Program, information processing method, and information processing apparatus

Abstract

Provided is a program or the like that can provide knowledge related to menopausal symptoms. A computer obtains symptom information related to menopausal symptoms of a user in accordance with a program. The computer inputs the obtained symptom information into a learning model that has been learned to output an advice related to the symptoms when symptom information related to menopausal symptoms is input, and obtains an advice corresponding to the input symptom information. Then, the computer outputs the obtained advice.

Description

Technical Field

[0001] This disclosure relates to procedures, information processing methods, and information processing apparatus. Background Technology

[0002] Patent Document 1 discloses a system capable of implementing psychotherapy, such as behavioral therapy, cognitive therapy, and guidance, for obese patients in medical institutions when necessary. By using the system disclosed in Patent Document 1, it is possible to correct patients' misconceptions about obesity and obesity treatment when necessary, and to equip them with accurate knowledge. Thus, for example, when an urge to eat arises, it is possible to control the urge based on correct knowledge, thereby effectively eliminating the urge.

[0003] Existing technical documents

[0004] Patent documents

[0005] Patent Document 1: Japanese Patent No. 6862009 Summary of the Invention

[0006] The problem the invention aims to solve

[0007] It is claimed that the average age of menopause for women is around 50, and the ten years encompassing the five years before and after menopause are generally referred to as menopause. It is known that various symptoms occur during menopause due to the complex involvement of multiple factors, including physical, psychological, and social factors. The symptoms experienced by each individual during menopause vary greatly, making it difficult for them to determine whether they should seek medical attention, and many also lack anyone to consult. The system in Patent Document 1, designed to provide accurate information to obese patients, fails to provide information related to menopausal symptoms.

[0008] The purpose of this disclosure is to provide a program, etc., that can provide knowledge related to menopausal symptoms.

[0009] Solution for solving the problem

[0010] One scheme of this disclosure involves a program that causes a computer to perform the following processes: acquiring symptom information related to the symptoms of a user experiencing menopause; inputting the acquired symptom information into a learning model to acquire suggestions corresponding to the input symptom information, wherein the learning model is learned to output suggestions related to the symptoms when symptom information related to menopause symptoms is input; and outputting the acquired suggestions.

[0011] Invention Effects

[0012] One of the solutions disclosed herein can provide knowledge related to menopausal symptoms. Attached Figure Description

[0013] Figure 1 This is a schematic diagram illustrating the structure of an information processing system.

[0014] Figure 2 This is a block diagram representing a structural example of a server and a user terminal.

[0015] Figure 3A This is a diagram illustrating an example of a suggested record layout for a database.

[0016] Figure 3B This is a schematic diagram illustrating an example of the record layout of a user's database.

[0017] Figure 4A This is a schematic diagram representing a structural example of a learning model.

[0018] Figure 4B This is a schematic diagram representing a structural example of a learning model.

[0019] Figure 5 This is a flowchart illustrating an example of a suggested processing procedure.

[0020] Figure 6 This is a flowchart illustrating an example of a suggested processing procedure.

[0021] Figure 7 This is a flowchart illustrating an example of a suggested processing procedure.

[0022] Figure 8A This is a schematic diagram representing an example of a screen display.

[0023] Figure 8B This is a schematic diagram representing an example of a screen display.

[0024] Figure 9A This is a schematic diagram representing an example of a screen.

[0025] Figure 9B This is a schematic diagram representing an example of a screen display.

[0026] Figure 10A This is a schematic diagram representing an example of a screen display.

[0027] Figure 10B This is a schematic diagram representing an example of a screen display.

[0028] Figure 11A This is a schematic diagram representing an example of a screen display.

[0029] Figure 11B This is a schematic diagram representing an example of a screen display.

[0030] Figure 12A This is a schematic diagram representing a structural example of an inspection and judgment model.

[0031] Figure 12BThis is a schematic diagram illustrating a structural example of the learning model in Implementation Method 2.

[0032] Figure 13 This is a flowchart illustrating an example of the proposed processing flow for Implementation Method 2.

[0033] Figure 14 This is a schematic diagram representing an example of a screen display.

[0034] Figure 15 This is a schematic diagram illustrating a variation of the symptom input screen.

[0035] Figure 16 This is a schematic diagram representing a variation of a historical scene.

[0036] Figure 17A This is a schematic diagram representing an example.

[0037] Figure 17B This is a schematic diagram representing an example. Detailed Implementation

[0038] The procedures, information processing methods, and information processing apparatus of this disclosure will now be described in detail with reference to the accompanying drawings illustrating their embodiments.

[0039] Implementation Method 1

[0040] The information processing system provides tips corresponding to the symptoms of menopause. Menopause generally refers to the ten-year period of five years before and five years after menopause. The average age of menopause for women is said to be around 50, so the period between 45 and 55 years old is called menopause. Figure 1 This is a schematic diagram illustrating an example of the structure of an information processing system. The information processing system in this embodiment includes a server 10 and multiple user terminals 20, which are connected via a network N such as the Internet. Multiple user terminals 20 are information terminals used by users who utilize services provided by the server 10.

[0041] Figure 2This is a block diagram illustrating a structural example of server 10 and user terminal 20. Server 10 is an information processing device capable of various information processing and information transmission and reception, and is composed of a server computer or personal computer, etc. Server 10 includes a control unit 11, a storage unit 12, a communication unit 13, an input unit 14, a display unit 15, a reading unit 16, etc., and these parts are interconnected via a bus. The control unit 11 is composed of one or more processors such as CPU (Central Processing Unit), MPU (Micro-Processing Unit), GPU (Graphics Processing Unit), TPU (Tensor Processing Unit), or AI chip (AI semiconductor). The control unit 11 executes the information processing and control processing that the server 10 should perform by appropriately executing the program 12P stored in the storage unit 12.

[0042] Storage unit 12 includes RAM (Random Access Memory), flash memory, hard disk, SSD (Solid State Drive), etc. Storage unit 12 stores the program 12P (program product, computer program) executed by control unit 11 and various data. Additionally, storage unit 12 temporarily stores data generated when control unit 11 executes program 12P. Furthermore, storage unit 12 stores, for example, a learning model 12M learned from training data through machine learning. Learning model 12M is a model that takes symptom information related to known menopausal symptoms as input and outputs information related to recommendations that should be provided to the user based on the input symptom information. Learning model 12M is envisioned as a program module constituting artificial intelligence software. Learning model 12M performs prescribed operations on input values ​​and outputs the operation results. Storage unit 12 stores data such as coefficients and thresholds of the function used to define this operation as learning model 12M. In addition, storage unit 12 also stores a suggestion DB12a and a user DB12b. The storage unit 12 may be composed of multiple storage devices. A portion of the storage unit 12 may be other storage devices connected to the server 10, or other storage devices that can communicate with the server 10.

[0043] The communication unit 13 is a communication module used for processing related to wired or wireless communication, and it sends and receives information with other devices via network N. Network N can be the Internet or a public switched telephone network, or it can be a LAN (Local Area Network) built within a facility equipped with server 10. The input unit 14 accepts user input and sends control signals corresponding to the operation to the control unit 11. The display unit 15 is a liquid crystal display or an organic EL display, etc., and displays various information according to instructions from the control unit 11. The input unit 14 and the display unit 15 can be a single touch panel.

[0044] The reading unit 16 reads information stored in portable storage media 10a such as CD (Compact Disc), DVD (Digital Versatile Disc), USB (Universal Serial Bus) storage, SD (Secure Digital) card, and CompactFlash (registered trademark). The program 12P and various data stored in the storage unit 12 can be read from the portable storage media 10a by the control unit 11 via the reading unit 16 and then stored in the storage unit 12. Furthermore, the program 12P and various data can be written to the storage unit 12 during the manufacturing stage of the server 10, or downloaded from other devices by the control unit 11 via the communication unit 13 and stored in the storage unit 12.

[0045] In this embodiment, server 10 can be a multi-computer system composed of multiple computers, or a virtual machine virtualized within a single device using software. Furthermore, when server 10 is composed of server computers, it can be a local server located within a facility where server 10 is located, or a cloud server connected via network N. The following description uses server 10 as an example of a single computer. Additionally, program 12P can be configured on a single computer or at a single site, or it can be distributed across multiple computers at multiple sites and interconnected via network N. Moreover, server 10 does not necessarily require input unit 14 and display unit 15; it can be a structure that accepts operations via connected computers, or a structure that outputs information to be displayed to an external display device.

[0046] User terminal 20 is a general-purpose information processing device such as a personal computer, smartphone, or tablet computer. User terminal 20 has a control unit 21, a storage unit 22, a communication unit 23, an input unit 24, a display unit 25, and a reading unit 26, all of which are connected via a bus. The control unit 21, storage unit 22, communication unit 23, input unit 24, display unit 25, and reading unit 26 of user terminal 20 have the same structure as the control unit 11, storage unit 12, communication unit 13, input unit 14, display unit 15, and reading unit 26 of server 10; therefore, detailed descriptions of the structure will not be repeated. Furthermore, in addition to storing the program 22P (program product) executed by control unit 21, the storage unit 22 of user terminal 20 also stores an application program (hereinafter referred to as health management application 22AP) for implementing the use of services provided by server 10 to receive suggestions corresponding to the user's symptoms.

[0047] In the information processing system described above, user terminal 20 receives input of symptom information related to the user's symptoms and sends the input symptom information to server 10. Server 10 acquires the symptom information sent by user terminal 20, determines suggestions corresponding to the acquired symptom information, and sends them to user terminal 20. Thus, suggestions corresponding to the user's symptoms are sent from server 10 to user terminal 20 and provided to the user of user terminal 20. In this embodiment, server 10 uses learning model 12M when determining suggestions based on the user's symptom information.

[0048] Figure 3A This is a schematic diagram illustrating an example of the recommended record layout for DB12a. Figure 3B This is a schematic diagram illustrating an example of the record layout for user DB12b. Recommendation DB12a is a database used to store information related to recommendations prepared for the various symptoms experienced by menopausal users. Figure 3A The suggested DB12a includes symptom information columns and suggestion information columns. It stores suggestion information related to the suggestions to be provided to users with each symptom, linked to symptom information associated with each symptom. Symptom information includes the category into which each symptom is classified and the items categorized within each category. Suggestion information includes the suggestion ID assigned to each suggestion, the type of suggestion, and the content of the suggestion. Suggestion content includes, for example, information on diet, exercise, sleep, stress-reducing behaviors, the use of electronic devices such as smartphones or computers, the use of medications including traditional Chinese medicine, information related to body care, information related to skin care, information related to lifestyle (habits), and information related to internal health (internal care). Suggested DB12a is not limited to... Figure 3AThe structure shown. Furthermore, the types of categories and items stored in the recommended DB12a, as well as the recommended types, are not limited to... Figure 3A The example shown.

[0049] User DB12b is a database used to store information about users who have registered to use the services provided by server 10. Figure 3B The user DB12b shown includes columns such as User ID, Name, Age, Medical History, and Input Information, storing user-related information in association with a unique identifier (User ID) assigned to each user. The Name, Age, and Medical History columns store the user's name, age, and medical history. The user's name, age, and medical history are entered, for example, through the user terminal 20 during user registration, and are retrieved from the user terminal 20 and stored in the user DB12b by the server 10. The Input Information column stores information related to user symptoms entered by the user through the user terminal 20. Input information includes the date, symptom information related to the user's symptoms, and practice information related to behaviors practiced by the user in their daily life. Symptom information includes... Figure 3A The suggestion is to include any of the symptoms registered in DB12a. When the severity (severity) of each symptom is entered, information indicating the severity of each symptom will also be included. Information indicating severity includes terms such as mild, moderate, severe, or most severe. Practical information includes, for example, the intake of nutritional supplements such as vitamins or iron, the implementation of exercises or massages such as yoga or pelvic floor muscle training, and the intake of medications for treating menopausal symptoms. The content stored in the user's DB12b is not limited to... Figure 3B The examples shown could be structures without a medical history column, or structures that store treatment history or medication history in addition to medical history.

[0050] Figure 4A This is a schematic diagram illustrating the structure of learning model 12M. Learning model 12M learns by taking symptom information related to a predetermined number of symptoms experienced by a menopausal user as input, and based on the input symptom information, performing calculations to infer suggestions that should be provided to the user for each symptom, and outputting the results. Figure 4A The example shown is configured to take five symptom information entries as input (symptom information from the first to the fifth symptom information) and output information related to suggestions for each pre-prepared symptom. The learning model 12M can be constructed using algorithms such as CNN (Convolutional Neural Network), decision tree, random forest, SVM (Support Vector Machine), and Transformer, or a combination of multiple algorithms.

[0051] Figure 4A The learning model 12M shown has five input nodes, through which symptom information (identification information that identifies each symptom) representing the symptoms selected by the user is input. Furthermore, in this embodiment, the user is configured to select up to five symptoms, and the learning model 12M inputs up to five symptom information. The learning model 12M uses various functions and thresholds to calculate the output value based on the input symptom information and outputs the calculated output value. Figure 4A The learning model 12M shown has multiple output nodes, each associated with a pre-prepared symptom, specifically with a symptom registered in the suggestion DB12a. Each output node outputs information related to suggestions (specifically, suggestion IDs) that should be provided to the user for the associated symptom. According to this structure, when a predetermined number (e.g., five) of symptom information are input into the learning model 12M, the learning model 12M outputs suggestion IDs for suggestions that should be provided to the user for each symptom.

[0052] The learning model 12M is generated through machine learning using training data. This training data associates symptom information used for learning with suggestion IDs (correct answer labels) representing suggestions to be given to users by doctors or other professionals based on that symptom information. The training data is generated by assigning suggestion IDs to patients receiving treatment for menopausal symptoms or those aware of their menopausal symptoms, based on symptom information for each patient. Furthermore, the suggestions for each symptom are appropriately set based on the severity of the symptom and the combination of co-occurring symptoms. The learning model 12M is learned to output information represented by the correct answer label (e.g., suggestion ID) from the output nodes corresponding to each symptom, given the symptom information contained in the training data. In the learning process, the learning model 12M performs calculations based on the input symptom information and computes the output values ​​from each output node. Then, the learning model 12M compares the calculated output values ​​of each output node with the information corresponding to the correct answer label (for output nodes corresponding to symptoms with a correct answer label, it is the suggestion ID of the correct answer label; for other output nodes, it is, for example, 0), and optimizes the parameters used for computation to make the two approximate. For example, it optimizes parameters such as the weights (association coefficients) between neurons in the learning model 12M using error backpropagation, steepest descent, etc. This yields the learning model M12, which, given a specified number of symptom information inputs, outputs the suggestion ID of the suggestion to be provided to the user from the output nodes associated with the symptoms of the input symptom information, and outputs, for example, 0 from other output nodes. Furthermore, the server 10 determines the output value corresponding to each symptom contained in the input data based on the output values ​​of each output node from the learning model 12M, and uses the determined output value as the suggestion ID of the suggestion corresponding to each symptom, thereby determining the suggestion ID of the suggestion to be provided to the user for each symptom for which symptom information has been input.

[0053] Figure 4B This is a schematic diagram illustrating the structure of a variant of learning model 12M, namely learning model 12Ma. Learning model 12Ma has the same structure as learning model 12M, but the symptom information used as input data includes, in addition to information related to a specified number of symptoms, information indicating the severity (degree) of each symptom. The severity information can be, for example, any one of mild, moderate, severe, or most severe. Learning model 12Ma is learned to perform calculations based on the input symptom information (information on each symptom and the severity information of each symptom), inferring suggestions that should be provided to the user for each symptom, and outputting the results of the calculations.

[0054] The learning model 12Ma is generated through machine learning using training data. This training data associates symptom information, including a specified number of symptoms and the severity of each symptom, with suggestion IDs (correct answer labels) for suggestions to be provided to the user for each symptom. The learning model 12Ma also learns that, given the symptom information contained in the training data, it outputs the information represented by the correct answer label from the output node corresponding to each symptom (for output nodes corresponding to symptoms with a set correct answer label, it is the suggestion ID of the correct answer label; for other output nodes, it is, for example, 0). This results in the learning model 12Ma, which, given symptom information containing a specified number of symptoms and the severity of each symptom, infers suggestions to be provided to the user for each symptom and outputs the inference results.

[0055] The learning of models 12M and 12Ma can be performed on server 10 or on other learning devices. The learned models 12M and 12Ma generated from learning on other learning devices are downloaded from the learning device to server 10 via network N or via portable storage medium 10a and stored in storage unit 12.

[0056] Learning models 12M and 12Ma are not limited to Figure 4A and Figure 4B The structure shown is as follows. For example, instead of outputting suggestion IDs (the suggestion IDs with the highest confidence) from each output node, which are inferred to be suggestions to be provided for the symptoms associated with each output node, a structure with multiple output nodes could be used, where each output node outputs the confidence level of each suggestion ID prepared for each symptom. Furthermore, the learning models 12M and 12Ma could, for example, be structures that input the user's age or the number of years since menopause in addition to the user's symptom information; alternatively, they could be structures that input one or more of the user's medical history, treatment history, medication history, etc. In this case, the learning models 12M and 12Ma infer suggestions to be provided to the user not only based on symptom information but also based on various user information. Additionally, the learning models 12M and 12Ma can be prepared for each attribute distinguished by age or the number of years since menopause.

[0057] Server 10 performs machine learning on the prescribed training data to generate learning models 12M and 12Ma in advance. Then, server 10 inputs the user's symptom information obtained from user terminal 20 into learning models 12M and 12Ma, and determines the suggestions to be provided to the user based on the output information from learning models 12M and 12Ma.

[0058] The following process will be described: In the information processing system of this embodiment, the user terminal 20 receives input of symptom information related to the user's symptoms, determines suggestions corresponding to the input symptom information in the server 10, and provides them to the user. Figures 5 to 7 This is a flowchart illustrating an example of a suggested processing flow. Figures 8A to 11B This is a schematic diagram illustrating an example of a screen display. In Figures 5 to 7 In the diagram, the left side represents the processing performed by the user terminal 20, and the right side represents the processing performed by the server 10.

[0059] When a user wants to obtain advice related to menopausal symptoms, the user terminal 20 launches the health management application 22AP. The control unit 21 of the user terminal 20 performs the following processing according to the health management application 22AP. When the health management application 22AP is launched, the control unit 21 determines whether the input of symptom information related to the user's symptoms is the first time (whether it is the first input) (S11). For example, the control unit 21 periodically, for example once a day, accepts the user's symptom information input through the input unit 24 and stores the input symptom information along with the date and time of input in the health management application 22AP or the storage unit 22. Therefore, if the symptom information is not stored in the health management application 22AP or the storage unit 22, the control unit 21 determines it as the first input; if the symptom information has already been stored, it determines it as a second or subsequent input.

[0060] When it is determined to be the first input (S11: Yes), the control unit 21 will, for example, perform the following: Figure 8A The initial input screen shown is displayed on display unit 25 (S12). Furthermore, during the initial input, control unit 21 can first display an input screen (not shown) containing the user's name, age, medical history, treatment history, medication history, and other personal information, accept the personal information through the input screen, and then display the initial input screen. Upon accepting the personal information, control unit 21 sends the accepted personal information to server 10 for user registration. Upon receiving personal information from user terminal 20, control unit 11 of server 10 issues a user ID and stores (registers) the received personal information in user DB12b of storage unit 12 in association with the issued user ID. Alternatively, control unit 21 of user terminal 20 can also store the accepted personal information in health management application 22AP or in storage unit 22.

[0061] Figure 8AThe initial input screen shown features a selection button group that allows users to choose up to five symptoms of interest. The selection button group includes icons (selection buttons) representing various menopausal symptoms, each categorized into four types: physical, sensitive areas, psychological, and sleep status (sleep state). The categories and types of symptoms displayed on the initial input screen are not limited to... Figure 8A For example, the control unit 21 accepts a selection of up to five symptoms via the initial input screen (S13). After accepting the symptom selection, the control unit 21 displays the icon of the selected symptom in a different way than the other icons. Figure 8A In this implementation, four symptoms were selected: hot flashes, headache, anxiety, and night sweats. The icons for each symptom were displayed in an inverted state. The initial input screen of this embodiment is configured to accept the selection of symptom types, but it can also be configured to accept the severity of the selected symptom in addition to the type. Specifically, on the initial input screen, when any symptom is selected, the control unit 21 displays as follows: Figure 9B The severity level input field R1 is shown. Through the severity level input field R1, you can input any of the following for the selected symptoms: mild, moderate, severe, or most severe.

[0062] A "Next" button is provided on the initial input screen, which is used to request suggestions corresponding to the selected symptom. Control unit 21 determines whether the "Next" button has been activated on the initial input screen (S14). If it is determined that it has not been activated (S14: No), it returns to step S13 and continues with symptom selection acceptance. If it is determined that the "Next" button has been activated (S14: Yes), control unit 21 stores the symptom information representing the selected symptom in association with the date (or date and time) at that moment (S15). For example, control unit 21 stores the symptom information in the health management application 22AP or in storage unit 22. Then, control unit 21 sends the symptom information to server 10 (S16) and requests suggestions corresponding to the symptom information from server 10. Furthermore, control unit 21 sends the symptom information along with the user ID assigned to the user and the date (or date and time) at that moment to server 10. In addition, when the control unit 21 receives input of the type of symptom and the severity of the symptom through the initial input screen, it stores the symptom information, including the type of symptom and the severity of the symptom, in the health management application 22AP or the storage unit 22, and sends it to the server 10.

[0063] When the control unit 11 of server 10 receives symptom information from user terminal 20 along with user ID and date, it stores the received symptom information in user DB12b in association with the received user ID and date (S17). Based on the symptom information obtained from user terminal 20, control unit 11 determines suggestions corresponding to each symptom selected by the user (S18). Specifically, control unit 11 inputs the symptom information (e.g., identification information of each symptom) obtained from user terminal 20 into learning model 12M, and obtains suggestion information corresponding to each symptom from each output node of learning model 12M. Specifically, control unit 11 obtains suggestion IDs from output nodes associated with the symptoms in the input symptom information, and obtains, for example, 0 from other output nodes. Furthermore, if the user selects fewer than five symptoms, control unit 11 inputs information (e.g., identification information) of each symptom from the input node for the number of selected symptoms, obtains suggestion IDs corresponding to each symptom from output nodes associated with the selected symptoms, and obtains, for example, 0 from other output nodes. Furthermore, when the symptom information includes the severity of each symptom, the control unit 11 inputs symptom information indicating the type and severity of each symptom. Figure 4B The learning model 12Ma is shown to have input nodes, and suggestions corresponding to each symptom are obtained from each output node. Here, the control unit 11 also obtains suggestions from output nodes associated with the symptoms of the input symptom information, and obtains suggestions such as 0 from other output nodes.

[0064] After determining the recommendations corresponding to each symptom, the control unit 11 generates, for example: Figure 8B The suggested screen shown is shown in S19. In this process, the control unit 11 reads the suggestion corresponding to each suggestion ID obtained in step S18 from the suggestion DB12a, and generates a suggested screen to display the read suggestions. Thus, a suggested screen is generated displaying suggestions related to each symptom selected by the user through the initial input screen. The control unit 11 sends the generated suggested screen to the user terminal 20 (S20), and the control unit 21 of the user terminal 20 receives the suggested screen from the server 10 and displays it on the display unit 25 (S21).

[0065] On the other hand, when the control unit 21 determines that it is not the first input (S11: No), it will proceed as follows: Figure 9A The top-level screen shown is displayed on display unit 25 (S22). The top-level screen displays the date of the current moment and has a fill button for instructing the user to process the symptoms entered at that moment. In addition, at the bottom of the top-level screen are: a suggestion button for instructing the user to view suggestions corresponding to the symptom information already entered by the user; a history button for instructing the user to view the symptom information entered in the past (input history); and a record button for instructing the user to register (record) new symptom information.

[0066] Control unit 21 determines whether the fill button on the top-level screen has been operated (S23). If it determines that the button has not been operated (S23: No), for example, it determines whether the suggestion button has been operated (S24). If it determines that the suggestion button has not been operated (S24: No), control unit 21 returns to step S23 and waits until the fill button or suggestion button is operated. Furthermore, when the history button on the top-level screen is operated, control unit 21 displays the history screen (reference) based on previously entered input information (symptom information and practice information). Figure 11B The input information is displayed on the display unit 25. Past input information can be saved in the health management application 22AP or the storage unit 22, or the control unit 21 can retrieve input information stored in the user DB12b on the server 10. Furthermore, when the record button on the top-level screen is pressed, the control unit 21 performs the same processing as when the fill button is pressed.

[0067] When it is determined that the fill button has been operated (S23: Yes), the control unit 21 will... Figure 9B The symptom input screen shown is displayed on display unit 25 (S25). Figure 9B The symptom input screen shown has the same characteristics as... Figure 8A The same selection button group is shown on the initial input screen. Furthermore, Figure 9B The selection button group on the screen has icons representing each symptom in five categories: body, sensitive areas, psychological state, sleep status, and bleeding. In addition, Figure 9B The selection button group on the screen has icons for each category to indicate asymptomatic, and is configured to allow users to input no symptoms for any category. Figure 9B The screen is set up with indicators for display. Figure 9B The symptom input screen shown includes symptom buttons and indicators. Figure 10A The practice button on the practice input screen, when activated, displays... Figure 10A The practice input screen shown is for accepting input related to the user's current practice. It includes a selection button group with icons representing various items the user might be performing. These selection buttons represent various items such as nutrients the user is ingesting, exercise the user is performing, medications the user is taking, and treatments the user is receiving. Each icon is categorized into four types: nutrition, lifestyle, therapy (medications), and therapy (other than medications). The items and categories displayed on the symptom input screen and the practice input screen are not limited to these categories. Figure 9B and Figure 10A Examples.

[0068] Control unit 21 judges that Figure 9BThe symptom input screen shows whether to accept the selection of any symptom (S26). When it is determined that a symptom has been selected (S26: Yes), as shown... Figure 9B The severity input field R1 is displayed above the symptom input screen (S27). Then, the control unit 21 accepts input of the severity level for the selected symptom via the severity input field R1 (S28). Upon accepting the severity level input, the control unit 21 displays, for example, the icon of the selected symptom inverted display. Furthermore, the control unit 21 can display the input severity level in association with the icon, allowing the user to confirm the severity level of each symptom they have entered.

[0069] exist Figure 9B Symptom input screen and Figure 10A The practice input screen includes a save button to indicate whether the entered symptom information and practice information are saved. Control unit 21 determines whether the save button on the symptom input screen has been activated (S29). If it determines that the button has not been activated (S29: No), it returns to step S26 and continues to accept input of the type and severity of symptoms. If it determines that the save button has been activated (S29: Yes), control unit 21 sends the symptom information indicating the type and severity of symptoms entered through the symptom input screen to server 10 (S30) and requests server 10 to save the symptom information. Here, control unit 21 also sends the symptom information along with the user ID and the date (or date and time) to server 10. Control unit 11 of server 10 stores the symptom information in association with the received user ID and date in user DB12b (S31). Alternatively, control unit 21 may also store the symptom information received through the symptom input screen in association with the date (or date and time) in health management application 22AP or storage unit 22.

[0070] When the control unit 21 determines that it is in Figure 9B If no symptom selection is accepted on the symptom input screen (S26: No), it is determined whether the practice button has been operated (S32). If it is determined that the practice button has not been operated (S32: No), the control unit 21 returns to step S26 to accept the symptom selection, or waits until the practice button is operated. If it is determined that the practice button has been operated (S32: Yes), the control unit 21 will proceed as follows: Figure 10A The practice input screen shown is displayed on the display unit 25 (S33), and practice information is received through the practice input screen (S34). Figure 10AThe practice input screen shown is categorized by the nutrients (nutrients) the user obtains through supplements, etc., and includes icons for B vitamins, vitamin C, vitamin E, magnesium, isoflavones / estradiol, iron, and multivitamin preparations, as well as an add button. In addition to the aforementioned nutrients, items related to nutrients and dietary lifestyles are also provided as nutrient categories, such as vitamin A, vitamin D, zinc, tryptophan, omega-3 fatty acids (DHA / EPA), probiotics, protein / amino acids, dietary fiber, healthy fats, low sugar, low fat, and low salt, and are configured to allow users to change the display / display of each item's icon. For example, when the add button is activated, the control unit 21 displays a selection screen (not shown) for indicating whether to display or not display the aforementioned items. This selection screen allows users to choose whether to display or not display each item. The control unit 21 displays the icon of the selected item in the nutrient category display bar of the practice input screen. The icons for each item are prepared by the health management application 22AP.

[0071] Regarding the lifestyle category, icons are provided to represent yoga, Pilates, pelvic floor muscle training, foot bath, and foot massage, as well as an add button. In addition to the above items, the lifestyle category also includes items related to exercise and lifestyle habits, such as increasing daily activity levels, stretching, calisthenics, walking, jogging, swimming / water training, Tai Chi, muscle training, bathing, mindfulness / meditation, aromatherapy (aromatic bath, bathing method, etc.), Epsom salt bath, skin moisturizing, digital detox, morning sun bathing, and sleep hygiene, and is configured to allow users to change the display / display of each item's icon. Therefore, when the add button for the lifestyle category is activated, the control unit 21 displays a selection screen (not shown) for indicating whether to display or not display the above items, and accepts the selection for displaying / displaying each item. The control unit 21 displays the icon of the selected item in the lifestyle category display bar of the practice input screen. Here, the icons for each lifestyle item are also provided by the health management application 22AP.

[0072] Regarding the therapy (medication) category, icons and additional buttons are provided to represent HRT (hormone replacement therapy), hormone medications other than HRT, and traditional Chinese medicine, respectively. In addition to the items mentioned above, items related to medication use in medical institutions are also provided as part of the therapy (medication) category, such as intravaginal estrogen administration and overactive bladder treatment, and are configured to toggle the display / display of each item's icon. Therefore, when the additional button for the therapy (medication) category is activated, the control unit 21 displays a selection screen (not shown) for indicating whether to display or not to display the aforementioned items, and accepts the selection for displaying / displaying each item. The control unit 21 displays the icon of the selected item in the therapy (medication) display bar of the practice input screen. Here, the icons for each therapy (medication) item are also provided by the health management application 22AP.

[0073] Regarding the category of "Therapies (excluding medication)," icons are provided to represent psychological counseling, acupuncture treatment, and bone setting / massage, respectively, as well as additional buttons. In addition to the items mentioned above, the "Therapies (excluding medication)" category also includes items related to treatments other than medication performed in medical institutions, such as cognitive behavioral therapy, reflexology, intravaginal laser irradiation, and aromatherapy (therapeutic methods), and is configured to toggle the display / display of each item's icon. Therefore, when the additional button for the "Therapies (excluding medication)" category is activated, the control unit 21 displays a selection screen (not shown) for indicating whether to display or not to display the aforementioned items, and accepts the selection to display or not display each item. The control unit 21 displays the icon of the selected item in the "Therapies (excluding medication)" display bar of the practice input screen. Here, the icons for each item in the "Therapies (excluding medication)" category are also prepared by the health management application 22AP.

[0074] Control unit 21 determines whether the save button on the practice input screen has been activated (S35). If it determines that the button has not been activated (S35: No), it returns to step S34 and continues to accept practice information input. If it determines that the save button has been activated (S35: Yes), control unit 21 sends the practice information representing the practice input through the practice input screen to server 10 (S36) and requests server 10 to save the practice information. Here, control unit 21 also sends the practice information along with the user ID and the date (or date and time) to server 10. Control unit 11 of server 10 stores the practice information in user DB12b in association with the received user ID and date (S37). Alternatively, control unit 21 may also store the practice information received through the practice input screen in association with the date (or date and time) in health management application 22AP or storage unit 22.

[0075] After processing step S30 or step S36, the control unit 21 generates a recording completion screen (S38) to notify that the recording of symptom information or practice information has been completed. Figure 10B This is an example of a screen indicating the completion of recording. For instance, control unit 21 compares symptom information entered chronologically within a recent specified period (e.g., one week, one month, etc.) with symptom information entered through the symptom input screen (the latest symptom information), determines whether there are any new symptoms, whether any symptoms have disappeared or lessened, or whether any symptoms have remained unchanged, and generates a message corresponding to the symptom change. For example, when control unit 21 determines that a new symptom has appeared, such as... Figure 10B The control unit 21 generates a message such as "New symptoms such as headache and hot flashes seem to have appeared" to display the message. Additionally, when the control unit 21 determines that symptoms have disappeared or lessened, it generates a message such as "The headache symptoms seem to have lessened," and when it determines that symptoms have not changed, it generates a message such as "The headache symptoms seem to be continuing." Furthermore, the health management application 22AP pre-prepares multiple messages to be provided to menopausal users, and the control unit 21 randomly selects a message or selects one based on changes in the user's symptoms (whether new symptoms have appeared, whether symptoms have disappeared or lessened, whether symptoms have not changed, etc.). Then, the control unit 21 generates a recording completion screen displaying the generated messages. Furthermore, in... Figure 10B The screen displays messages as "Today's Messages," either randomly or based on changes in the user's symptoms.

[0076] In addition, the control unit 21 can also compare the practice information entered in sequence within the most recent specified period with the practice information entered through the practice input screen (the latest practice information) to determine whether there are any changes in the practice content, such as whether there are any newly started practices, whether there are any stopped or interrupted practices, or whether there are any practices that continue to be executed. It then generates a message corresponding to the changes in the practice content and displays it on the recording completion screen.

[0077] The control unit 21 displays the generated recording completion screen on the display unit 25 (S39) and determines whether the OK button on the recording completion screen has been operated (S40). If the control unit 21 determines that the OK button has not been operated (S40: No), it waits until the OK button is operated. If it determines that the OK button has been operated (S40: Yes), it returns to step S22 and... Figure 9A The top-level screen shown is displayed on the display unit 25 (S22). When the control unit 21 determines that the suggestion button in the top-level screen has been operated (S24: Yes), it sends the user's user ID to the server 10 and requests suggestions from the server 10 corresponding to the user's symptom information (S41).

[0078] When the control unit 11 of server 10 receives a suggestion request, it reads the user's symptom information from user DB12b (S42). For example, the control unit 11 reads symptom information entered in the past month (e.g., symptom information entered regularly, such as once a day). Furthermore, the read symptom information is not limited to the past month's information; it can also be information from the past week, two weeks, two months, etc. Based on the read symptom information, the control unit 11 counts the frequency of occurrence (occurrence frequency) of each symptom and determines five symptoms in descending order of frequency (S43). Then, based on the symptom information of the five determined symptoms, the control unit 11 determines suggestions corresponding to each of the five symptoms (S44). Step S44 is the same process as step S18; the control unit 11 inputs the symptom information of the determined symptoms (e.g., identification information of each symptom) into the input node of the learning model 12M and obtains the suggestion ID from the output node associated with the symptom of the input symptom information. Here, when the symptom information includes the severity of each symptom, the control unit 11 also inputs the identification information and severity of each symptom into the input node of the learning model 12Ma, and obtains the suggestion ID corresponding to each symptom from the output node associated with the symptom in the input symptom information. Furthermore, when the severity of each symptom is input, the control unit 11 can input the latest severity of the determined symptom into the learning model 12Ma, or it can calculate the average severity over a specified period and input it into the learning model 12Ma.

[0079] Based on the suggestion ID for each symptom determined in step S44, control unit 11 reads the suggestion corresponding to the suggestion ID from suggestion DB12a and generates a suggestion as shown in step S44. Figure 11A The suggested screen shown is (S45). Figure 11A The suggested screen has the same Figure 8B The suggestion screen has the same structure, providing suggestions that take into account the user's symptoms entered within the specified period. The control unit 11 sends the generated suggestion screen to the user terminal 20 (S46), and the control unit 21 of the user terminal 20 receives the suggestion screen from the server 10 and displays it on the display unit 25 (S47).

[0080] Figure 11B This shows an example of the history screen displayed when the history button on the screen is pressed. For example, when the history button is pressed on the top-level screen, the control unit 21 will display the history screen as follows: Figure 11BThe historical screen shown is displayed on the display unit 25. Specifically, the control unit 21 acquires symptom information stored in the health management application 22AP or the storage unit 22, or symptom information stored in the user DB12b on the server 10. For example, the control unit 21 acquires symptom information for the most recent two months. Then, in the two months of symptom information, the control unit 21 counts the frequency of occurrence of each symptom in the first half of the month based on the symptom information for the first half of the month, and counts the frequency of occurrence of each symptom in the second half of the month based on the symptom information for the second half of the month. In addition, the control unit 21 identifies the symptoms that occur frequently in the second half of the month, and counts the frequency of occurrence of each symptom in each week of the second half of the month. Furthermore, the control unit 21 compares the frequency of occurrence of each symptom in the first half of the month with the frequency of occurrence of each symptom in the second half of the month, determines whether the frequency of occurrence is increasing, decreasing, or remaining constant, and determines the change of each symptom. In addition, the control unit 21 calculates, for example, the rate of increase or decrease in frequency of occurrence as the change of each symptom. The control unit 21 generates a historical screen displaying the information acquired as described above. Therefore, the control unit 21 is able to generate, for example... Figure 11B The historical scene shown. Figure 11B The screen displays the most frequent symptoms and their frequency in the second half of the month (e.g., this month), shows messages related to symptoms that have changed compared to the first half of the month (e.g., last month), and further displays a chart to show the frequency of each symptom in each week of the second half of the month.

[0081] In addition to considering the frequency of occurrence of each symptom (the selected frequency), the control unit 21 can also consider the severity level input for each symptom to determine the change of each symptom from last month to this month. For example, for each occurrence, a higher severity level can be assigned a greater weight, thereby calculating the occurrence score of each symptom, comparing the occurrence scores of each symptom in the first half of the month and the second half of the month, and determining the change of each symptom by calculating the increase or decrease rate of the occurrence score. Figure 11B The screen displays the history for this month. It can also be configured to switch to the history for the previous month (last month) when a specified action is performed (e.g., a swipe to the right), and switch to the history for the next month when a specified action is performed (e.g., a swipe to the left).

[0082] also, Figure 11BThe screen includes a "This Month" button for displaying the current month's historical screen and a "This Year" button for displaying the current year's historical screen. When the "This Year" button is activated, the control unit 21 generates and displays a historical screen based on symptom information from the current year. Specifically, the control unit 21 counts the frequency of each symptom throughout the year based on the symptom information from the current year, identifies the most frequent symptoms, and counts the frequency of each symptom in each month of the current year. Furthermore, if the control unit 21 has stored symptom information from the previous year, it compares the frequency of each symptom from the previous year with the frequency of each symptom in the current year to determine changes in the frequency of each symptom. Therefore, the control unit 21 displays the most frequent symptoms and their frequencies from the current year, displays messages related to symptoms that have changed compared to the previous year, and generates a historical screen (not shown) displaying a chart showing the frequency of each symptom in each month of the current year.

[0083] Through the above processing, in this embodiment, when a user first inputs symptom information, suggestions corresponding to the inputted symptoms can be provided to the user. Furthermore, when a user continuously and regularly inputs symptom information over a specified period (e.g., one month, one week, etc.), suggestions corresponding to frequently occurring symptoms can be provided to the user. Therefore, users can obtain knowledge related to the symptoms they are concerned about and receive advice for alleviating those symptoms.

[0084] This embodiment is configured to determine frequently occurring symptoms based on symptom information input by the user through the symptom input screen within a specified period, and to determine recommendations to be provided to the user based on the frequently occurring symptoms. Alternatively, recommendations to be provided to the user can be determined based on chronological symptom information input within the specified period. For example, learning models 12M and 12Ma can be configured to output suggestion IDs corresponding to each symptom after chronological symptom information is input. Furthermore, for example, similar to a smartwatch, it can collaborate with a measurement device used to measure the user's biometric information such as body temperature, blood pressure, heart rate, pulse rate, and blood glucose levels, determining recommendations to be provided to the user based on symptom information input by the user through the initial input screen and the symptom input screen, as well as the user's biometric information measured by the measurement device. In this case, learning models 12M and 12Ma can also be configured to input biometric information in addition to symptom information, and output suggestion IDs corresponding to each symptom based on the input symptom information and biometric information.

[0085] In this embodiment, the specific processing of suggestions corresponding to the symptom information input by the user can be configured to be performed locally by the user terminal 20. For example, by storing the learning models 12M and 12Ma in the storage unit 22 of the user terminal 20, the control unit 21 inputs the symptom information received through the input unit 24 into the learning models 12M and 12Ma, and determines the suggestions to be provided to the user based on the output information from the learning models 12M and 12Ma. In this case, the control unit 21 of the user terminal 20 can generate suggestions based on the determined suggestions. Figure 8B and Figure 11A The suggested screen is shown on the display unit 25. Even in this configuration, the same processing as in this embodiment can be achieved, and the same effect can be obtained.

[0086] In this embodiment, Figure 5 Step S18 and Figure 7 The determination of the suggestions corresponding to each symptom in step S44 is not limited to the structure using learning models 12M and 12Ma, but can also be a rule-based structure. For example, a database (DB) can be used, which records suggestion IDs associated with each of multiple sets containing a specified number of symptoms and the severity of each symptom. In this case, the control unit 11 determines the suggestion ID of each symptom corresponding to the symptom information input by the user based on the DB, and determines the suggestion with the determined suggestion ID as the suggestion to be provided to the user for each symptom. Furthermore, for example, an importance (priority) can be assigned to each suggestion prepared for each symptom, and a score can be calculated for each symptom based on the combination of a specified number of symptoms selected by the user and the severity of each symptom, and the suggestion with the importance corresponding to the calculated score can be determined as the suggestion to be provided to the user for each symptom. The score for each symptom can be configured to be calculated based on a score set for each symptom, a score set for the combination of symptoms, and a score set for the severity of each symptom. Furthermore, the control unit 11 can also determine a predetermined number of symptoms for which suggestions should be provided to the user based on the symptom information of the time sequence input by the user, and determine suggestions corresponding to the input symptom information of the time sequence for the determined symptoms. Even in this configuration, the same processing as in this embodiment can be achieved, and the same effect can be obtained.

[0087] Implementation Method 2

[0088] The information processing system described herein uses image processing to determine the test results obtained from a hormone level testing kit for examining a user, and provides suggestions corresponding to the determination results. The information processing system of this embodiment can be used by... Figure 1 and Figure 2The information processing system of Embodiment 1 shown is implemented using the same apparatus; therefore, the description of the structure of each apparatus will not be repeated. Furthermore, the user terminal 20 of this embodiment, in addition to having… Figure 2 In addition to the structure shown, it also includes a camera (not shown), and the storage unit 22 stores the inspection and judgment model 22M (reference). Figure 12A The inspection results are determined using a learning model.

[0089] A camera is a shooting device that includes lenses and imaging elements. In this embodiment, the camera performs processing to acquire one image (still image) per operation of the shutter button, or it can perform processing to acquire 15 or 30 frames of images (video) per second, for example, per operation of the shutter button. The inspection and judgment model 22M can be built into the health management application 22AP and stored in the storage unit 22, or it can be stored in a storage unit 22 separate from the health management application 22AP. The inspection and judgment model 22M is, for example, a model learned from training data through machine learning, and can be envisioned as a program module constituting artificial intelligence software. In addition to being stored in the user terminal 20, the inspection and judgment model 22M can also be stored in the storage unit 12 of the server 10, or it can be stored on other servers besides the server 10. In this case, the user terminal 20 can be configured to access the server 10 or other servers storing the inspection and judgment model 22M to read the inspection and judgment model 22M, or it can be configured to perform the judgment processing using the inspection and judgment model 22M on the server 10 or other servers.

[0090] Figure 12A This is a schematic diagram representing a structural example of the inspection and judgment model 22M. Figure 12AThe test determination model 22M shown is a model that takes an image of a test kit used to measure hormone levels (specifically, an image of the test kit's decision lines) as input, performs calculations to infer the test results obtained from the test kit based on the input image, and outputs the calculation results. The test kit can be, for example, a urine test kit that can be used at home, and can measure the presence and amount of hormones in the test subject. Furthermore, the test kit indicates the presence or absence of hormones by the presence or absence of decision lines, and the amount of hormones is indicated by the thickness or color intensity of the decision lines. The test kit can detect female hormones such as FSH (follicle-stimulating hormone) or LH (luteinizing hormone), and can measure one or more hormones. In addition, the test kit is not limited to urine tests; it can also be a test kit that uses saliva or blood to measure hormone levels. The test results output from the test determination model 22M represent the amount of each hormone in the test subject using the test kit. The test determination model 22M can be constructed using algorithms such as CNN, decision tree, random forest, SVM, and Transformer, or by combining multiple algorithms.

[0091] Figure 12A The inspection and judgment model 22M shown takes the image captured by the inspection kit as input through the input node, calculates the output value using various functions and thresholds based on the input image, and outputs the calculated output value. Figure 12A The test judgment model 22M shown has multiple output nodes, each of which is associated with multiple hormones that are the test objects. The measured amount of the associated hormone (the test result obtained from the test kit) is output from each output node.

[0092] The diagnostic model 22M is generated by learning through training data that correlates images taken with a diagnostic kit used for training with the correct hormone measurements (measurements) determined by doctors or others based on the kit's determination lines. The training data is generated based on the results of tests conducted on patients undergoing treatment for menopausal symptoms, patients aware of their menopausal symptoms, and patients without menopausal symptoms. The diagnostic model 22M is learned to output the correct hormone measurements when given the images included in the training data. Specifically, the diagnostic model 22M performs calculations based on the input images, inferring the amounts of each hormone based on the diagnostic lines of the diagnostic kit in the images, and obtaining the inference results. Then, the diagnostic model 22M compares the inferred hormone amounts with the correct measurements, optimizing parameters such as the weights (association coefficients) between nodes in the diagnostic model 22M to make the two approximate. Parameter optimization methods can include steepest descent, backpropagation, etc. Thus, the test judgment model 22M is obtained, which, when inputting an image captured by the test kit, outputs the measured amount of each hormone as the test result obtained from the test kit.

[0093] Depend on Figure 12A The measured values ​​of each hormone output by the inspection and judgment model 22M shown are represented by continuous values. As a so-called regression problem, the inspection and judgment model 22M outputs any one of the continuous values. However, it can also be viewed as a classification problem, judging any one of the selectable values ​​for each hormone's measured value. In this case, the inspection and judgment model 22M is configured to determine the optimal value from a set of preset values ​​as the measured value for each hormone and output the judgment result. Alternatively, the inspection and judgment model 22M can be configured to determine whether the amount of each hormone is within the normal range based on the input captured image and output the judgment result. The learning of the inspection and judgment model 22M can be performed on the server 10 or other learning devices. The learned inspection and judgment model 22M, generated by learning on the server 10 or other learning devices, is, for example, embedded in the health management application 22Ap and downloaded to the user terminal 20 and stored in the storage unit 22.

[0094] Figure 12B This is a schematic diagram illustrating a structural example of the learning model 12Mb in Embodiment 2. The learning model 12Mb of this embodiment has the same characteristics as... Figure 4AThe learning model 12M of Implementation Method 1 shown has the same structure. As input data, in addition to symptom information for each symptom, hormone test results are also input. Hormone test results can be the measured levels of each hormone inferred from images taken by the test kit using the test determination model 22M, or they can be test results determined based on the measured levels of each hormone and questionnaire responses (e.g., test results related to the stage or severity of menopausal disorders). Furthermore, and Figure 4B Similarly, the learning model 12Mb of this embodiment can be configured to take into account symptom information including the severity of each symptom as input.

[0095] Figure 13 This is a flowchart illustrating an example of the provision process flow for the proposed implementation method 2. Figure 14 This is a schematic diagram representing an example of a screen display. Figure 13 The process shown is in Figures 5 to 7 The process shown includes steps S51 to S60 added between steps S25 and S26. Regarding... Figures 5 to 7 The same steps will not be repeated here. Additionally, in Figure 13 in, omit Figures 5 to 7 A diagram showing steps other than S25 to S26.

[0096] In this embodiment, the control unit 21 of the user terminal 20 proceeds to step S51 after processing in step S25. Furthermore, the symptom input screen displayed by the control unit 21 in step S25 includes, in addition to, […]. Figure 9B The structure shown also includes a "Test Kit Reading" button to indicate the reading of images captured by the test kit (see reference). Figure 15 (B1 in the example). Therefore, the control unit 21 determines whether the "Check Kit Reading" button in the symptom input screen has been operated (S51). If it is determined that it has not been operated (S51: No), the process proceeds to step S26. In this case, the control unit 21 performs the same processing as in Embodiment 1.

[0097] When the "Read Test Kit" button is activated (S51: Yes), the control unit 21 activates the camera (S52) and begins displaying the images acquired by the camera on the display unit 25. The user confirms the displayed images, for example, by taking a still image with the entire test kit within the shooting range. When the control unit 21 receives the user's shooting instruction via the input unit 24, it takes a picture of the test kit (S53) and acquires the image of the test kit. The control unit 21 stores the acquired image in the storage unit 22. The control unit 21 infers the test results obtained from the test kit based on the acquired image (S54). Specifically, the control unit 21 inputs the image of the test kit into the test determination model 22M and acquires the measured values ​​of each hormone as output values ​​from the test determination model 22M.

[0098] Secondly, the control unit 21 will be as follows Figure 14 The questionnaire screen is displayed on the display unit 25 (S55), and responses to each questionnaire are received through the questionnaire screen (S56). The control unit 21 determines whether the OK button on the questionnaire screen has been operated (S57). If it is determined that it has not been operated (S57: No), it returns to step S56 and continues to receive questionnaire responses. If it is determined that the OK button has been operated (S57: Yes), the control unit 21 determines the hormone test results based on the examination results inferred in step S54 and the questionnaire responses obtained in step S56 (S58). The hormone test results may be the measured amounts of each hormone, which are modified based on the questionnaire responses from the test results obtained from the test kit, or the stage or severity of menopausal disorders determined based on the test results obtained from the test kit and the questionnaire responses. The control unit 21 stores the determined hormone test results in the storage unit 22.

[0099] Hormone test results can be determined based on rules or using a learning model. For example, a database (DB) can be used, which records hormone test results associated with the status of each hormone in relation to each of multiple sets including hormone measurements and questionnaire responses. In this case, control unit 21 can determine the hormone test results corresponding to the hormone measurements and questionnaire responses based on the DB. Alternatively, importance (priority) can be assigned to each hormone or questionnaire, and a score for the user's hormone test can be calculated by considering weights corresponding to the importance, and the hormone test result corresponding to the calculated score can be determined. Alternatively, a learning model can be used to determine hormone test results, which takes hormone measurements and questionnaire responses as input, infers the user's hormone test results based on the input data, and outputs the results. For example, a database can be used to determine the hormone test results. Figure 12AThe 22M test model is configured to take the images captured by the test kit and the answers to each questionnaire as inputs, and output the hormone test results.

[0100] Control unit 21 sends the determined hormone test results to server 10 (S59). Control unit 21 may send the hormone test results along with the user ID and the date (or date and time) to server 10, and control unit 11 of server 10 stores the hormone test results in association with the receiving user ID and date in, for example, user DB12b (S60). Alternatively, control unit 21 may store the hormone test results in association with the date (or date and time) in health management application 22AP or storage unit 22. The above processing structure determines the hormone test results based on the measured amounts of each hormone determined from the images captured by the test kit and the responses to each questionnaire. Alternatively, the measured amounts of each hormone may be directly used as the hormone test results. In this case, control unit 21 may directly send the test results inferred in step S54 and the accepted questionnaire responses in step S56 as the hormone test results to server 10, or it may skip steps S55 to S57 and directly send the test results inferred in step S54 to server 10.

[0101] After processing in step S59, control unit 21 proceeds to step S26. Thereby, it accepts input of symptoms and their severity, as well as practice information, and stores the received symptom information and practice information in user DB12b of server 10. Furthermore, in server 10 of this embodiment, in step S44, in addition to the symptom information determined in step S43, control unit 11 also inputs the hormone test results stored in user DB12b in step S60 into learning model 12Mb, and obtains the suggestion ID for each symptom as output data from learning model 12Mb, thereby determining suggestions corresponding to each symptom. Furthermore, when symptom information and hormone test results obtained in a time sequence are stored in user DB12b, the time sequence data of symptom information and hormone test results is input into learning model 12Mb. Figure 12B The learning model 12Mb can also be configured to input, in addition to the symptom information of each symptom, the measured values ​​of each hormone and the answers to each questionnaire survey. In this case, in addition to the symptom information of the symptoms determined in step S43, the control unit 11 will also input the measured values ​​of each hormone and the answers to each questionnaire survey determined by the inspection and judgment model 22M into the learning model 12Mb and determine the recommendations corresponding to each symptom.

[0102] In this embodiment, the same effects as in Embodiment 1 are achieved. Furthermore, in this embodiment, in addition to information related to the symptoms input by the user, suggestions considering the test results obtained from the testing kit can also be provided to the user. Therefore, suggestions more suitable for the user's condition can be provided.

[0103] Implementation Method 3

[0104] In the information processing system of embodiments 1 to 2, a modified example of the display screen of the user terminal 20 will be described. Figure 15 This is a schematic diagram illustrating a variation of the symptom input screen. Figure 9B The example shown is a variation of the symptom input screen. Figure 15 The screen includes buttons for symptoms, causes, actions, and notes. Figure 15 This indicates that the symptom button is selected. Figure 15 The screen displays icons representing four options related to emotions and bleeding, and also includes a "Read Test Kit" button (B1). Therefore, in Figure 15 Similar to Embodiment 2, the screen can also capture images of the test kit, infer hormone test results based on the images, and register them to server 10. Additionally, in Figure 15 The screen displays input buttons for various symptoms across multiple categories, including physical, sensitive areas, psychological, and sleep conditions, allowing users to indicate whether they are asymptomatic or have any level of severity. Figure 15 The screen allows users to input the status of various symptoms across different categories with a single action (such as clicking an input button), thus enabling seamless communication with the system. Figure 9B The graphics are improved compared to the previous version, offering enhanced operability. Figure 15 In the screen, when the action button is pressed, the user terminal 20 displays the same information as... Figure 10A The same input screen is used.

[0105] When in Figure 15 When the "Reason" button is pressed on the screen, the user terminal 20 displays a "Reason Input Screen" (not shown). This screen accepts input regarding the user's current state and includes input buttons for expressing concerns or stress, such as work-related issues, workplace relationships, household chores, family relationships, diet, and sleep. The control unit 21 of the user terminal 20 receives this "Reason" input and sends it to the server 10, where it is stored in the user's DB12b. Additionally, in... Figure 15When the memo button is activated on the screen, the user terminal 20 displays a memo input screen (not shown). The memo input screen is used to accept input such as memos or diaries, for example, text data input via the input unit 24. The control unit 21 of the user terminal 20 uses this screen to send the received memos or diaries to the server 10 and store them in the user DB12b. The reason for storing them in the user DB12b and the memos themselves can be used by the server 10 when making recommendations. For example, the server 10 can... Figure 12B The learning model 12Mb shown is configured to accept input of causes and memos, in addition to hormone test results and symptom information for each symptom. In this case, causes and memos, along with symptom information and hormone test results, can be read from the user DB12b and input into the learning model 12Mb to determine recommendations for each symptom.

[0106] Figure 16 This is a schematic diagram illustrating a variation of a historical scene. Figure 11B The historical image shown is a variation. Figure 16 The screen shows that during the specified period (in Figure 16 The data includes the five most frequent symptoms appearing within a month, time-series data of hormone test results (measurements of each hormone), and the five most frequent reasons entered within the specified period. Regarding the time-series data of hormone test results, if multiple hormone test results are stored in the user's DB12b, a chart representing the changes in multiple hormone test results is created and displayed. Additionally, in... Figure 16 The screen displays information such as changes in symptoms and their frequency compared to those of the previous month. Figure 16 The images make it easy to identify the most frequent symptoms and their causes, and also allow for a clear understanding of the progression of hormone test results.

[0107] In the information processing system of this embodiment, only the symptom input screen and the history screen differ from those of embodiments 1 and 2. It can perform the same processing as embodiments 1 and 2 and achieve the same results. In this embodiment, for... Figure 15 and Figure 16 The items displayed on the screen shown can be selected to be shown or hidden, and the operability can be improved by customizing the items displayed according to the user's preferences.

[0108] Implementation Method 4

[0109] An information processing system that uses a language model to generate suggestions for users is described. The information processing system of this embodiment can be used by... Figure 1 and Figure 2The information processing system of Embodiment 1 shown is implemented using the same apparatus, therefore the structure of each apparatus will not be described in detail. Furthermore, the server 10 of this embodiment, in addition to... Figure 2 The structure shown also includes a language model (not shown) stored in the storage unit 12. Furthermore, in this embodiment, in addition to storing symptom information and practice information, the user DB12b stored in the storage unit 12 also stores (accumulates) cause information and memo information input via the symptom input screen, similar to that in Embodiment 2.

[0110] A language model is a general-purpose large language model (LLM) built by pre-learning from large groups of articles. The language model is learned to perform operations on input data, such as text data described in natural language, to generate output data corresponding to the content of the input data, and then output the generated output data. Language models can be constructed using algorithms such as GPT (Generative Pre-trained Transformer)-3, GPT-3.5, and GPT-4, or by combining multiple algorithms. Furthermore, language models are not limited to the aforementioned Transformer-based models. Language models are envisioned as program modules constituting artificial intelligence software. In addition to being configured to be stored in storage unit 12, language models can also be accessed and read by a language processing server storing the language model from server 10. Furthermore, the language model can be fine-tuned using user symptom information, hormone test results, and suggestions to be provided to the user. In addition to fine-tuning the overall model, the language model can also be fine-tuned using techniques such as LoRA (Low-Rank Adaptation) to fine-tune the parameters of only a subset of layers.

[0111] Figure 17A and Figure 17B This is a schematic diagram representing an example. Figure 17A It is used for passing through Figure 8A The example shown is an instruction from the language model to generate suggestions for the user when symptom information is entered for the first time on the initial input screen. Figure 17B It is used for passing through Figure 9B The example shown is an instruction from the language model to generate suggestions for the user when symptom information is entered on the symptom input screen. Figure 17AExamples include command statements for the processing to be performed, and articles related to symptom information that should be referenced when generating recommendations. Command statements may include articles instructing on the content of the recommendations, articles instructing on considerations when generating recommendations, and articles instructing on the output format, etc. Articles instructing on the output format may, for example, include an article such as, "Please create a weekly plan regarding diet and exercise. Here, please list the diet and exercise to be performed each day from Monday to Sunday." Storage unit 12 stores... Figure 17A The prompt only contains a template of command statements. The control unit 11 of server 10, based on this prompt template, reads and inserts the user's symptom information from user DB12b, suggesting the generation of the object, thereby creating a... Figure 17A The prompt shown.

[0112] exist Figure 17B Examples include command statements for the processes to be executed, and articles related to symptom information, cause information, practice information, and memo information that should be referenced when generating recommendations. Storage unit 12 stores information related to... Figure 17B The prompt only contains a template of command statements. The control unit 11 of server 10, based on this prompt template, reads and inserts the user's symptom information, cause information, practice information, and memo information from user DB12b, and thus creates a prompt template. Figure 17B The prompt shown.

[0113] The information processing system of this embodiment executes and Figures 5 to 7 The same process applies. Furthermore, in this embodiment, the server 10 uses a language model when determining suggestions in steps S18 and S44. For example, in step S18, the control unit 11 reads the symptom information stored in the user DB12b in step S17 from the user DB12b and inserts it into a prompt template containing command statements, thereby creating... Figure 17A The control unit 11 then inputs the created prompts into the language model and obtains suggestions for each symptom generated by the language model. The control unit 11 then performs processing after step S19. Additionally, in step S44, the control unit 11 reads the user's cause information, practice information, and memo information from the user DB12b, and inserts them, along with the symptom information read in step S42, into a prompt template containing command statements, thereby creating... Figure 17BThe control unit 11 then inputs the created prompts into the language model and obtains suggestions for each symptom generated by the language model. The control unit 11 then performs processing after step S45. Furthermore, when the user DB12b stores past symptom information, cause information, practice information, and memo information, the control unit 11 can insert past (e.g., the most recent week, month, or several months) symptom information, cause information, practice information, and memo information into the prompts.

[0114] Through the above processing, this embodiment can generate suggestions to be provided to the user by taking into account the symptom information, cause information, practice information, and memo information input by the user. Furthermore, the control unit 11 of the server 10 can, for example, after the processing in steps S31 and S37, generate suggestions or messages using a language model based on the symptom information, cause information, practice information, and memo information stored in the user's DB12b at that time, and generate... Figure 10B The recording completion screen shown is provided to the user terminal 20. In this case, the control unit 21 of the user terminal 20 can obtain the recording completion screen generated by the server 10 in step S38 and display it on the display unit 25.

[0115] In this embodiment, by Figure 17A and Figure 17B The prompts contain articles related to the user's preferences and their liking for suggestions (e.g., whether they prefer a simple one-sentence suggestion or a suggestion with detailed content), thus instructing the language model to generate suggestions corresponding to the user's preferences. In this case, suggestions that match the user's preferences can be generated.

[0116] In the information processing system of this embodiment, when the server 10 determines a suggestion, the process of generating the suggestion using a language model differs from that in embodiments 1 to 3, while other processes are the same as in embodiments 1 to 3. Therefore, the same effect can be obtained. Furthermore, in this embodiment, since a language model is used to generate suggestions, suggestions based on various information related to the user can be generated, and suggestions suitable for the user's state can be provided.

[0117] All aspects of the embodiments disclosed herein should be considered as examples rather than limiting descriptions. The scope of the invention is defined by the claims rather than by the foregoing meaning, and is intended to include all modifications within the meaning and scope equivalent to the claims.

[0118] The items described in the above embodiments can be combined with each other. Furthermore, the independent and dependent claims described in the claims can be combined in any combination, regardless of the form of reference. Moreover, although the claims may use a form that refers to two or more other claims (multiple claim form), this is not a limitation. A multiple claim form that refers to at least one multiple claim (multiple reference multiple claim) may also be used.

[0119] Explanation of reference numerals in the attached figures

[0120] 10 servers

[0121] 11 Control Department

[0122] 12 Storage Department

[0123] 13 Ministry of Communications

[0124] 14 Input Section

[0125] 15 Display Section

[0126] 20 User Terminals

[0127] 21 Control Department

[0128] 22 Storage Department

[0129] 23 Ministry of Communications

[0130] 12M learning model

Claims

1. A program, characterized in that, The computer will perform the following processes: Obtain symptom information related to menopausal symptoms in users; The acquired symptom information is input into the learning model, and suggestions corresponding to the input symptom information are obtained, wherein the learning model is learned to output suggestions related to the symptoms when symptom information related to menopausal symptoms is input; Output the obtained recommendations.

2. The procedure according to claim 1, characterized in that, The learning model is learned to output suggestions related to each of the multiple symptoms when given symptom information related to multiple symptoms. The program causes the computer to perform the following processes: When symptom information related to multiple symptoms of the user is obtained, the obtained symptom information related to multiple symptoms is input into the learning model, and suggestions related to each of the multiple symptoms are obtained.

3. The procedure according to claim 1 or 2, characterized in that, The symptom information includes information related to hot flashes, headaches, palpitations, shoulder pain, anxiety, or sleep patterns.

4. The procedure according to claim 1 or 2, characterized in that, The recommendations include those related to physical care or lifestyle habits.

5. The procedure according to claim 1 or 2, characterized in that, The program causes the computer to perform the following processes: Periodically obtain the user's symptom information; Based on the symptom information obtained within the specified period, identify the symptoms that occur most frequently; The learning model inputs symptom information related to the identified symptoms and obtains suggestions corresponding to the input symptom information.

6. The procedure according to claim 1 or 2, characterized in that, The symptom information includes the type and severity of the symptoms. The learning model is learned to output suggestions related to the symptoms when given symptom information including the type and severity of the symptoms.

7. The procedure according to claim 1 or 2, characterized in that, The program causes the computer to perform the following processes: Based on the symptom information obtained in chronological order, the changes in the symptoms are determined; Output the identified changes in the symptoms.

8. The procedure according to claim 1 or 2, characterized in that, The program causes the computer to perform the following processes: Images were captured using a test kit that examined the user's hormone levels. The acquired image is input into the inspection result determination learning model, and the inspection result obtained from the inspection kit in the input image is obtained, wherein the inspection result determination learning model is learned to output the inspection result obtained from the inspection kit in the captured image when the captured image of the inspection kit is input. The learning model is learned to output suggestions related to the symptoms when given symptom information related to menopausal symptoms and test results from the test kit. The acquired symptom information and the test results obtained from the test kit are input into the learning model, and suggestions corresponding to the input symptom information and test results are obtained.

9. The procedure according to claim 8, characterized in that, The program causes the computer to perform the following processes: Obtain responses to questionnaires related to the user's symptoms; Based on the test results obtained by the learning model from the test kit and the answers to the questionnaire, the hormone test results are determined. The learning model is learned to output suggestions related to the symptoms when given symptom information related to menopausal symptoms and the hormone test results; The acquired symptom information and the determined hormone test results are input into the learning model, and suggestions corresponding to the input symptom information and hormone test results are obtained.

10. The procedure according to claim 1 or 2, characterized in that, The program causes the computer to perform the following processes: Based on the obtained symptom information, a language model is used to generate suggestions corresponding to the symptom information.

11. A program, characterized in that, The computer will perform the following processes: Obtain symptom information related to the symptoms of multiple menopausal users in chronological order; Based on the multiple symptom information obtained within the specified period, a specified number of symptoms should be recommended to the user; For a given number of symptoms, recommendations are made corresponding to multiple symptom information obtained within the given period; Output the determined recommendations.

12. An information processing method, characterized in that, The computer performs the following processing: Obtain symptom information related to menopausal symptoms in users; The acquired symptom information is input into the learning model, and suggestions corresponding to the input symptom information are obtained, wherein the learning model is learned to output suggestions related to the symptoms when symptom information related to menopausal symptoms is input; Output the obtained recommendations.

13. An information processing apparatus comprising a control unit, characterized in that, The control unit is used for: Obtain symptom information related to menopausal symptoms in users; The acquired symptom information is input into the learning model, and suggestions corresponding to the input symptom information are obtained, wherein the learning model is learned to output suggestions related to the symptoms when symptom information related to menopausal symptoms is input; Output the obtained recommendations.

Images