value on a 10-point scale. Note that, for example, the answers to the questions about the subject's health status and lifestyle shown in Figure 9 may be entered as binary values ​​of "yes" or "no".

[0042] Furthermore, a function button (send button) for sending answers to questions is displayed on the question answer screen. When the send button is pressed, the terminal processing unit 21 of the terminal device 2 sends the answers to the entered questions regarding health status, lifestyle, or diet to the information processing device 1. The processing unit 11 of the information processing device 1, having obtained the answers to the questions regarding health status, lifestyle, or diet, stores the answers to the questions regarding health status, lifestyle, or diet in the question answer DB 123.

[0043] FIG. 10 is an explanatory diagram showing an example of input of reference information to the language model M related to the action target output process. FIG. 11 is an explanatory diagram showing an example of input of a prompt to the language model M related to the action target output process. The language model M is configured by, for example, GPT (Generative Pretrained Transformer). The language model M has an aspect as an LLM (Large Language Model) that outputs according to the instructions included in the prompt based on the input prompt and data, and an aspect as a VLM (Vision Language Model) that outputs according to the instructions included in the prompt based on the input prompt and image. Further, the language model M may be an MLLM (Multimodal Large Language Model).

[0044] In the action target output process, the language model M is input with a meal image, records of the subject DB121, glucose DB122, question and answer DB123, action target examples, and a prompt. The meal images input to the language model M include the meal images of breakfast, lunch, and dinner stored in the glucose DB122 input to the language model M. Note that the content of the meal output by inputting the meal image into an image recognition model such as Yolo (You only Look at once) may be input to the language model M.

[0045] The record of the subject DB121 input to the language model M in the action target output process is a record read out using the subject ID of the subject for whom the action target is output as a key. The glucose DB122 input to the language model M in the action target output process is a glucose DB122 having the glucose data ID stored in the read record of the subject DB121 as attribute information. The question and answer DB123 input to the language model M in the action target output process is a question and answer DB123 having the question and answer ID stored in the read record of the subject DB121 as attribute information. The record of the subject DB121, the glucose DB122, and the question and answer DB123 constitute the subject data.

[0046] In the action target output process, the action target examples input to the language model M include a plurality of pre-prepared action target examples and examples of conditions for outputting each action target example. The action target examples are stored in the storage unit 12 of the information processing apparatus 1, for example, and are read out by the processing unit 11.

[0047] In the action target output process, the prompt input to the language model M is composed of an instruction, reference information, and an output format. The instruction of the prompt shown in FIG. 11 includes an instruction to output a chart showing a plurality of indexes related to eating behavior, a chart showing a plurality of indexes related to eating preferences, a chart showing a plurality of indexes related to seasoning, a plurality of action targets, and comments related to eating behavior and eating preferences, referring to the input action target examples based on the input target person data.

[0048] The reference information of the prompt shown in FIG. 11 includes a target person ID that serves as a key for reading the record of the target person for whom the action target is output, a glucose data ID included in the attribute information of the glucose DB 122 related to the target person, and a question answer ID included in the attribute information of the question answer DB 123 related to the target person.

[0049] The output format of the prompt shown in FIG. 11 includes specifying the format of the information output by the language model M. The specification of the output format of the chart showing a plurality of indexes related to eating behavior is "A chart with axes of the total indexes of recognition related to physique and weight, eating motivation, proxy eating, hunger or satiety, eating manner, diet content, regularity of eating habits, and other indexes based on the answers to questions related to eating behavior. However, the average value of data of the same sex as the target person is superimposed and displayed". The specification of the output format of the chart showing a plurality of indexes related to eating preferences is "A chart with axes of high-fat salty, low-fat salty, high-fat sweet, and low-fat sweet based on the answers to questions related to eating preferences". The specification of the output format of the chart showing a plurality of indexes related to seasoning is "A chart with axes of sweet, salty, sour, and greasy based on the answers to questions related to seasoning".

[0050] The output format for multiple behavioral goals is specified as "goals related to changes in what is eaten to improve lifestyle habits, based on responses to questions about health status, lifestyle, or diet, meal photos, and glucose data." The output format for comments on eating behavior and food preferences is specified as "trends analyzed from responses to questions about eating behavior and food preferences, showing the tendencies of the subject's eating behavior and food preferences."

[0051] Figure 12 is an explanatory diagram showing an example of the output of the language model M related to the behavioral goal output processing. The language model M outputs a chart showing multiple indicators related to eating behavior (eating behavior chart), a chart showing multiple indicators related to food preferences (food preference chart), a chart showing multiple indicators related to seasoning (seasoning chart), multiple behavioral goals, and comments related to eating behavior and food preferences in the specified format according to the instructions and output format included in the input prompt. The processing unit 11 of the information processing device 1 acquires the chart showing multiple indicators related to eating behavior, a chart showing multiple indicators related to food preferences, a chart showing multiple indicators related to seasoning, multiple behavioral goals, and comments related to eating behavior and food preferences output by the language model M, and outputs them to the terminal device 2. Note that the average for the same sex as the subject in the eating behavior chart shown in Figure 13 is indicated by a dotted line.

[0052] Figure 13 is an explanatory diagram showing an example of a behavioral goal selection screen. The terminal processing unit 21 of the terminal device 2, which has acquired a chart showing multiple indicators related to eating behavior, a chart showing multiple indicators related to food preferences, a chart showing multiple indicators related to seasoning, multiple behavioral goals, and comments related to eating behavior and food preferences output by the language model M from the information processing device 1, displays the chart showing multiple indicators related to eating behavior, a chart showing multiple indicators related to food preferences, a chart showing multiple indicators related to seasoning, multiple behavioral goals, and comments related to eating behavior and food preferences on the display unit 25.

[0053] Each of the multiple behavioral goals is marked with a checkbox to allow the participant to select which behavioral goal they wish to attempt. The behavioral goal selection screen also displays a function button (selection complete button) for sending the selected behavioral goal to the information processing device 1. When the checkbox is checked and the selection complete button is pressed, the checked behavioral goal is sent to the information processing device 1 as the selected behavioral goal. When the processing unit 11 of the information processing device 1 obtains the selected behavioral goal, it stores the selected behavioral goal in the behavioral goal DB 124 related to the participant ID of the participant who selected the behavioral goal.

[0054] Figure 14 is an explanatory diagram showing an example of an implementation status input screen. The terminal processing unit 21 of the terminal device 2 displays an implementation status input screen on the display unit 25 for receiving input from the subject on the status of implementation of the behavioral goals. The implementation status input screen displays a date selection field and the behavioral goals selected by the subject, which have been read from the behavioral goal DB 124 by the information processing device 1.

[0055] Each action goal is accompanied by a checkbox to allow users to input that they have performed the action related to that goal. The implementation status input screen also displays a function button (implementation status transmission button) for sending the entered implementation status. If the implementation status transmission button is pressed after a checkbox is checked, the checked action goals are sent to the information processing device 1 as action goals performed on the selected date (implementation date). When the processing unit 11 of the information processing device 1 obtains the implementation status of an action goal, it stores the implementation status (whether or not it was performed) of each action goal in the field related to the selected date in the action goal DB 124 related to the subject ID of the person who entered the implementation status. Records for action goals that are checked on the implementation status input screen store a flag indicating "Performed," while records for action goals that are not checked store a flag indicating "Not performed."

[0056] Figure 15 is a flowchart illustrating an example of behavioral goal output and selection processing. When the processing unit 11 of the information processing device 1 receives an instruction to output behavioral goals from, for example, the terminal device 2, it starts the behavioral goal output and selection processing. The processing unit 11 of the information processing device 1 uses the subject ID of the subject for whom behavioral goals are to be output as a key to read the record per subject from the subject DB 121, the subject's meal photos, the subject's glucose DB 122, and the subject's question answer DB 123 from the storage unit 12 (S1). The processing unit 11 inputs the read record from the subject DB 121, meal photos, glucose DB 122, and question answer DB 123 into the language model M (S2). The processing unit 11 obtains a chart showing multiple indicators related to eating behavior (eating behavior chart), a chart showing multiple indicators related to food preferences (food preference chart), a chart showing multiple indicators related to seasoning (seasoning chart), multiple behavioral goals, and comments related to eating behavior and food preferences output by the language model M (S3). The processing unit 11 transmits to the terminal device 2 a chart showing multiple indicators related to eating behavior, a chart showing multiple indicators related to food preferences, a chart showing multiple indicators related to seasoning, multiple behavioral goals, and comments related to eating behavior and food preferences, all output by the language model M (S4).

[0057] The terminal processing unit 21 of the terminal device 2, which has acquired from the information processing device 1, receives a chart showing multiple indicators related to eating behavior, a chart showing multiple indicators related to food preferences, a chart showing multiple indicators related to seasoning, multiple behavioral goals, and comments related to eating behavior and food preferences, which has been output by the language model M. The terminal processing unit 21 of the terminal device 2, which has acquired a chart showing multiple indicators related to eating behavior, a chart showing multiple indicators related to food preferences, a chart showing multiple indicators related to seasoning, multiple behavioral goals, and comments related to eating behavior and food preferences, displays a behavioral goal selection screen on the display unit 25, which displays a chart showing multiple indicators related to eating behavior, a chart showing multiple indicators related to food preferences, a chart showing multiple indicators related to seasoning, multiple behavioral goals, and comments related to eating behavior and food preferences, which has been output by the language model M. The terminal processing unit 21 receives input for the selection of a behavioral goal (S6). The terminal processing unit 21 transmits the selected behavioral goal to the information processing device 1 (S7) and terminates the process.

[0058] The processing unit 11 of the information processing device 1, having acquired the selected behavioral target from the terminal device 2, stores the acquired behavioral target in the behavioral target field of the behavioral target DB 124, which includes the subject's subject ID (identification information) as attribute information (S8), and then terminates the process. In other words, the selected behavioral target is stored in association with the subject's identification information.

[0059] Figure 16 is an explanatory diagram showing an example of the implementation status input registration process. For example, when the processing unit 11 of the information processing device 1 receives an instruction from the terminal device 2 to register the implementation status of an action goal, it starts the implementation status registration process. The processing unit 11 of the information processing device 1 reads the action goal selected by the person from the action goal DB 124, which contains the person ID of the person whose action goal implementation status is to be registered as attribute information (S11). The processing unit 11 of the information processing device 1 transmits the read action goal to the terminal device 2 (S12).

[0060] The terminal processing unit 21 of the terminal device 2, upon receiving the action targets read from the action target DB 124 from the information processing device 1, displays the implementation status input screen on the display unit 25 (S13). The terminal processing unit 21 of the terminal device 2 accepts the selection of the date on which the implementation status will be entered and the input of whether or not each action target has been implemented (S14). The terminal processing unit 21 transmits the selected date and whether or not each action target has been implemented to the information processing device 1 (S15), and terminates the process.

[0061] The processing unit 11 of the information processing device 1, having acquired the selected date and whether or not each action objective was carried out from the terminal device 2, stores whether or not each action objective was carried out in the implementation status field for the selected date related to the record of each action objective in the action objective DB 124 (S16), and then terminates the process.

[0062] With the above configuration and processing, it is possible to output behavioral goals for improving the subject's lifestyle based on the subject's answers to questions about their diet, meal photos, and glucose data. Furthermore, by accepting the subject's selection of behavioral goals and registering the implementation status of the selected behavioral goals on a daily basis, it is possible to encourage the subject to continue taking actions to improve their lifestyle. Reference information may include, for example, participation status in workshops to improve lifestyle, or usage status of an application for inputting implementation status of behavioral goals. In this embodiment, the eating behavior chart, food preference chart, seasoning chart, multiple behavioral goals, and comments on eating behavior and food preference are output by the language model M, but are not limited to this. The eating behavior chart, food preference chart, and seasoning chart may be created using a rule-based method according to the values ​​of each item in the answers to questions about diet, for example, by classifying questions about meals into indicators on each axis of the chart, and calculating the average value of the answers to the questions classified on each axis as the numerical value of the indicator. Furthermore, multiple behavioral goals or comments related to eating behavior and food preferences may be selected from a template of behavioral goals or comments related to eating behavior and food preferences, for example, by pattern matching, according to the values ​​of each item in the answers to the questions about eating. Alternatively, multiple behavioral goals or comments related to eating behavior and food preferences may be selected from a template of behavioral goals or comments related to eating behavior and food preferences, according to a rule-based approach, according to the values ​​of each item in the answers to the questions about eating. In this case, multiple behavioral goals may be output according to rules, for example, if the value of the answer to a particular question is above a certain level, a specific behavioral goal may be output.

[0063] (Embodiment 2) In Embodiment 2, when the processing unit 11 of the information processing device 1 outputs behavioral targets using the language model M, it inputs the subject's implementation status for previously selected behavioral targets into the language model M. The differences between Embodiment 2 and Embodiment 1 will be explained below.

[0064] Figure 17 is an explanatory diagram showing an example of inputting reference information to the language model M related to the behavior goal output processing according to Embodiment 2. Figure 18 is an explanatory diagram showing an example of inputting prompts to the language model M related to the behavior goal output processing according to Embodiment 2. In the behavior goal output processing according to Embodiment 2, the language model M is input with a behavior goal DB 124 relating to the target person for whom the behavior goal is output. The behavior goal DB 124 stores the implementation status of previously selected behavior goals. That is, the processing unit 11 of the information processing device 1 outputs a new behavior goal based on the implementation status of previously selected behavior goals after a predetermined period has elapsed since the language model M output the behavior goal.

[0065] The prompt instructions input to the language model M according to Embodiment 2 include instructions to output a chart showing multiple indicators related to eating behavior, a chart showing multiple indicators related to food preferences, a chart showing multiple indicators related to seasoning, multiple behavioral goals, and comments related to eating behavior and food preferences, based on the input subject data and the subject's implementation status toward the behavioral goals, and referring to the input example of behavioral goals. Furthermore, the prompt instructions according to Embodiment 2 include instructions to change the numerical values ​​of the input behavioral goals and output them as new behavioral goals according to the implementation status toward the input behavioral goals.

[0066] The reference information of the prompt input to the language model M according to Embodiment 2 includes the behavior goal ID included in the attribute information of the behavior goal DB 124 pertaining to the subject.

[0067] Figure 19 is an explanatory diagram showing an example of the output of the language model M related to the behavior goal output processing according to Embodiment 2. The behavior goals output in the behavior goal output processing according to Embodiment 2 include behavior goals in which the numerical values ​​of the input behavior goals (behavior goals selected in the past) have been changed. In the example shown in Figure 19, the numerical value of the previously selected behavior goal "Walk for 10 minutes or more after eating" (see Figure 13) has been changed, and a new behavior goal "Walk for 15 minutes or more after eating" is output. New behavior goals are output with priority given to those with a low frequency of implementation among the input behavior goals, or behavior goals other than the input behavior goals, such as those listed in the example behavior goals.

[0068] With the above configuration and processing, the processing unit 11 of the information processing device 1 outputs new behavioral targets based on the implementation status of previously selected behavioral targets, making it possible to output behavioral targets that are more appropriate to the subject's current eating habits or behavioral habits.

[0069] (Embodiment 3) The information processing system S according to Embodiment 3 includes a medical terminal 4. The medical terminal 4 is a terminal used by medical professionals such as doctors, nurses, or nutritionists. The medical terminal 4 receives input of approval information to be output to the terminal device 2 (the subject's terminal) regarding the behavioral goals output by the language model M, and input of advice from medical professionals to the subject regarding consultations about the implementation of the behavioral goals, which are entered into the terminal device 2 via the chat function. The differences between Embodiment 3 and Embodiments 1 and 2 will be explained below.

[0070] Figure 20 is an explanatory diagram showing an overview of the information processing system S according to Embodiment 3. The information processing system S according to Embodiment 3 includes a medical terminal 4. The medical terminal 4 acquires a plurality of behavioral goals output by the language model M from the information processing device 1 and accepts the selection of a behavioral goal from among the plurality of behavioral goals that is recommended to be implemented by the subject. The medical terminal 4 transmits the selected behavioral goal (with approval information entered) to the information processing device 1. The medical terminal 4 also acquires a consultation regarding the implementation of the behavioral goal entered in the terminal device 2 from the information processing device 1, accepts the input of advice from a medical professional regarding the consultation, and transmits the input advice to the information processing device 1. The information processing device 1, having acquired the advice regarding the consultation transmitted to the information processing device 1, transmits the advice to the subject's terminal device 2.

[0071] Figure 21 is a block diagram showing an example configuration of the medical terminal 4. The medical terminal 4 is, for example, a smartphone, a tablet, or a personal computer. In this embodiment, the example of the medical terminal 4 being a smartphone will be described. The medical terminal 4 comprises a terminal processing unit 41, a storage unit 42, a communication unit 43, an input unit 44, and a display unit 45. The terminal processing unit 41 is composed of a CPU or MPU, and performs various control processing, calculation processing, etc. The medical terminal 4 may perform all or part of the processing performed by the information processing device 1 or the language model server 3.

[0072] The memory unit 42 stores an application program Pb that displays questions about meals obtained from the information processing device 1 and accepts input for advice, and displays multiple behavioral goals obtained from the information processing device 1 and accepts selection of behavioral goals. The application program Pb is provided to the medical terminal 4, for example, using a storage medium 42a. The terminal processing unit 41 of the medical terminal 4 may obtain the application program Pb using the internet and store it in the memory unit 42.

[0073] The input unit 44 receives input from healthcare professionals regarding the selection of behavioral goals to be recommended for the subject, and advice for consultations from the subject regarding the implementation of these behavioral goals. The display unit 45 displays the behavioral goals output by the language model M acquired from the information processing device 1, as well as a chat screen for receiving advice for consultations from the subject. If the medical terminal 4 is a smartphone, the input unit 44 and the display unit 45 are integrated via a touch panel.

[0074] Figure 22 is an explanatory diagram showing an example of an approval information input screen. When the terminal processing unit 41 of the medical terminal 4 acquires the behavioral target output by the language model M from the information processing device 1, it displays the subject ID related to the subject for which the behavioral target was output, and the acquired behavioral target, on the display unit 45. On the approval information input screen, each behavioral target is marked with a checkbox for accepting the input of approval information. Medical professionals input approval information by checking the checkbox for the behavioral target that they recommend the subject perform from the displayed behavioral targets. The approval information input screen also displays a function button (approval button) for sending the behavioral target for which approval information has been entered to the information processing device 1. For example, if a subject ID displayed on the approval information input screen is selected, the terminal processing unit 41 may acquire subject data related to that subject from the information processing device 1 and display it on the display unit 45.

[0075] If a checkbox is checked on the approval information input screen and the approval button is pressed, the checked behavioral objective is sent to the information processing device 1 as a behavioral objective for which approval information has been entered (selected as a behavioral objective to be recommended to the subject). When the processing unit 11 of the information processing device 1 obtains a behavioral objective for which approval information has been entered, it sends the behavioral objective for which approval information has been entered to the terminal device 2.

[0076] Figure 23 is an explanatory diagram showing an example of a chat screen. In Figure 23, the chat screens of terminal device 2 and medical terminal 4 are shown. In the chat screen of terminal device 2, the content entered in medical terminal 4 is displayed on the left, and the content entered in terminal device 2 is displayed on the right. The terminal processing unit 21 of terminal device 2 receives input from the subject regarding the behavioral goals being implemented and consultations regarding the implementation of the behavioral goals, and transmits the entered content to medical terminal 4 via the information processing unit 1.

[0077] In the chat screen of medical terminal 4, the content entered in terminal device 2 is displayed on the left, and the content entered in medical terminal 4 is displayed on the right. The terminal processing unit 41 of medical terminal 4 receives input from medical professionals regarding advice for consultations sent from terminal device 2, and transmits the input content to terminal device 2 via information processing unit 1. The content input and output in the chat function between terminal device 2 and medical terminal 4 is stored in the storage unit 12 of information processing unit 1.

[0078] Figure 24 is a flowchart showing an example of the behavior target output and selection process according to Embodiment 3. When the processing unit 11 of the information processing device 1 receives an instruction to output a behavior target from the terminal device 2, for example, it starts the behavior target output and selection process. S21 to S23 are the same processes as S1 to S3 shown in Figure 15. The processing unit 11 of the information processing device 1 transmits the multiple behavior targets output by the language model M to the medical terminal 4 (S24).

[0079] The terminal processing unit 41 of the medical terminal 4, which has acquired multiple behavioral goals output by the language model M from the information processing device 1, displays an approval information input screen on the display unit 45 (S25). The terminal processing unit 41 accepts the input of approval information for the behavioral goals (S26). The terminal processing unit 41 transmits the behavioral goals with the input approval information to the information processing device 1 (S27).

[0080] The processing unit 11 of the information processing device 1, which has acquired the behavioral target with approval information from the medical terminal 4, transmits to the terminal device 2 a chart showing multiple indicators related to eating behavior, a chart showing multiple indicators related to food preferences, a chart showing multiple indicators related to seasoning, comments related to eating behavior and food preferences, and the behavioral target with approval information (S28). S29 to S32 are the same processes as S5 to S8 shown in Figure 15.

[0081] With the above configuration and processing, the healthcare professional selects from among the multiple behavioral goals output by the language model M that are most suitable for improving the subject's lifestyle habits, and transmits them to the subject's terminal device 2. Furthermore, communication between the subject and the healthcare professional is realized in a simple manner. This makes it possible to encourage the subject to continue taking actions that improve their lifestyle habits.

[0082] (Embodiment 4) The processing unit 11 of the information processing device 1 according to Embodiment 4 outputs advice to the subject by inputting the consultation regarding the implementation of the target behavior entered by the subject in the chat function into the language model M. The following describes the differences between Embodiment 4 and Embodiments 1 to 3.

[0083] Figure 25 is an explanatory diagram showing an example of input to the language model M related to the advice output processing. In the advice output processing, the language model M is input with a meal image, a record from the subject DB 121, glucose DB 122, question answer DB 123, behavioral goal DB 124, and a prompt. The meal image, the record from the subject DB 121, glucose DB 122, question answer DB 123, and behavioral goal DB 124 input in the advice output processing are the same as those input to the language model M in the behavioral goal output processing (see Figures 10 and 17).

[0084] In the advice output process, the prompt input to the language model M consists of instructions, reference information, and consultation content. The prompt instructions shown in Figure 25 include instructions to output advice regarding the consultation content concerning the implementation of the behavioral goals of the subject, based on the input subject data and the implementation status of the behavioral goals recorded in the behavioral goal DB 124.

[0085] The prompt reference information shown in Figure 25 includes a subject ID that serves as a key for reading the subject's record for outputting behavioral goals, a glucose data ID included in the attribute information of the glucose DB 122 relating to the subject, a question answer ID included in the attribute information of the question answer DB 123 relating to the subject, and a behavioral goal ID included in the attribute information of the behavioral goal DB 124 relating to the subject.

[0086] The prompt consultation content shown in Figure 25 includes the behavioral goals that the subject is implementing and the consultation from the subject regarding the implementation of the behavioral goals, as entered in the chat screen according to Embodiment 4 (see Figure 27). In the example shown in Figure 25, the consultation is entered as, "I really need to use oil, what should I do?"

[0087] Figure 26 is an explanatory diagram showing an example of the output of the language model M involved in the advice output processing. The language model M outputs advice regarding the subject's consultation content concerning the implementation of behavioral goals, according to the instructions contained in the input prompt. In the example shown in Figure 26, the advice output is, "Using plenty of olive oil when grilling will make your food juicy and good for your health."

[0088] Figure 27 is an explanatory diagram showing an example of a chat screen according to Embodiment 4. The chat screen according to Embodiment 4 is displayed on the display unit 25 of the terminal device 2. In the chat screen according to Embodiment 4, the content output by the language model M is displayed on the left, and the content entered in the terminal device 2 is displayed on the right. The terminal processing unit 21 of the terminal device 2 receives input from the subject regarding the behavioral goals being implemented and consultations regarding the implementation of the behavioral goals, and transmits the input content to the medical terminal 4 via the information processing unit 1. The terminal processing unit 21 of the terminal device 2 also obtains advice output by the language model M in response to the input consultation from the information processing unit 1 and displays it on the chat screen.

[0089] With the above configuration and processing, the processing unit 11 of the information processing device 1 can easily output advice in response to consultations from the subject regarding the implementation of behavioral goals by using the language model M.

[0090] The embodiments disclosed herein should be considered in all respects as illustrative and not restrictive. The technical features described in each embodiment can be combined with one another, and the scope of the present invention is intended to include all modifications within the claims and scopes equivalent to the claims. Furthermore, the independent and dependent claims described in the claims can be combined with each other in any combination, regardless of the form of reference. Moreover, the claims use a multi-claim format in which claims refer to two or more other claims (multi-claim format), but are not limited thereto. They may also be described using a multi-claim format in which at least one multi-claim refers to another multi-claim (multi-multi-claim format).

[0091] 1: Information processing device 11: Processing unit 12: Storage unit 12a: Storage medium 13: Communication unit 2: Terminal device 21: Terminal processing unit 22: Storage unit 22a: Storage medium 23: Communication unit 24: Input unit 25: Display unit 26: Imaging unit 3: Language model server M: Language model 4: Medical terminal 41: Terminal processing unit 42: Storage unit 43: Communication unit 44: Input unit 44a: Storage medium 45: Display unit 121: Target DB 122: Glucose DB 123: Question and answer DB 124: Behavioral goal DB N: Network P: Program Pa: Application program Pb: Application program

Claims

1. A program that causes a computer to perform the following processes: acquire subject data including answers to questions about the subject's diet, food photos, and glucose data; output multiple behavioral goals related to lifestyle improvement based on the subject data; and store the behavioral goal selected by the subject from the multiple behavioral goals in association with the subject's identification information.

2. The program according to claim 1, wherein the questions relating to meals include questions relating to eating behavior and questions relating to food preferences.

3. The program according to claim 1 or 2, which outputs a chart showing multiple indicators related to eating behavior, a chart showing multiple indicators related to food preferences, and comments related to eating behavior and food preferences, in conjunction with the multiple behavioral goals.

4. The program according to claim 1 or 2, which outputs comments on eating behavior and food preferences by providing a language model with answers to the questions on eating, including questions on eating behavior and questions on food preferences.

5. A program according to claim 1 or 2 that outputs a plurality of behavioral goals by providing a language model with answers to the questions relating to meals, including questions relating to eating behavior and questions relating to food preferences, the meal photographs, and the glucose data, along with a plurality of pre-prepared examples of behavioral goals.

6. The program according to claim 1 or 2, which displays input fields for the status of implementation of a selected action goal and fields for selecting the date of implementation of the action goal, and accepts input of the status of implementation and selection of the date of implementation.

7. The program according to claim 1 or 2, which outputs a new behavioral goal after a predetermined period has elapsed since outputting the behavioral goal, based on the answers to the questions about meals, including questions about eating behavior and questions about food preferences, meal photos, glucose data, and the status of implementation of the selected behavioral goal.

8. The program according to claim 7, which changes the numerical value of an action target according to the progress toward the selected action target and outputs a new action target.

9. The program according to claim 1 or 2, which outputs the multiple behavioral goals to the terminal of a healthcare professional, and when approval information from the healthcare professional is entered, outputs the multiple behavioral goals to the terminal of the subject.

10. The program according to claim 1 or 2, which uses a chat function to perform input and output of consultations from the subject regarding the implementation of behavioral goals and advice from healthcare professionals to the consultations.

11. The program according to claim 1 or 2, which outputs advice regarding the implementation of behavioral goals from the subject, input via the chat function, the subject data, the behavioral goals, and the status of implementation of the behavioral goals to a language model, and displays the outputted advice via the chat function.

12. An information processing method that acquires subject data including the subject's answers to questions about their diet, meal photos, and glucose data; outputs multiple behavioral goals related to lifestyle improvement based on the subject data; and stores the behavioral goal selected by the subject from the multiple behavioral goals in association with the subject's identification information.

13. An information processing device comprising a processing unit that acquires subject data including the subject's answers to questions about their diet, meal photos, and glucose data; outputs multiple behavioral goals related to lifestyle improvement based on the subject data; and stores the behavioral goal selected by the subject from the multiple behavioral goals in association with the subject's identification information.

Images