Program, information processing method, and information processing device

Abstract

Provided is a program, etc., capable of providing knowledge about symptoms of a menopausal period.　A computer acquires symptom information pertaining to the symptom of a user in the menopausal period in accordance with the program. The computer inputs the acquired symptom information into a learning model trained so as to output, upon receiving input of symptom information pertaining to symptoms of the menopausal period, tips relating to the symptom, and acquires tips corresponding to the input symptom information. The computer then outputs the acquired tips.

Description

Program, information processing method, and information processing device

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

[0002] Patent Literature 1 discloses a system that can provide obese patients with psychological therapies such as behavioral therapy, cognitive therapy, and coaching, which are currently performed at medical institutions, at the required timing. By using the system disclosed in Patent Literature 1, it is possible to correct the patient's misinterpretations regarding obesity and obesity treatment at the required timing and provide the patient with correct knowledge. As a result, when an appetite urge arises, for example, the appetite urge can be controlled based on correct knowledge, and the appetite urge can be effectively eliminated.

[0003] Patent No. 6862009

[0004] The average age at which women reach menopause is said to be around 50 years old, and the 10-year period consisting of the five years before menopause and the five years after menopause is generally referred to as the menopause. It is known that various symptoms appear during menopause due to the complex interaction of various factors, including physical, psychological, and social factors. Because symptoms of menopause vary greatly from person to person, it is difficult for women to determine whether or not they should visit a medical institution, and many women have no one to consult with. The system in Patent Document 1 provides accurate information to obese patients, but does not provide information about menopausal symptoms.

[0005] The present disclosure aims to provide a program or the like that can provide knowledge about menopausal symptoms.

[0006] A program according to one aspect of the present disclosure causes a computer to execute a process of acquiring symptom information regarding a user's symptoms during menopause, inputting the acquired symptom information into a learning model trained to output tips regarding the symptoms when symptom information regarding menopausal symptoms is input, acquiring tips according to the input symptom information, and outputting the acquired tips.

[0007] In one aspect of the present disclosure, knowledge regarding menopausal symptoms can be provided.

[0008] 1 is an explanatory diagram showing an example of the configuration of an information processing system; FIG. 2 is a block diagram showing an example of the configuration of a server and a user terminal; FIG. 3 is an explanatory diagram showing an example of the record layout of a tips DB; FIG. 4 is an explanatory diagram showing an example of the record layout of a user DB; FIG. 4 is an explanatory diagram showing an example of the configuration of a learning model; FIG. 5 is an explanatory diagram showing an example of the configuration of a learning model; FIG. 6 is a flowchart showing an example of a tips provision processing procedure; FIG. 7 is a flowchart showing an example of a tips provision processing procedure; FIG. 8 is an explanatory diagram showing an example of a screen; FIG. 9 is an explanatory diagram showing an example of a screen; FIG. 10 is an explanatory diagram showing an example of a screen; FIG. 11 is an explanatory diagram showing an example of a screen; FIG. 12 is an explanatory diagram showing an example of a screen; FIG. 13 is an explanatory diagram showing an example of a screen; FIG. 14 is an explanatory diagram showing an example of a screen;

[0009] Hereinafter, a program, an information processing method, and an information processing device according to the present disclosure will be described in detail with reference to the drawings illustrating embodiments thereof.

[0010] (Embodiment 1) This section describes an information processing system that provides tips tailored to the symptoms of menopausal users. Menopause generally refers to the 10-year period consisting of the five years before menopause and the five years after menopause. Since the average age of menopause for women is said to be around 50, the period from approximately 45 to 55 years of age is referred to as menopause. FIG. 1 is an explanatory diagram showing an example configuration of an information processing system. The information processing system of this embodiment includes a server 10 and multiple user terminals 20, which are communicatively connected via a network N such as the Internet. The user terminals 20 are information terminals used by users who use services provided by the server 10, and multiple user terminals 20 are provided.

[0011] FIG. 2 is a block diagram showing an example configuration of the server 10 and the user terminal 20. The server 10 is an information processing device capable of various information processing and transmitting and receiving information, and is configured as a server computer, a personal computer, or the like. The server 10 includes a control unit 11, a memory unit 12, a communication unit 13, an input unit 14, a display unit 15, a reading unit 16, and the like, which are interconnected via a bus. The control unit 11 is configured using one or more processors, such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), a GPU (Graphics Processing Unit), a TPU (Tensor Processing Unit), or an AI chip (AI semiconductor). The control unit 11 executes a program 12P stored in the memory unit 12 as needed to perform information processing and control processing to be performed by the server 10.

[0012] The memory unit 12 includes RAM (Random Access Memory), flash memory, a hard disk, an SSD (Solid State Drive), etc. The memory unit 12 stores a program 12P (program product, computer program) executed by the control unit 11 and various data. The memory unit 12 also temporarily stores data generated when the control unit 11 executes the program 12P. The memory unit 12 also stores a learning model 12M that has learned training data, for example, through machine learning. The learning model 12M is a model that inputs symptom information related to symptoms known to occur during menopause and outputs information about tips to be provided to users based on the input symptom information. The learning model 12M is expected to be used as a program module constituting artificial intelligence software. The learning model 12M performs a predetermined calculation on the input values ​​and outputs the calculation results. The memory unit 12 stores data such as coefficients and thresholds of functions that define this calculation as the learning model 12M. The memory unit 12 also stores a tips DB 12a and a user DB 12b. The memory unit 12 may be composed of multiple memory devices, and part of the memory unit 12 may be another memory device connected to the server 10, or another memory device with which the server 10 can communicate.

[0013] The communication unit 13 is a communication module for performing processes related to wired or wireless communication, and transmits and receives information to and from other devices via the network N. The network N may be the Internet or a public telephone network, or may be a LAN (Local Area Network) established within the facility where the server 10 is installed. The input unit 14 accepts operation input by a user and sends a control signal corresponding to the operation content to the control unit 11. The display unit 15 is a liquid crystal display, an organic EL display, or the like, and displays various information according to instructions from the control unit 11. The input unit 14 and the display unit 15 may be a touch panel integrally configured.

[0014] The reading unit 16 reads information stored in a portable storage medium 10a, such as a CD (Compact Disc), a DVD (Digital Versatile Disc), a USB (Universal Serial Bus) memory, an SD (Secure Digital) card, or a CompactFlash (registered trademark). The program 12P and various data stored in the storage unit 12 may be read by the control unit 11 from the portable storage medium 10a via the reading unit 16 and stored in the storage unit 12. The program 12P and various data may be written to the storage unit 12 during the manufacturing stage of the server 10, or may be downloaded by the control unit 11 from another device via the communication unit 13 and stored in the storage unit 12.

[0015] In this embodiment, the server 10 may be a multi-computer consisting of multiple computers, or may be a virtual machine virtually constructed by software within a single device. Furthermore, when the server 10 is configured as a server computer, it may be a local server installed in the facility where the server 10 is located, or a cloud server connected to the facility via a network N. The following description assumes that the server 10 is a single computer. The program 12P may be deployed on a single computer or at a single site, or may be distributed across multiple sites and executed on multiple computers interconnected via the network N. Furthermore, the input unit 14 and the display unit 15 are not essential to the server 10. The server 10 may be configured to accept operations via a connected computer or to output information to be displayed to an external display device.

[0016] The user terminal 20 is a general-purpose information processing device such as a personal computer, smartphone, or tablet terminal. The user terminal 20 includes a control unit 21, a memory unit 22, a communication unit 23, an input unit 24, a display unit 25, and a reading unit 26, which are connected via a bus. The control unit 21, the memory unit 22, the communication unit 23, the input unit 24, the display unit 25, and the reading unit 26 of the user terminal 20 each have the same configuration as the control unit 11, the memory unit 12, the communication unit 13, the input unit 14, the display unit 15, and the reading unit 26 of the server 10, and therefore detailed description of the configurations will be omitted. In addition to the program 22P (program product) executed by the control unit 21, the memory unit 22 of the user terminal 20 also stores an application program (hereinafter referred to as a healthcare app 22AP) for implementing a process of receiving tips tailored to the user's symptoms using a service provided by the server 10.

[0017] In the information processing system configured as described above, the user terminal 20 receives input of symptom information related to the user's symptoms and transmits the input symptom information to the server 10. The server 10 acquires the symptom information transmitted by the user terminal 20, identifies tips corresponding to the acquired symptom information, and transmits the tips to the user terminal 20. As a result, tips corresponding to the user's symptoms are transmitted from the server 10 to the user terminal 20 and provided to the user of the user terminal 20. The server 10 of this embodiment uses the learning model 12M when identifying tips from the user's symptom information.

[0018] FIG. 3A is an explanatory diagram showing an example of the record layout of the tips DB 12a, and FIG. 3B is an explanatory diagram showing an example of the record layout of the user DB 12b. The tips DB 12a is a database that stores information about tips prepared for each of multiple symptoms experienced by menopausal users. The tips DB 12a shown in FIG. 3A includes a symptom information column and a tip information column, and stores tip information about tips to be provided to users for each symptom in association with symptom information about each symptom. The symptom information includes categories into which each symptom is classified and items classified into each category. The tip information includes a tip ID assigned to each tip, the type of tip, and the tip content. The tip content includes, for example, information about diet, exercise, sleep, actions for stress reduction, usage of electronic devices such as smartphones or computers, use of medicines including herbal medicines, information about body care, information about skin care, information about lifestyle (living habits), information about inner health (inner care), etc. The tips DB 12a is not limited to the configuration shown in Fig. 3A. The types of categories, items, and tips stored in the tips DB 12a are not limited to the examples shown in Fig. 3A.

[0019] The user DB 12b is a database that stores information about users who have registered to use services provided by the server 10. The user DB 12b shown in FIG. 3B includes a user ID column, a name column, an age column, a medical history column, an input information column, etc., and stores information about users in association with identification information (user ID) uniquely assigned to each user. The name column, age column, and medical history column store the user's name, age, and medical history column. The user's name, age, and medical history column are input via the user terminal 20 when the user registers, for example, and are acquired by the server 10 from the user terminal 20 and stored in the user DB 12b. The input information column stores information about the user's symptoms, etc., input by the user via the user terminal 20. The input information includes a date, symptom information about the user's symptoms, and effort information about the behaviors the user is engaging in as part of their lifestyle habits. The symptom information includes any of the symptoms registered in the tips DB 12a in FIG. 3A , and if the level (severity) of each symptom is entered, information indicating the severity of each symptom is also included. The severity information may be, for example, mild, moderate, severe, or very severe. The activity information may include, for example, taking supplements such as vitamins or iron, performing exercises or massages such as yoga or pelvic floor muscle training, or taking medication for menopausal symptoms. The contents stored in the user DB 12b are not limited to the example shown in FIG. 3B , and may be configured without a medical history column, or may store, in addition to the medical history, a medical history or medication history.

[0020] FIG. 4A is an explanatory diagram showing an example configuration of the learning model 12M. The learning model 12M is trained to input symptom information regarding a predetermined number of symptoms experienced by a menopausal user, perform calculations to estimate tips to be provided to the user for each symptom based on the input symptom information, and output the calculation results. In the example shown in FIG. 4A, five pieces of symptom information (first symptom information to fifth symptom information) are input, and the learning model 12M is configured to output information regarding tips for each of the pre-prepared symptoms. The learning model 12M can be configured using algorithms such as a convolutional neural network (CNN), a decision tree, a random forest, a support vector machine (SVM), or a transformer, or it may be configured by combining multiple algorithms.

[0021] The learning model 12M shown in FIG. 4A has five input nodes, through which symptom information (identification information that can identify each symptom) indicating a symptom selected by the user is input. In this embodiment, the user is configured to select up to five symptoms, and up to five pieces of symptom information are input to the learning model 12M. The learning model 12M calculates output values ​​based on the input symptom information using various functions, thresholds, etc., and outputs the calculated output values. The learning model 12M shown in FIG. 4A has multiple output nodes, each of which is associated with a pre-prepared symptom, specifically, each symptom registered in the tips DB 12a. Each output node outputs information about tips to be provided to the user for the associated symptom (specifically, a tip ID). With this configuration, when a predetermined number of pieces of symptom information (e.g., five) are input, the learning model 12M outputs a tip ID of a tip to be provided to the user for each symptom.

[0022] The learning model 12M is generated by machine learning using training data that associates symptom information for training with tip IDs (correct labels) of tips to be provided to users, as determined by a doctor or other professional based on the symptom information. The training data is generated by assigning tip IDs of tips provided by a doctor or other professional to each patient for symptom information of each patient who is undergoing treatment for menopausal symptoms or who is aware of menopausal symptoms. Tips for each symptom are appropriately set depending on the severity of each symptom, the combination of symptoms that occur simultaneously, and the like. The learning model 12M learns to output information indicated by the correct labels (e.g., tip IDs) from output nodes corresponding to each symptom when symptom information included in the training data is input. During the learning process, the learning model 12M performs calculations based on the input symptom information and calculates output values ​​from each output node. The learning model 12M then compares the calculated output value of each output node with information corresponding to the correct label (the tip ID of the correct label for the output node corresponding to the symptom for which the correct label is set, and, for example, 0 for other output nodes), and optimizes parameters used in the calculation process so that the two values ​​approximate each other. For example, parameters such as weights (coupling coefficients) between neurons in the learning model 12M are optimized using backpropagation, steepest descent, or the like. As a result, when a predetermined number of pieces of symptom information are input, the learning model 12M outputs the tip ID of a tip to be provided to the user from the output node associated with the symptom of the input symptom information, and outputs, for example, 0 from the other output nodes. The server 10 identifies the output value corresponding to each symptom included in the input data from the output values ​​of each output node of the learning model 12M, and sets the identified output value as the tip ID of the tip corresponding to each symptom, thereby identifying the tip ID of a tip to be provided to the user for each symptom for which symptom information has been input.

[0023] 4B is an explanatory diagram showing an example of the configuration of learning model 12Ma, which is a modified example of learning model 12M. Learning model 12Ma has the same configuration as learning model 12M, but the symptom information, which is input data, includes information indicating the severity (degree) of each symptom in addition to information about a predetermined number of symptoms. The information indicating the severity is, for example, mild, moderate, severe, or very severe. Learning model 12Ma is trained to perform calculations to estimate tips to be provided to the user for each symptom based on the input symptom information (information about each symptom and information about the severity of each symptom) and output the calculation results.

[0024] The learning model 12Ma is generated by machine learning using training data that associates symptom information, including a predetermined number of symptoms and the severity of each symptom, with tip IDs (correct labels) of tips to be provided to the user for each symptom. The learning model 12Ma also learns to output information indicated by the correct label from the output node corresponding to each symptom when symptom information included in the training data is input (the tip ID of the correct label for the output node corresponding to the symptom for which the correct label is set, and, for example, 0 for other output nodes). This results in a learning model 12Ma that, when symptom information, including a predetermined number of symptoms and the severity of each symptom, estimates tips to be provided to the user for each symptom and outputs the estimation results.

[0025] The learning of the learning models 12M and 12Ma may be performed on the server 10 or on another learning device. The trained learning models 12M and 12Ma generated by training on another learning device are downloaded from the learning device to the server 10 via the network N or via the portable storage medium 10a and stored in the storage unit 12.

[0026] The learning models 12M and 12Ma are not limited to the configurations shown in FIGS. 4A and 4B . For example, instead of a configuration in which each output node outputs the tip ID of a tip estimated to be provided for a symptom associated with the output node (the tip ID of the tip with the highest degree of certainty), a configuration may be provided in which multiple output nodes are provided for each symptom, outputting a degree of certainty for each tip ID of a tip prepared for the symptom. Furthermore, the learning models 12M and 12Ma may be configured to input, for example, the user's age or the number of years since menopause in addition to the user's symptom information, or to input one or more of the user's medical history, medical treatment history, medication history, etc. In this case, the learning models 12M and 12Ma are configured to estimate tips to be provided to the user based on not only the symptom information but also various user information. Furthermore, the learning models 12M and 12Ma may be prepared for each attribute classified by age or the number of years since menopause.

[0027] The server 10 performs machine learning to learn predetermined training data to generate learning models 12M and 12Ma in advance. The server 10 then inputs the symptom information of the user acquired from the user terminal 20 into the learning models 12M and 12Ma, and identifies tips to be provided to the user based on the output information from the learning models 12M and 12Ma.

[0028] The following describes the process in the information processing system of this embodiment, in which the user terminal 20 accepts input of symptom information related to the user's symptoms, and tips corresponding to the input symptom information are identified by the server 10 and provided to the user. Figures 5 to 7 are flowcharts showing an example of the process for providing tips, and Figures 8A to 11B are explanatory diagrams showing example screens. In Figures 5 to 7, the left side shows the process performed by the user terminal 20, and the right side shows the process performed by the server 10.

[0029] When a user wants to obtain tips about menopausal symptoms, the user launches the healthcare app 22AP on the user terminal 20. The control unit 21 of the user terminal 20 executes the following process in accordance with the healthcare app 22AP. When the healthcare app 22AP is launched, the control unit 21 determines whether or not this is the first time the user has input symptom information about their symptoms (whether or not this is the first input) (S11). For example, the control unit 21 periodically accepts the user's input of symptom information via the input unit 24, such as once a day, and stores the input symptom information together with the input date and time in the healthcare app 22AP or the memory unit 22. Therefore, if symptom information is not stored in the healthcare app 22AP or the memory unit 22, the control unit 21 determines that this is the first input, and if symptom information is already stored, the control unit 21 determines that this is the second or subsequent input.

[0030] If it is determined that the input is the first time (S11: YES), the control unit 21 displays an initial input screen, such as that shown in FIG. 8A, on the display unit 25 (S12). In the case of initial input, the control unit 21 may first display an input screen (not shown) for personal information such as the user's name, age, medical history, treatment history, and medication history, accept the personal information via the input screen, and then display the initial input screen. When the control unit 21 accepts the personal information, it transmits the accepted personal information to the server 10 to register the user. When the control unit 11 of the server 10 receives personal information from the user terminal 20, it issues a user ID and stores (registers) the received personal information in the user DB 12b of the storage unit 12 in association with the issued user ID. The control unit 21 of the user terminal 20 may store the accepted personal information in the healthcare application 22AP or the storage unit 22.

[0031] The initial input screen shown in FIG. 8A includes a group of selection buttons that allow the user to select up to five symptoms of concern. The group of selection buttons includes icons (selection buttons) representing each of a plurality of symptoms (items) that appear during menopause, and the icons are displayed in four categories: body, delicate zone, mind, and sleep status (sleep state). The symptom categories and symptoms in each category displayed on the initial input screen are not limited to the example shown in FIG. 8A. The control unit 21 accepts the selection of up to five symptoms via the initial input screen (S13). Upon accepting the symptom selection, the control unit 21 displays the icon for the selected symptom in a different manner from the other icons. In FIG. 8A, four symptoms, hot flashes, headaches, anxiety, and night sweats, are selected, and the icons for each symptom are displayed in an inverted state. The initial input screen of this embodiment is configured to accept the selection of symptom type, but may also be configured to accept the severity of the selected symptom in addition to the symptom type. Specifically, when any symptom is selected on the initial input screen, the control unit 21 may be configured to display a severity input field R1 as shown in FIG. 9B and accept input of either mild, moderate, severe, or most severe for the selected symptom via the severity input field R1.

[0032] The initial input screen includes a Next button for requesting tips corresponding to the selected symptom. The control unit 21 determines whether the Next button has been pressed on the initial input screen (S14). If it determines that the Next button has not been pressed (S14: NO), the control unit 21 returns to step S13 and continues accepting symptom selections. If it determines that the Next button has been pressed (S14: YES), the control unit 21 stores symptom information indicating the selected symptom in association with the current date (or date and time) (S15). The control unit 21 stores the symptom information, for example, in the healthcare application 22AP or the storage unit 22. The control unit 21 then transmits the symptom information to the server 10 (S16) and requests tips corresponding to the symptom information from the server 10. The control unit 21 transmits the symptom information to the server 10 together with the user ID assigned to the user and the current date (or date and time). In addition, when the control unit 21 receives input of the symptom type and severity via the initial input screen, it stores the symptom information including the symptom type and severity in the healthcare app 22AP or the memory unit 22 and transmits it to the server 10.

[0033] When the control unit 11 of the server 10 receives symptom information from the user terminal 20 along with the user ID and date, the control unit 11 stores the received symptom information in the user DB 12b in association with the received user ID and date (S17). The control unit 11 identifies tips corresponding to each symptom selected by the user based on the symptom information acquired from the user terminal 20 (S18). Here, the control unit 11 inputs the symptom information acquired from the user terminal 20 (e.g., identification information for each symptom) into the learning model 12M and acquires information on tips corresponding to each symptom from each output node of the learning model 12M. Specifically, the control unit 11 acquires tip IDs from output nodes associated with the symptoms in the input symptom information and acquires, for example, 0 from other output nodes. Furthermore, if the user selects fewer than five symptoms, the control unit 11 inputs information on each symptom (e.g., identification information) from input nodes equal to the number of selected symptoms, acquires tip IDs corresponding to each symptom from output nodes associated with the selected symptoms, and acquires, for example, 0 from other output nodes. Furthermore, if the symptom information includes the severity of each symptom, the control unit 11 inputs symptom information indicating the type and severity of each symptom to each input node of the learning model 12Ma shown in Fig. 4B and obtains a tips ID corresponding to each symptom from each output node. Here, too, the control unit 11 obtains tips IDs from output nodes associated with the symptoms in the input symptom information and obtains, for example, 0 from other output nodes.

[0034] After identifying tips corresponding to each symptom, the control unit 11 generates a tips screen as shown in FIG. 8B (S19). Here, the control unit 11 reads tips corresponding to the tips IDs for each symptom acquired in step S18 from the tips DB 12a and generates a tips screen displaying the read tips. This generates a tips screen displaying tips related to each symptom selected by the user via the initial input screen. The control unit 11 transmits the generated tips screen to the user terminal 20 (S20), and the control unit 21 of the user terminal 20 receives the tips screen from the server 10 and displays it on the display unit 25 (S21).

[0035] On the other hand, if the control unit 21 determines that this is not the first input (S11: NO), it displays a top screen as shown in Fig. 9A on the display unit 25 (S22). The top screen displays the current date and has an entry button for instructing execution of a process for inputting the user's symptoms at that time. The bottom of the top screen is provided with a TIPS button for instructing the user to view tips corresponding to symptom information already entered by the user, a HISTORY button for instructing the user to view symptom information previously entered by the user (input history), and a RECORD button for instructing the user to register (record) new symptom information.

[0036] The control unit 21 determines whether the entry button on the top screen has been operated (S23). If it determines that the entry button has not been operated (S23: NO), it determines whether, for example, the tips button has been operated (S24). If it determines that the tips button has not been operated (S24: NO), the control unit 21 returns to step S23 and waits until the entry button or the tips button is operated. Note that, if the history button on the top screen is operated, the control unit 21 displays a history screen (see FIG. 11B ) on the display unit 25 based on previously input information (symptom information and effort information). The previously input information may be stored, for example, in the healthcare app 22AP or the memory unit 22, or the control unit 21 may acquire input information stored in the user DB 12b of the server 10. Note that, if the record button on the top screen is operated, the control unit 21 performs the same processing as when the entry button is operated.

[0037] If it is determined that the entry button has been operated (S23: YES), the control unit 21 displays a symptom input screen as shown in FIG. 9B on the display unit 25 (S25). The symptom input screen shown in FIG. 9B has a group of selection buttons similar to those of the initial input screen shown in FIG. 8A. The selection buttons on the screen of FIG. 9B include icons indicating symptoms for five categories: body, delicate areas, mind, sleep status, and bleeding. The selection buttons on the screen of FIG. 9B also include an icon indicating no symptoms for each category, allowing the user to input the absence of symptoms for each category. The screen of FIG. 9B has a symptom button for displaying the symptom input screen shown in FIG. 9B and an effort button for displaying the effort input screen shown in FIG. 10A. When the effort button is operated, the effort input screen shown in FIG. 10A is displayed. The effort input screen is a screen that accepts input of information regarding the user's efforts and has a group of selection buttons including icons (selection buttons) indicating each of multiple items that the user may perform. The selection buttons on the effort input screen include icons indicating each item such as the nutrients the user is taking, the exercise the user is performing, the medications the user is taking, and the treatments the user is receiving, and the icons are displayed in four categories: nutrition, lifestyle, therapy (medication), and therapy (non-medication). The items and categories of each item displayed on the symptom input screen and the effort input screen are not limited to the examples in FIGS. 9B and 10A.

[0038] The control unit 21 determines whether or not the selection of any symptom has been accepted on the symptom input screen shown in Fig. 9B (S26). If it is determined that a selection has been made (S26: YES), the control unit 21 displays a severity input field R1 on the symptom input screen as shown in Fig. 9B (S27). Then, the control unit 21 accepts an input of the severity of the selected symptom via the severity input field R1 (S28). When the control unit 21 accepts the input of the severity, it displays the icon of the selected symptom, for example, by highlighting it. Note that the control unit 21 may also display the input severity in association with the icon, in which case the user can confirm the severity of each symptom they have entered.

[0039] The symptom input screen of Fig. 9B and the effort input screen of Fig. 10A each include a SAVE button for instructing the user to save the input symptom information and effort information. The control unit 21 determines whether the SAVE button has been operated on the symptom input screen (S29). If it determines that the SAVE button has not been operated (S29: NO), the control unit 21 returns to step S26 and continues accepting input of the symptom type and severity. If it determines that the SAVE button has been operated (S29: YES), the control unit 21 transmits symptom information indicating the symptom type and severity entered via the symptom input screen to the server 10 (S30) and requests the server 10 to save the symptom information. Here, too, the control unit 21 transmits the symptom information together with the user ID and the current date (or time and date) to the server 10, and the control unit 11 of the server 10 stores the symptom information in the user DB 12b in association with the received user ID and date (S31). The control unit 21 may store the symptom information received via the symptom input screen in the healthcare application 22AP or the storage unit 22 in association with the current date (or date and time).

[0040] If the control unit 21 determines that a symptom selection has not been accepted on the symptom input screen of FIG. 9B ( S26: NO), it determines whether the action button has been operated ( S32 ). If it determines that the action button has not been operated ( S32: NO), the control unit 21 returns to step S26 and waits until a symptom selection is accepted or the action button is operated. If it determines that the action button has been operated ( S32: YES), the control unit 21 displays an action input screen as shown in FIG. 10A on the display unit 25 ( S33 ) and accepts the input of action information via the action input screen ( S34 ). The action input screen shown in FIG. 10A includes icons representing the categories of nutrients (nutrients) the user is taking through supplements, etc., including B vitamins, vitamin C, vitamin E, magnesium, isoflavone / equol, iron, and multivitamins, as well as an Add button. In addition to the nutrients mentioned above, the nutrition category includes other nutritional and dietary related items 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, low salt, etc., and the icon for each item can be switched between visible and hidden. For example, when the Add button is operated, the control unit 21 displays a selection screen (not shown) for instructing whether to display or hide each of the above-mentioned items, and accepts the display / hide selection for each item via the selection screen. The control unit 21 displays the icon of the selected item in the nutrition category display field on the activity input screen. The icon for each item is provided in the healthcare app 22AP.

[0041] The lifestyle category includes icons representing yoga, Pilates, pelvic floor muscle training, foot baths, and foot massage, as well as an add button. In addition to the above-mentioned items, lifestyle category items include exercise and lifestyle habits such as increasing daily activity, stretching, listening to the radio, walking, jogging, swimming / water training, tai chi, muscle training, bathing, mindfulness / meditation, aromatherapy (aromatherapy, ablution, etc.), Epsom salt baths, skin moisturizing, digital detox, morning sun exposure, and sleep hygiene. The icons for each item can be toggled on and off. Therefore, when the lifestyle category add button is operated, the control unit 21 displays a selection screen (not shown) for instructing whether to display or hide each of the above-mentioned items and accepts the display / hide selection for each item. The control unit 21 displays the icon for the selected item in the lifestyle category display field of the activity input screen. Again, the icons for each lifestyle item are provided in the healthcare app 22AP.

[0042] The Therapy (Medical) category includes icons for HRT (Hormone Replacement Therapy), hormones other than HRT, and herbal medicines, as well as an Add button. In addition to the above-mentioned items, the Therapy (Medical) category also includes items related to medication administered at medical institutions, such as intravaginal estrogen administration and overactive bladder medication, and the icons for each item can be switched between visible and hidden. Therefore, when the Add button for the Therapy (Medical) category is operated, the control unit 21 displays a selection screen (not shown) for instructing whether to display or hide each of the above-mentioned items and accepts the display / hide selection for each item. The control unit 21 displays the icon of the selected item in the Therapy (Medical) display field on the effort input screen. Again, the icons for each Therapy (Medical) item are provided in the healthcare app 22AP.

[0043] The Therapy (Non-Medicine) category includes icons for counseling, acupuncture, and chiropractic / massage, as well as an Add button. In addition to the above-mentioned items, the Therapy (Non-Medicine) category also includes items related to non-medication treatments performed at medical institutions, such as cognitive behavioral therapy, reflexology, intravaginal laser irradiation, and aromatherapy (treatment method), and the icons for each item can be switched between visible and hidden. Therefore, when the Add button for the Therapy (Non-Medicine) category is operated, the control unit 21 displays a selection screen (not shown) for instructing whether to display or hide each of the above-mentioned items and accepts the display / hide selection for each item. The control unit 21 displays the icon of the selected item in the Therapy (Non-Medicine) display field on the effort input screen. Again, the icons for each Therapy (Non-Medicine) item are provided in the healthcare application 22AP.

[0044] The control unit 21 determines whether the save button on the effort input screen has been operated (S35). If it determines that the save button has not been operated (S35: NO), the control unit 21 returns to step S34 and continues accepting the input of effort information. If it determines that the save button has been operated (S35: YES), the control unit 21 transmits effort information indicating the efforts entered via the effort input screen to the server 10 (S36) and requests the server 10 to save the effort information. Here, the control unit 21 also transmits the effort information together with the user ID and the current date (or time and date) to the server 10, and the control unit 11 of the server 10 stores the effort information in the user DB 12b in association with the received user ID and date (S37). Note that the control unit 21 may store the effort information received via the effort input screen in the healthcare app 22AP or the memory unit 22 in association with the current date (or time and date).

[0045] After processing step S30 or step S36, the control unit 21 generates a recording completion screen notifying the user that recording of the symptom information or effort information has been completed (S38). FIG. 10B shows an example of the recording completion screen. For example, the control unit 21 compares the symptom information input chronologically over the most recent predetermined period (e.g., one week, one month, etc.) with the symptom information input via the symptom input screen (the most recent symptom information), identifies changes in symptoms, such as whether or not a new symptom has appeared, whether or not a symptom has disappeared or alleviated, or whether or not a symptom has remained unchanged, and generates a message corresponding to the symptom change. For example, when the control unit 21 identifies a newly appeared symptom, it generates a message such as "It appears that new symptoms, such as headache and hot flashes, have appeared," as shown on the screen in FIG. 10B. When the control unit 21 identifies a symptom that has disappeared or alleviated, it generates a message such as "It appears that your headache symptoms have alleviated," and when the control unit 21 identifies a symptom that has remained unchanged, it generates a message such as "It appears that your headache symptoms are continuing." Furthermore, the healthcare application 22AP has prepared a plurality of messages to be provided to the menopausal user, and the control unit 21 selects an arbitrary message randomly or in accordance with changes in the user's symptoms (whether or not new symptoms have appeared, whether or not symptoms have disappeared or been alleviated, whether or not or not symptoms have not changed, etc.). The control unit 21 then generates a recording completion screen displaying the generated message. Note that, on the screen of FIG. 10B , the message selected randomly or in accordance with changes in the user's symptoms is displayed as "Today's Message."

[0046] In addition, the control unit 21 may compare the effort information input in chronological order over the most recent specified period with the effort information input via the effort input screen (latest effort information), identify changes in the effort content, such as whether or not an effort has been started for the first time, whether or not an effort has been stopped or interrupted, or whether or not an effort is still being carried out, and generate a message according to the change in the effort content and display it on the recording completion screen.

[0047] 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 the control unit 21 determines that the OK button has been operated (S40: YES), it returns to step S22 and displays the top screen shown in Figure 9A on the display unit 25 (S22). If the control unit 21 determines that the tips button on the top screen has been operated (S24: YES), it transmits the user's user ID to the server 10 and requests tips corresponding to the user's symptom information from the server 10 (S41).

[0048] When the control unit 11 of the server 10 receives a tips request, it reads the user's symptom information from the user DB 12b (S42). For example, the control unit 11 reads symptom information input over the past month (e.g., chronological symptom information input periodically, such as once a day). Note that the symptom information to be read is not limited to information from the past month, but may be information from the past week, two weeks, or two months. Based on the read symptom information, the control unit 11 counts the frequency of occurrence (occurrence) of each symptom and identifies five symptoms in descending order of frequency of occurrence (S43). Then, based on the symptom information of the identified five symptoms, the control unit 11 identifies tips corresponding to each of the five symptoms (S44). Step S44 is the same process as step S18. The control unit 11 inputs symptom information of the identified symptoms (e.g., identification information for each symptom) to the input node of the learning model 12M and obtains tip IDs from the output nodes associated with the symptoms in the input symptom information. Here too, when the symptom information includes the severity of each symptom, the control unit 11 inputs the identification information and severity of each symptom to the input node of the learning model 12Ma, and acquires a tips ID corresponding to each symptom from the output node associated with the symptom in the input symptom information. When inputting the severity of each symptom, the control unit 11 may input the latest severity of the identified symptom to the learning model 12Ma, or may calculate the average severity over a predetermined period and input it to the learning model 12Ma.

[0049] Based on the tips ID for each symptom identified in step S44, the control unit 11 reads the tips corresponding to the tips ID from the tips DB 12a and generates a tips screen as shown in Fig. 11A (S45). The tips screen in Fig. 11A has the same configuration as the tips screen in Fig. 8B, but presents tips that take into account the user's symptoms input during a specified period. The control unit 11 transmits the generated tips screen to the user terminal 20 (S46), and the control unit 21 of the user terminal 20 receives the tips screen from the server 10 and displays it on the display unit 25 (S47).

[0050] 11B shows an example of a history screen displayed when the history button on the screen is operated. For example, when the history button is operated on the top screen, the control unit 21 displays a history screen such as that shown in FIG. 11B on the display unit 25. Specifically, the control unit 21 acquires symptom information stored in the healthcare app 22AP or the memory unit 22, or symptom information stored in the user DB 12b of the server 10. For example, the control unit 21 acquires symptom information for the most recent two months. Then, for the two months of symptom information, the control unit 21 counts the occurrence frequency of each symptom in the first month based on the symptom information for the first month, and counts the occurrence frequency of each symptom in the second month based on the symptom information for the second month. The control unit 21 also identifies symptoms with a high occurrence frequency in the second month and counts the occurrence frequency of each symptom for each week in the second month. Furthermore, the control unit 21 compares the occurrence frequency of each symptom in the first month with the occurrence frequency of each symptom in the second month, determines whether the occurrence frequency is increasing, decreasing, or remaining the same, and identifies changes in each symptom. The control unit 21 calculates, for example, the rate of increase or decrease in the occurrence frequency as the change in each symptom. The control unit 21 generates a history screen displaying the information acquired as described above. This allows the control unit 21 to generate a history screen such as that shown in FIG. 11B. The screen in FIG. 11B displays symptoms and their occurrence frequencies that occur frequently in the second month (e.g., this month), displays messages regarding symptoms that have changed compared to the first month (e.g., last month), and further displays a graph showing the occurrence frequency of each symptom for each week in the second month.

[0051] The control unit 21 may identify the change in each symptom from last month to this month by considering not only the occurrence frequency (selected frequency) of each symptom but also the severity level entered for each symptom. For example, the control unit 21 may calculate an occurrence score for each symptom by assigning a higher weight to each occurrence of the symptom with a higher severity, compare the occurrence scores for each symptom in the first month and the second month, and calculate the rate of increase or decrease in the occurrence score to identify the change in each symptom. The screen in FIG. 11B shows the history screen for this month. When a predetermined operation (e.g., a swipe to the right) is performed, the screen may be switched to display the history screen for the previous month (last month). Furthermore, when a predetermined operation (e.g., a swipe to the left) is performed, the screen may be switched to display the history screen for the month following the currently displayed history screen.

[0052] 11B also includes a "This Month" button for displaying the current month's history screen and a "This Year" button for displaying the current year's history screen. When the "This Year" button is pressed, the control unit 21 generates and displays a history screen based on the current year's symptom information. Specifically, the control unit 21 counts the occurrence frequency of each symptom for the current year based on the current year's symptom information, identifies frequently occurring symptoms, and further counts the occurrence frequency of each symptom for each month of the current year. Furthermore, when the control unit 21 stores symptom information for the past year, it compares the occurrence frequency of each symptom for the past year with the occurrence frequency of each symptom for the current year and identifies changes in the occurrence frequency of each symptom. This allows the control unit 21 to display frequently occurring symptoms and their occurrence frequencies for the current year, display messages regarding symptoms that have changed compared to the past year, and generate a history screen (not shown) displaying a graph showing the occurrence frequency of each symptom for each month of the current year.

[0053] By the above-described process, in this embodiment, when a user inputs symptom information for the first time, tips corresponding to the input symptoms can be provided to the user. Also, when the user continues to input symptom information periodically for a predetermined period (e.g., one month, one week, etc.), tips corresponding to frequently occurring symptoms can be provided to the user. Thus, the user can acquire knowledge about symptoms that concern them and receive advice for alleviating the symptoms.

[0054] This embodiment is configured to identify frequently occurring symptoms from symptom information input by a user over a predetermined period via a symptom input screen, and to identify tips to be provided to the user based on the frequently occurring symptoms. Alternatively, tips to be provided to the user may be identified based on time-series symptom information input over a predetermined period. For example, the learning models 12M and 12Ma may be configured to receive time-series symptom information and output tip IDs for tips corresponding to each symptom based on the input time-series symptom information. Furthermore, a smartwatch, for example, may be configured to link with a measuring device that measures the user's biological information, such as body temperature, blood pressure, heart rate, pulse rate, and blood glucose level, and to identify tips to be provided to the user based on the symptom information input by the user via the initial input screen and the symptom input screen and the user's biological information measured by the measuring device. In this case, the learning models 12M and 12Ma may be configured to receive symptom information and biological information input, and to output tip IDs for tips corresponding to each symptom based on the input symptom information and biological information.

[0055] In this embodiment, the process of identifying tips based on symptom information entered 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 memory unit 22 of the user terminal 20, the control unit 21 can input symptom information received via the input unit 24 into the learning models 12M and 12Ma and identify tips 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 a tips screen such as those shown in Figures 8B and 11A based on the identified tips and display it on the display unit 25. Even with this configuration, the same processing as in this embodiment is possible and the same effects can be obtained.

[0056] In this embodiment, the process of identifying tips for each symptom in step S18 in FIG. 5 and step S44 in FIG. 7 is not limited to a configuration using the learning models 12M and 12Ma, but may be a rule-based configuration. For example, a database in which tips IDs for tips to be provided to the user are associated with each of multiple sets including a predetermined number of symptoms and their severity levels can be used. In this case, the control unit 11 identifies the tips IDs for each symptom corresponding to the symptom information entered by the user from the database, and identifies the tips associated with the identified tips IDs as the tips for each symptom to be provided to the user. Alternatively, for example, the control unit 11 may assign an importance (priority) to each tip prepared for each symptom, calculate a score for each symptom based on a combination of a predetermined number of symptoms selected by the user and the severity level of each symptom, and identify tips with an importance corresponding to the calculated score as the tips to be provided to the user for each symptom. The score for each symptom here can be calculated based on, for example, a score set for each symptom, a score set for a combination of symptoms, and a score set for the severity of each symptom. Furthermore, the control unit 11 may be configured to identify a predetermined number of symptoms for which tips should be provided to the user based on the time-series symptom information input by the user, and then identify tips for the identified symptoms based on the input time-series symptom information. Even with this configuration, it is possible to perform the same processing as in this embodiment, and to obtain the same effects.

[0057] (Embodiment 2) An information processing system will be described that uses image processing to determine the test results of a test kit that tests a user's hormone levels and provides tips based on the determination results. The information processing system of this embodiment can be realized using devices similar to those of the information processing system of embodiment 1 shown in Figures 1 and 2, so a description of the configuration of each device will be omitted. In addition to the configuration shown in Figure 2, the user terminal 20 of this embodiment has a camera (not shown) and stores a test determination model 22M (see Figure 12A, a learning model for determining test results) in the memory unit 22.

[0058] The camera is an imaging device having a lens, an imaging element, etc. In this embodiment, the camera captures one image (still image) in response to a single operation of the capture button. However, the camera may also be configured to capture, for example, 15 or 30 frames of images (video) per second in response to an operation of the capture button. The inspection judgment model 22M may be incorporated into the healthcare app 22AP and stored in the storage unit 22, or may be stored in the storage unit 22 separately from the healthcare app 22AP. The inspection judgment model 22M is a model that has learned training data, for example, through machine learning, and is intended for use as a program module constituting artificial intelligence software. Instead of being stored in the user terminal 20, the inspection judgment model 22M may be stored in the storage unit 12 of the server 10 or in a server other than the server 10. In this case, the user terminal 20 may be configured to access the server 10 or another server that stores the inspection judgment model 22M and read the inspection judgment model 22M, or a judgment process using the inspection judgment model 22M may be performed in the server 10 or another server.

[0059] FIG. 12A is an explanatory diagram showing an example configuration of the test determination model 22M. The test determination model 22M shown in FIG. 12A inputs a captured image of a test kit measuring hormone levels (specifically, a captured image of the test kit's test line), performs a calculation to estimate the test results of the test kit based on the input captured image, and outputs the calculation result. The test kit may be, for example, a urine test kit that can be performed at home, and may be a test kit capable of measuring the presence and amount of a target hormone. The test kit indicates the presence or absence of a hormone by the presence or absence of a test line, and the amount of hormone is indicated by the thickness or color intensity of the test line. The test kit may be capable of detecting female hormones, such as FSH (follicle-stimulating hormone) or LH (luteinizing hormone), and may measure one or more types of hormones. The test kit is not limited to urine tests and may also be a test kit that measures hormone levels using saliva or blood. The test results output from the test determination model 22M indicate the amount of each hormone to be determined by the test kit. The inspection and judgment model 22M may be configured using an algorithm such as CNN, decision tree, random forest, SVM, or Transformer, or may be configured by combining multiple algorithms.

[0060] The test determination model 22M shown in Fig. 12A receives a photographed image of the test kit via an input node, calculates an output value based on the input photographed image using various functions and thresholds, etc., and outputs the calculated output value. The test determination model 22M shown in Fig. 12A has multiple output nodes, each of which is associated with one of multiple hormones to be tested, and outputs the measured amount of the associated hormone (the test result from the test kit) from each output node.

[0061] The test determination model 22M is generated by learning using training data that associates images captured using a training test kit with the measured amounts (correct measured amounts) of each hormone determined by a doctor or other doctor based on the test kit's test line. The training data is generated based on test results using the test kit for patients undergoing treatment for menopausal symptoms, patients who are aware of menopausal symptoms, and patients who do not have menopausal symptoms. The test determination model 22M learns to output the correct measured amount for each hormone when the captured images included in the training data are input. Specifically, the test determination model 22M performs calculations based on the input captured images, estimates the amount of each hormone based on the test kit's test line in the captured images, and obtains the estimated results. The test determination model 22M then compares the estimated amount of each hormone with the correct measured amount and optimizes parameters such as the weights (coupling coefficients) between nodes in the test determination model 22M so that the estimated amount of each hormone approximates the correct measured amount. The parameter optimization method can be a steepest descent method, an error backpropagation method, or the like. This provides a test determination model 22M that outputs the measured amounts of each hormone, which are the test results obtained by the test kit, when a photographed image of the test kit is input.

[0062] The measured amounts of each hormone output by the test judgment model 22M shown in FIG. 12A are numerical values ​​expressed as continuous values, and the test judgment model 22M outputs one of the continuous values ​​as a so-called regression problem. However, it may be configured to treat it as a classification problem and determine one of selectable numerical values ​​for each measured amount of hormone. In this case, the test judgment model 22M is configured to determine the optimal value as the measured amount of each hormone from multiple preset numerical values ​​and output the determination result. The test judgment model 22M may also be configured to determine whether the amount of each hormone is within the normal range based on the input photographed image and output the determination result. The test judgment model 22M may be trained on the server 10 or on another learning device. The trained test judgment model 22M generated by training on the server 10 or another learning device is incorporated into, for example, the Elscare app 22Ap, downloaded to the user terminal 20, and stored in the memory unit 22.

[0063] FIG. 12B is an explanatory diagram showing an example of the configuration of the learning model 12Mb of embodiment 2. The learning model 12Mb of this embodiment has a configuration similar to that of the learning model 12M of embodiment 1 shown in FIG. 4A, but in addition to symptom information for each symptom, hormone test results are input as input data. The hormone test results may be the measured amounts of each hormone estimated from images captured by a test kit using the test determination model 22M, or may be test results (e.g., check results for the phase or severity of menopausal symptoms) identified based on the measured amounts of each hormone and questionnaire responses, as described below. Furthermore, the learning model 12Mb of this embodiment may be configured to input symptom information including the severity of each symptom, similar to the learning model 12Ma shown in FIG. 4B.

[0064] Figure 13 is a flowchart showing an example of the procedure for providing tips in embodiment 2, and Figure 14 is an explanatory diagram showing an example screen. The process shown in Figure 13 is the process shown in Figures 5 to 7, with steps S51 to S60 added between steps S25 and S26. Explanations of steps that are the same as those in Figures 5 to 7 will be omitted. Also, in Figure 13, steps other than steps S25 to S26 in Figures 5 to 7 are omitted.

[0065] In this embodiment, the control unit 21 of the user terminal 20 proceeds to step S51 after processing step S25. Note that the symptom input screen displayed by the control unit 21 in step S25 includes, in addition to the configuration shown in FIG. 9B, a "Load Test Kit" button (see B1 in FIG. 15) for instructing the loading of a captured image of a test kit. Therefore, the control unit 21 determines whether the "Load Test Kit" button has been operated on the symptom input screen (S51), and if it determines that it has not been operated (S51: NO), proceeds to step S26. In this case, the control unit 21 executes the same processing as in embodiment 1.

[0066] When it is determined that the "Read Test Kit" button has been operated (S51: YES), the control unit 21 activates the camera (S52) and starts displaying an image acquired via the camera on the display unit 25. The user checks the displayed image and, for example, takes a still image with the entire test kit within the shooting range. When the control unit 21 receives an instruction to take a photo from the user via the input unit 24, it takes a photo of the test kit (S53) and acquires the captured image of the test kit. The control unit 21 stores the acquired captured image in the memory unit 22. The control unit 21 estimates the test results of the test kit based on the acquired captured image (S54). Here, the control unit 21 inputs the captured image of the test kit into the test determination model 22M and acquires the measured amount of each hormone as an output value from the test determination model 22M.

[0067] Next, the control unit 21 displays a questionnaire screen as shown in FIG. 14 on the display unit 25 (S55) and accepts responses to each questionnaire via the questionnaire screen (S56). The control unit 21 determines whether the OK button on the questionnaire screen has been pressed (S57). If the control unit 21 determines that the OK button has not been pressed (S57: NO), the process returns to step S56 and continues accepting questionnaire responses. If the control unit 21 determines that the OK button has been pressed (S57: YES), the control unit 21 identifies hormone test results based on the test results estimated in step S54 and the questionnaire responses acquired in step S56 (S58). The hormone test results may be measured amounts of each hormone obtained by correcting the test results obtained using a test kit based on the questionnaire responses, or may be the phase or severity of menopausal symptoms determined based on the test results obtained using a test kit and the questionnaire responses. The control unit 21 stores the identified hormone test results in the memory unit 22.

[0068] The determination of hormone test results may be performed using a rule-based configuration or a learning model. For example, a database may be used in which hormone test results indicating the state of each hormone are registered in association with each of multiple sets including the measured amount of each hormone and each questionnaire response. In this case, the control unit 21 can identify the hormone test results corresponding to each measured amount of hormone and each questionnaire response from the database. An importance (priority) may be assigned to each hormone or each questionnaire, and the user's hormone test score may be calculated taking into account the weighting according to the importance, and the hormone test results may be determined according to the calculated score. Alternatively, the hormone test results may be determined using a learning model that inputs the measured amount of each hormone and each questionnaire response and estimates and outputs the user's hormone test results based on the input data. For example, the test determination model 22M in FIG. 12A may be configured to input the captured image of the test kit and each questionnaire response and output the hormone test results.

[0069] The control unit 21 transmits the identified hormone test results to the server 10 (S59). The control unit 21 transmits the hormone test results to the server 10, for example, along with the user ID and the current date (or time and date). The control unit 11 of the server 10 stores the hormone test results in the user DB 12b, for example, in association with the received user ID and date (S60). The control unit 21 may also store the hormone test results in the healthcare app 22AP or the memory unit 22 in association with the current date (or time and date). In the above-described process, the hormone test results are identified from the measured amounts of each hormone estimated from the captured image of the test kit and the responses to the questionnaire. However, the measured amounts of each hormone may be used directly as the hormone test results. In this case, the control unit 21 may transmit the test results estimated in step S54 and the questionnaire responses received in step S56 directly to the server 10 as the hormone test results. Alternatively, the control unit 21 may transmit the test results estimated in step S54 directly to the server 10 without performing steps S55 to S57.

[0070] After processing step S59, the control unit 21 proceeds to step S26. This accepts input of symptoms, their severity, and effort information, and stores the accepted symptom information and effort information in the user DB 12b of the server 10. In the server 10 of this embodiment, in step S44, the control unit 11 inputs the hormone test results stored in the user DB 12b in step S60 into the learning model 12Mb in addition to the symptom information of the symptoms identified in step S43, and acquires a tips ID for each symptom as output data from the learning model 12Mb, thereby identifying tips corresponding to each symptom. If symptom information and hormone test results acquired in chronological order are stored in the user DB 12b, the time-series data of the symptom information and hormone test results is input into the learning model 12Mb. In addition, the learning model 12Mb in Figure 12B may be configured to input the measured amount of each hormone and each questionnaire response in addition to the symptom information of each symptom.In this case, the control unit 11 may input the measured amount of each hormone determined using the test determination model 22M and each questionnaire response into the learning model 12Mb in addition to the symptom information of the symptom identified in step S43, to identify tips corresponding to each symptom.

[0071] This embodiment also provides the same effects as the above-described embodiment 1. Furthermore, this embodiment can provide the user with tips that take into account the test results from the test kit in addition to the information about each symptom entered by the user. This makes it possible to provide tips that are more suited to the user's condition.

[0072] (Embodiment 3) A modified example of the display screen of the user terminal 20 in the information processing systems of Embodiments 1 and 2 will be described. FIG. 15 is an explanatory diagram showing a modified example of the symptom input screen, which is a modified example of the symptom input screen shown in FIG. 9B. The screen of FIG. 15 includes a Symptoms button, a Causes button, an Action button, and a Memo button. FIG. 15 shows the state in which the Symptoms button is selected. The screen of FIG. 15 also includes icons indicating four options for mood and bleeding, as well as a "Load Test Kit" button B1. Thus, as in Embodiment 2, the screen of FIG. 15 also allows users to acquire photographed images of the test kit, estimate hormone test results from the photographed images, and register them in the server 10. The screen of FIG. 15 also includes input buttons for entering "no symptoms" or any severity level for each symptom in multiple categories, including body, intimate area, mind, and sleep status. The screen of FIG. 15 allows users to enter the status of each symptom in each category with a single operation (e.g., by tapping the input button), thereby improving operability compared to the screen of FIG. 9B. When the effort button is operated on the screen of FIG. 15, the user terminal 20 displays an effort input screen similar to that of FIG. 10A.

[0073] When the factor button is operated on the screen of FIG. 15 , the user terminal 20 displays a factor input screen (not shown). The factor input screen is a screen for accepting input of factors related to the user's current state, and is provided with input buttons for entering the user's worries or stresses, such as work content, workplace relationships, housework, relationships with family, diet, sleep, etc. The control unit 21 of the user terminal 20 accepts the input of factors via this screen, transmits the accepted factors to the server 10, and stores them in the user DB 12b. Furthermore, when the memo button is operated on the screen of FIG. 15 , the user terminal 20 displays a memo input screen (not shown). The memo input screen is a screen for accepting input of memos, diary entries, etc., and accepts input of text data via the input unit 24, for example. The control unit 21 of the user terminal 20 transmits the memos, diary entries, etc. accepted via this screen to the server 10 and stores them in the user DB 12b. The factors, memos, etc. stored in the user DB 12b in this manner may be used by the server 10 when identifying tips. For example, the learning model 12Mb shown in Figure 12B may be configured to input factors and notes in addition to the hormone test results and symptom information for each symptom.In this case, tips for each symptom may be identified by reading the factors and notes along with the symptom information and hormone test results from the user DB 12b and inputting them into the learning model 12Mb.

[0074] FIG. 16 is an explanatory diagram showing a modified example of the history screen, which is a modified example of the history screen shown in FIG. 11B . The screen of FIG. 16 displays the five most frequently occurring symptoms among those that occurred during a predetermined period (one month in FIG. 16 ), time-series data on hormone test results (measured amounts of each hormone), and the five most frequently entered factors among those that were input during the predetermined period. If multiple hormone test results are stored in the user DB 12b, the time-series data on hormone test results is displayed as a graph showing changes in the multiple hormone test results. The screen of FIG. 16 also displays a message regarding symptoms that have changed by comparing, for example, symptoms and their frequency of occurrence this month with symptoms and their frequency of occurrence last month. The screen of FIG. 16 allows the user to easily identify frequently occurring symptoms and factors, and intuitively grasp the progress of hormone test results.

[0075] In the information processing system of this embodiment, the symptom input screen and history screen are different from those of embodiments 1 and 2, but the same processes as those of embodiments 1 and 2 can be executed and similar effects can be obtained. In this embodiment, too, it is possible to select whether to display or not each item displayed on the screens shown in Figures 15 and 16, and operability can be improved by customizing the display items to the user's preferences.

[0076] (Embodiment 4) An information processing system that generates tips to be provided to a user using a language model will be described. The information processing system of this embodiment can be realized by devices similar to those of the information processing system of embodiment 1 shown in Figures 1 and 2, so a description of the configuration of each device will be omitted. Note that the server 10 of this embodiment stores a language model (not shown) in the storage unit 12 in addition to the configuration shown in Figure 2. Furthermore, in this embodiment, in addition to symptom information and effort information as input information, the user DB 12b stored in the storage unit 12 stores (accumulates) cause information and memo information input via the symptom input screen as in embodiment 2.

[0077] The language model is a general-purpose large-scale language model (LLM) constructed by pre-training using a large set of sentences, etc. The language model is trained to perform calculations based on input data, such as text data written in natural language, to generate output data corresponding to the content of the input data and output the generated output data. The language model can be constructed using algorithms such as GPT (Generative Pre-trained Transformer)-3, GPT-3.5, and GPT-4, or can be constructed by combining multiple algorithms. Note that the language model is not limited to the transformer-based model described above. The language model is expected to be used as a program module constituting artificial intelligence software. Instead of storing the language model in the memory unit 12, the server 10 may access a language processing server that stores the language model and read it out. The language model may also be fine-tuned using the user's symptom information, hormone test results, tips to be provided to the user, etc. In addition to fine-tuning the entire language model, the parameters of only some layers may be fine-tuned using techniques such as LoRA (Low-Rank Adaptation).

[0078] 17A and 17B are explanatory diagrams showing example prompts. FIG. 17A is an example of a prompt for instructing a language model to generate tips to be provided to a user when symptom information is input for the first time via the initial input screen shown in FIG. 8A , and FIG. 17B is an example of a prompt for instructing a language model to generate tips to be provided to a user when symptom information is input for the first time via the symptom input screen shown in FIG. 9B . The example of FIG. 17A includes a command statement for a process to be executed and a sentence regarding symptom information as information to be referenced when generating the tips. The command statement may include a sentence specifying the content of the tips, a sentence specifying matters to be considered when generating the tips, a sentence specifying the output format, etc. The sentence specifying the output format may include, for example, a sentence such as, "Please create a one-week diet and exercise plan. In that case, please list the diet and exercise to be performed for each day of the week from Monday to Sunday." A prompt template containing only the command statement for the prompt in Figure 17A is stored in the memory unit 12, and the control unit 11 of the server 10 creates a prompt such as that shown in Figure 17A by reading symptom information of the user for whom tips are to be generated from the user DB 12b and inserting it into this prompt template.

[0079] The example in Fig. 17B includes a command statement for the process to be executed and sentences related to symptom information, cause information, effort information, and memo information as information to be referenced when generating tips. A prompt template containing only command statements for the prompt in Fig. 17B is stored in the storage unit 12. The control unit 11 of the server 10 reads the symptom information, cause information, effort information, and memo information of the user for whom tips are to be generated from the user DB 12b and inserts them into this prompt template to create a prompt such as that shown in Fig. 17B.

[0080] The information processing system of this embodiment executes the same processes as those shown in FIGS. 5 to 7 . The server 10 of this embodiment uses a language model when identifying tips in steps S18 and S44. For example, in step S18, the control unit 11 reads the symptom information stored in the user DB 12b in step S17 from the user DB 12b and inserts it into a prompt template containing command statements, thereby creating the prompt shown in FIG. 17A . The control unit 11 then inputs the created prompt into the language model and acquires tips for each symptom generated by the language model. The control unit 11 then performs the processes from step S19 onward. In addition, in step S44, the control unit 11 reads the cause information, effort information, and memo information of the user for whom tips are to be generated from the user DB 12b and inserts them, along with the symptom information read in step S42, into a prompt template containing command statements, thereby creating the prompt shown in FIG. 17B . The control unit 11 then inputs the created prompt into the language model and acquires tips for each symptom generated by the language model. Thereafter, the control unit 11 performs the processes from step S45 onward. Note that, if past symptom information, cause information, effort information, and memo information are stored in the user DB 12b, the control unit 11 may insert the past (e.g., past week, month, or several months) symptom information, cause information, effort information, and memo information into the prompt.

[0081] The above-described process in this embodiment enables tips to be provided to the user to be generated taking into consideration the symptom information, cause information, effort information, and memo information input by the user. Furthermore, after the processes of steps S31 and S37, for example, the control unit 11 of the server 10 may generate tips or messages using a language model based on the symptom information, cause information, effort information, and memo information stored in the user DB 12b at that time, and generate a recording completion screen as shown in FIG. 10B and provide it to the user terminal 20. In this case, the control unit 21 of the user terminal 20 may acquire the recording completion screen generated by the server 10 and display it on the display unit 25 in step S38.

[0082] In this embodiment, the prompts in Figures 17A and 17B may include sentences about the user's preferences and tastes for tips (for example, whether the user wants simple, concise tips or detailed tips), which may instruct the language model to generate tips according to the user's preferences and tastes. In this case, tips that match the user's preferences can be generated.

[0083] In the information processing system of this embodiment, when the server 10 identifies tips, the process of generating tips using a language model differs from that of embodiments 1 to 3, but the other processes are the same as those of embodiments 1 to 3, so similar effects can be obtained. Also, in this embodiment, tips are generated using a language model, so tips can be generated based on various information about the user, making it possible to provide tips that are suited to the user's condition.

[0084] The embodiments disclosed herein are to be considered in all respects as illustrative and not restrictive. The scope of the present invention is defined by the claims, not by the above meaning, and is intended to include all modifications within the meaning and scope of the claims.

[0085] The features described in each of the above-mentioned embodiments can be combined with each other. Furthermore, the independent claims and dependent claims described in the claims can be combined with each other in any and all combinations, regardless of the reference format. Furthermore, while the claims use a format in which a claim references two or more other claims (multiple claim format), this is not limited to this format. A multiple claim (multi-multi claim) that references at least one other multiple claim may also be used.

[0086] REFERENCE SIGNS LIST 10 Server 11 Control unit 12 Storage unit 13 Communication unit 14 Input unit 15 Display unit 20 User terminal 21 Control unit 22 Storage unit 23 Communication unit 12M Learning model

Claims

1. A program that causes a computer to execute the following processes: acquire symptom information regarding a user's menopausal symptoms; input the acquired symptom information into a learning model trained to output tips regarding the symptoms when symptom information regarding menopausal symptoms is input; acquire tips corresponding to the input symptom information; and output the acquired tips.

2. The program of claim 1, which causes the computer to execute a process in which the learning model is trained to output tips for each of a plurality of symptoms when symptom information for a plurality of symptoms is input, and when symptom information for a plurality of symptoms of the user is acquired, the acquired symptom information for the plurality of symptoms is input into the learning model and tips for each of the plurality of symptoms are acquired.

3. The program according to claim 1 or 2, wherein the symptom information includes information regarding hot flashes, headaches, palpitations, stiff shoulders, anxiety, or sleep state.

4. The program according to claim 1 or 2, wherein the tips include tips on body care or tips on lifestyle habits.

5. The program according to claim 1 or 2, which causes the computer to execute the following process: periodically acquiring symptom information of the user; identifying frequently occurring symptoms based on the symptom information acquired during a specified period; inputting symptom information relating to the identified symptoms into the learning model; and acquiring tips according to the input symptom information.

6. The program according to claim 1 or 2, wherein the symptom information includes the type and severity of the symptom, and the learning model is trained to output tips related to the symptom when the symptom information including the type and severity of the symptom is input.

7. The program according to claim 1 or 2, which causes the computer to execute a process of: identifying a change in the symptom based on the symptom information acquired in time series; and outputting the identified change in the symptom.

8. The program according to claim 1 or 2, which causes the computer to execute the following processes: acquiring an image of a test kit for testing the user's hormone levels; inputting the acquired image into a learning model for determining test results, which has been trained to output test results using the test kit in the captured image when an image of the test kit is input, and acquiring test results using the test kit in the input image; the learning model has been trained to output tips related to menopausal symptoms when symptom information related to the symptoms and test results using the test kit are input; inputting the acquired symptom information and test results using the test kit into the learning model, and acquiring tips corresponding to the input symptom information and test results.

9. The program of claim 8, which causes the computer to execute the following processes: acquiring responses to a questionnaire regarding the user's symptoms; identifying hormone test results based on the test results obtained using the test kit using the learning model for determining test results and the responses to the questionnaire; the learning model being trained to output tips regarding the symptoms when symptom information regarding menopausal symptoms and the hormone test results are input; and inputting the acquired symptom information and the identified hormone test results into the learning model, and acquiring tips corresponding to the input symptom information and the hormone test results.

10. The program according to claim 1 or 2, which causes the computer to execute a process of generating tips according to the acquired symptom information using a language model.

11. A program that causes a computer to execute the following processes: acquiring multiple pieces of symptom information relating to a user's menopausal symptoms in chronological order; identifying a predetermined number of symptoms for which tips should be provided to the user based on the multiple pieces of symptom information acquired over a predetermined period; identifying tips for the identified predetermined number of symptoms in accordance with the multiple pieces of symptom information acquired over the predetermined period; and outputting the identified tips.

12. An information processing method in which a computer executes the following processes: acquiring symptom information regarding a user's menopausal symptoms; inputting the acquired symptom information into a learning model trained to output tips regarding the symptoms when symptom information regarding menopausal symptoms is input; acquiring tips corresponding to the input symptom information; and outputting the acquired tips.

13. An information processing device having a control unit, wherein the control unit: acquires symptom information regarding a user's menopausal symptoms; inputs the acquired symptom information into a learning model trained to output tips regarding the symptoms when symptom information regarding menopausal symptoms is input; acquires tips corresponding to the input symptom information; and outputs the acquired tips.

Images