System and method for providing health suggestions

Abstract

In order to effectively use biological data obtained in daily life of a user to maintain and enhance health of the user and to provide the user with further customized health suggestions convenient for the user to use, the system for providing health suggestions for the user comprises: a sensor provided in a facility, continuously collecting biological data of the user; an analysis device configured to analyze the biological data collected by the sensor and determine a health state of the user; a program generation device configured to generate a health enhancement program corresponding to the user using the determination result of the health state; and a suggestion generation device configured to obtain a health suggestion for the user by inputting the determined health state, the generated health enhancement program, and additional information relating to the user into a large-scale language model.

Description

TECHNICAL FIELD

[0001] The present application relates to a system and method for providing a health recommendation corresponding to a health state of a user of a facility. BACKGROUND

[0002] In recent years, smart homes have attracted attention. A smart home generally refers to a home in which information technology (IT) is used to control devices that use electricity or gas, such as lighting fixtures, cooking appliances, and refrigeration and heating equipment within the home, thereby optimally controlling energy consumption.

[0003] The applicant has proposed, in Patent Literature 1 (Japanese Patent No. 7285886), a health promotion program providing system including a sensor provided within a home that continuously collects biological data of an occupant of the home, an analysis device that analyzes the collected biological data and determines a health state of the occupant, and a program generation device that generates a health promotion program corresponding to the occupant based on a determination result of the health state, and a health promotion program providing method corresponding to the system.

[0004] The system and method have an advantage that it is possible to effectively utilize biological data obtained in a daily life of a user to maintain and promote health of the user.

[0005] PRIOR ART DOCUMENTS

[0006] PATENT LITERATURE

[0007] Patent Literature 1: Japanese Patent No. 7285886 SUMMARY

[0008] PROBLEMS TO BE SOLVED BY THE INVENTION

[0009] The applicant has further conducted research and found a method of providing a user with a further customized user-friendly health recommendation.

[0010] That is, an object of the present application is to provide a system and method that can effectively utilize biological data obtained in a daily life of a user to maintain and promote health of the user, and can provide the user with a further customized user-friendly health recommendation.

[0011] TECHNICAL MEANS FOR SOLVING PROBLEMS

[0012] To solve the above-described technical problem, one embodiment of the present application provides a system for providing a health recommendation for a user, including:

[0013] a structure provided within a facility, including a sensor that continuously collects biological data of the user;

[0014] an analysis device configured to perform analysis of the biological data collected in the sensor, and determine a health state of the user;

[0015] a program generation device configured to generate a health improvement program corresponding to the user using a result of the determination of the health state; and

[0016] a suggestion generation device configured to input the determined health state, the generated health improvement program, and additional information related to the user into a large language model to obtain a health suggestion for the user.

[0017] In the system of the present application, preferably, the additional information includes at least one of the following information: basic information including at least one of age, gender, and nationality of the user; biological information including at least one of height, weight, body composition, basal metabolism, and blood pressure of the user; external information including at least one of finance and economic status, logistics, season, trend, temperature, and humidity; and inherent information including at least one of interest and hobby of the user, preference for things and food, allergy, physical characteristics, economic status of the individual or family, and mood on the day.

[0018] In addition, in the system of the present application, preferably, the additional information includes a reaction of the user to the health suggestion.

[0019] In addition, in the system of the present application, preferably, the biological data includes at least one of data representing expression, heartbeat, oxygen saturation, carbon dioxide exhalation, surface body temperature, deep body temperature, skin protein analysis, body composition, autonomic nerve, HbA1c, amount of water in the body, foot pressure distribution, walking posture, walking speed, joint movable area change, center of gravity change, activity amount, muscle electricity, electrocardiogram, brain wave, standing and sitting posture, body temperature, blood pressure, blood flow, heart rate, respiration, sweating, eyeball, sleep time, amount and time of excretion, blood components, urine components, saliva components, images in the oral cavity, and fecal components.

[0020] Further, in the system of the present application, preferably, the analysis device is configured to output the health state of the user from the collected biological data using a prediction model obtained by machine learning of a learning model configured to have input of biological data of a person and output of a health state of the person, using at least one of data indicating an expression, a heartbeat, an oxygen saturation, a carbon dioxide exhalation amount, a surface body temperature, a deep body temperature, a skin protein analysis, a body composition, an autonomic nerve, HbAlc, an amount of water in the body, a foot pressure distribution, a walking posture, a walking speed, a change in a joint movable area, a change in a center of gravity, an activity amount, an electromyogram, an electrocardiogram, a brain wave, a standing and sitting posture, a body temperature, a blood pressure, a blood flow, a heart rate, a respiration, a perspiration, an eyeball, a sleep time, an amount and a time of excretion, a blood component, a urine component, a saliva component, an image in the oral cavity, and a fecal component of the person, and a health state of the person including an energy consumption amount and a motion effect of the person as teacher data, and to have input of biological data of a person and output of a health state of the person as learning data.

[0021] Further, in the system of the present application, preferably, the program generation unit is configured to:

[0022] extract a biological feature of the user by comparing the collected biological data with a prescribed evaluation index;

[0023] generate a health improvement program for the user from the collected biological data using an evaluation model obtained by machine learning of a learning model configured to have input of biological data and output of an evaluation of a health improvement program, using a prescribed evaluation index and an evaluation value indicating an effect of a health improvement program as teacher data;

[0024] output the extracted biological feature and the generated health improvement program.

[0025] Another aspect of the present application provides a method for providing a health recommendation to a user, the computer executing the steps of:

[0026] implementing analysis of biological data of a user collected in a sensor provided in a structure in a facility, determining a health state of the user;

[0027] generating a health improvement program corresponding to the user using a result of the determination of the health state;

[0028] inputting the determined health state, the generated health improvement program, and additional information related to the user into a large-scale language model to obtain a health recommendation for the user.

[0029] Effects of the Invention

[0030] According to the present application, it is possible to provide a system and method that can effectively utilize biological data obtained in a user's daily life to maintain and improve the user's health, and that can provide the user with further customized, user-friendly health advice. BRIEF DESCRIPTION OF DRAWINGS

[0031] FIG. 1 is a schematic diagram showing an overall view of an exemplary embodiment of the present application.

[0032] FIG. 2 is a diagram showing an example of a check gate including the display device 10.

[0033] FIG. 3 is a diagram showing another example of the display device 10.

[0034] FIG. 4 is a diagram showing still another example of the display device 10.

[0035] FIG. 5 is a block diagram showing the overall structure of the system 1 for providing health advice.

[0036] FIG. 6 is a flowchart showing the processing steps for providing health advice.

[0037] FIG. 7 is a diagram showing the overall view of the processing including the generation, provision, and user's response of health advice. DETAILED DESCRIPTION

[0038] Hereinafter, representative embodiments of the system and method of the present application will be described in detail with reference to the accompanying drawings. However, the present application is not limited to these drawings. In addition, the drawings are diagrams for conceptually explaining the present application, and the size, ratio, or amount is sometimes exaggerated or simplified as needed for easy understanding.

[0039] 1. Overall image of the present embodiment

[0040] Referring to FIG. 1 , the overall view of the system 1 around the present embodiment is explained.

[0041] The facility 70 in the present embodiment (the present application) is a concept including all facilities (also referred to as a facility, an institution, a site, a place, or an activity base, etc.) capable of using the system 1. As examples of the facility 70, there are a residence, a workplace (an office, a factory, etc.), an educational facility such as a school, a medical facility such as a hospital and a medical examination center, a nursing facility, a care facility, a support facility, an inpatient facility for the elderly, a sports facility, an entertainment facility, a meeting station, an accommodation facility, a station, and an airport. The facility 70 can also include a mobile body such as a vehicle, a ship, and an airplane.

[0042] At each of the periphery of the prescribed place or area in the facility 70, various sensors are provided, and biological data of the user is sensed in the sensors. In the present application, the configuration in which the periphery of the prescribed place or area is provided with various sensors is referred to as a "check gate". The types of sensors that can be used here are, for example, an image sensor, a temperature sensor, a weight sensor, etc., which are described in detail later. These sensors can be provided individually in the facility 70, or can be assembled in the display device 10 constituting a main body of the check gate.

[0043] Here, the biological data refers to all data obtained directly from the user or indirectly through prescribed processing, and includes, for example, image data obtained by photographing the user, and data obtained by analyzing the image data.

[0044] In the biological data, not only dynamic data such as a walking posture, a walking speed, a change in a joint movable range, a change in a center of gravity, and an activity amount, but also static data including time series data such as electromyography, electrocardiography, and brain waves, a standing and sitting posture, a body temperature, a blood pressure and a blood flow, a heart rate, a respiration, sweating, an eyeball, a sleep time, an amount and a time of excretion, a blood component, a blood sugar value, a urine component, a saliva component, an image in the oral cavity (tongue, pharynx, etc.), a fecal component, an image of a face, a moisture and an oil of skin, etc.

[0045] The biological data can also be stored in a database (DB) together with other types of data. The database (DB) can be created by a computer belonging to the system 1, or by an external computer 80.

[0046] The system 1 acquires the user's biological data via the display device 10 and the other device 30, and evaluates the user's health state by computer analysis of the acquired biological data. All or a part of the above computer analysis can be performed by the display device 10, or can also be performed by another computer within the facility 70 or the external computer 80. In this case, the analysis result can be transmitted from the other computer or the external computer 80 to the display device 10.

[0047] Examples of health evaluation are described in detail later, such as tongue diagnosis, examination of cognitive function • stress, blood pressure, screening of infectious diseases, prediction of dementia, prediction of posture control, examination of mineral balance, estimation of grip strength, estimation of body composition, estimation of lower extremity joint disease, estimation of heartbeat • pulse, and provision of functions customized according to the place or location where the display device 10 is introduced. Also, biological data is naturally acquired in the user's life.

[0048] Based on the evaluation result of the health state obtained, a health promotion program customized for each user can also be prepared. Here, as the health promotion program prepared, various programs and various health information for maintaining • promoting the user's physical and mental health can be listed in addition to exercise programs.

[0049] For example, for a user evaluated as being in an excessive stress state at the workplace, measures for stress mitigation such as business improvement, acquisition of leave, implementation of moderate exercise, securing of sufficient sleep, and consultation with a doctor can be proposed. Or, an exercise menu that matches the user's health state can also be prompted.

[0050] Or, for a user evaluated as being in an insufficient exercise state at the workplace, implementation of moderate exercise can be proposed, for example.

[0051] Or, for a user suspected of having insufficient or excessive nutrition, a diet menu can be proposed.

[0052] Or, for a user evaluated as having concerns about cognitive function, consultation with a doctor can be proposed.

[0053] Or, for a user suspected of having an infectious disease, consultation with a doctor can be proposed.

[0054] Or, for a user evaluated as having a disrupted mineral balance, improvement of meals or a visit to a doctor can be proposed.

[0055] The health promotion program is provided to the user together with the evaluation result of the health state, and the user implements the program.

[0056] Then, by repeatedly performing the above sensing, the determination of the health state, and the making and the user's implementation of the health improvement program, the improvement of the health function of the user is realized.

[0057] For example, by only the user standing in front of the display device 10, the display device 10 acquires various health data related to the user. The display device 10 can also be utilized as a usual mirror, and thus can be naturally introduced in life while ensuring the available mirror quality.

[0058] In the display device 10 (or the external computer 80), the artificial intelligence model integrates and analyzes the health data stored in the data server and the health data sensed by the display device 10 every day. As a result, it is possible to perceive the body abnormality that the user, his / her family, medical-related persons, or nursing-related persons cannot easily notice, and to make the health suggestion for each user, the alarm. Thereby, it is possible to contribute to the sub-health improvement or the improvement of the health.

[0059] At this time, it is preferable to input the determined health state, the generated health improvement program, and the additional information related to the user to the large-scale language model to obtain the health suggestion customized for the user. Thereby, it is possible to obtain the health suggestion that is convenient for the user to use, to promote the understanding of the user related to the health suggestion, and to expect that the user can easily practice the health suggestion.

[0060] Further, the collected biological data, the evaluation result of the health state, and the made health improvement program are aggregated in the data server or the like, and under the permission of the user, it is also possible to be prompted to the medical-related persons such as doctors, and to be utilized by the medical-related persons. In addition, the generated determination result of the health state can also be prompted to the medical-related persons under the permission of the user, and to be utilized by the medical-related persons.

[0061] Further, the collected biological data and the generated determination result of the health state can also be utilized in the administration and regional society under the permission of the user in order to form a more comfortable and habitable community, the health management of the residents, and the old people can continue to live independently at home.

[0062] 2. Details of the system of the present embodiment

[0063] The system 1 of the present embodiment realizes the following functions: a sensing function of collecting biological data from the user within the facility 70; an analysis and determination function of determining the health state of the user by analyzing the collected biological data; a program generation function of generating a health improvement program based on the determination result of the health state; and a suggestion generation function of obtaining a health suggestion for the user by inputting the determined health state, the generated health improvement program, and additional information related to the user to a large-scale language model.

[0064] To realize such a function, the system 1 includes a display device 10 and a model generation apparatus 20 (refer to FIG. 5 ). The system 1 can further include an output apparatus 40, a storage apparatus 50, and a suggestion generation apparatus 60. These constituent elements are communicably connected via a wired or wireless communication network. The display device 10, the model generation apparatus 20, the model generation apparatus 20, the output apparatus 40, the storage apparatus 50, and the suggestion generation apparatus 60 can be separate or can be integrated.

[0065] The storage apparatus 50 can be provided as a constituent element of a data center (not shown) in which not only collected data but also collected samples (for example, samples of saliva, feces) can be stored. The data center in this case is also referred to as a biobank.

[0066] Hereinafter, these constituent elements will be described in detail.

[0067] 2-1. Display device

[0068] Refer to FIGS. 2 to 4 The display device 10 will be described. FIGS. 2 to 4 Various modes of the display device 10 are shown.

[0069] The display device 10 is capable of displaying various characters and images.

[0070] The display device 10 can be installed in a facility 70 or can be placed in the facility 70. As a place where the display device 10 is installed, in a house, for example, a hall, a washroom, a living room, a parlor can be cited. In addition, in a collective house, an entrance can be cited, in a lodging facility, an entrance or a living room can be cited, in an office, an entrance or a workroom, a fitness or sports site, a beauty salon, a beauty service counter can be cited.

[0071] The display device 10 is capable of acquiring various biological data. The biological data that can be acquired in the display device 10 is described above. The display device 10 can automatically start acquisition of biological data when it is determined based on a detection result of a human sensor that a user approaches to a prescribed range from the display device 10, for example, or can start acquisition of biological data according to an operation of the display device 10 by the user. As the operation of the user, for example, a press or a touch of a button of the display device 10 can be cited, but is not limited thereto.

[0072] The display device 10 can also be configured to include a mirror 11, a display 12, an image sensor 13 (video sensor), a processor, a memory, and a communication device, for example. Such a display device 10 can also be said to be an Internet of Things (IoT) device called a smart mirror.

[0073] That is, the display device 10 has a mirror surface 11. The mirror surface 11 can reflect the entire body of the user, can reflect the upper body, or can mainly reflect the face. The mirror surface 11 can be configured as a laminate of a transparent plate-shaped glass and a metal-made reflecting surface (a back mirror), for example.

[0074] A display 12 is assembled in the mirror surface 11. As the display 12, a liquid crystal display, an organic EL display, or a screen of a video projector can be used, for example.

[0075] An image sensor 13 is embedded or installed in the mirror surface 11. The image sensor 13 acquires image data by photographing the user located in front of the display device 10. The image sensor 13 is a semiconductor image sensor, and can be a CCD image sensor or a CMOS image sensor, for example.

[0076] The wavelength band of the image sensor 13 is not limited to visible light. That is, the image sensor 13 can be an infrared temperature camera, a dual spectrum imaging camera, or an RFID sensor.

[0077] The image data acquired by the image sensor 13 is transmitted to a processor and a memory (not shown) to perform various processes or be saved. The processor can include a central operation processing device (CPU), a graphics processing unit (GPU), or a microprocessor, for example. The memory can include a random access memory (RAM) and a read only memory (ROM).

[0078] Here, an example of a process performed on acquired biological data and a result of the process is described:

[0079] - an estimation result of a bacterial flora in the oral cavity or a bacterial flora in the intestines obtained from analysis of a tongue image, and an evaluation of a disease risk based on the estimation result

[0080] - an evaluation of cognitive function and stress predicted from an expression, a line of sight, and a pulse wave obtained from analysis of a face image

[0081] - detection of body temperature, heartbeat fluctuation, and CO2 exhalation amount obtained from analysis of a face image, and a health examination and a screening of infectious diseases based on the detection result. In addition, the face image can be used for individual recognition and authentication.

[0082] - detection of amyloid β obtained from analysis of an image of a palm, and an evaluation of a dementia risk predicted from an accumulation amount of amyloid β. In addition, the image of the palm can be used for individual recognition and authentication.

[0083] - detection of dizziness or shaking obtained from analysis of a whole body image, and prediction of a functional abnormality of the nervous system based on the detection result

[0084] - detection of harmful metals or minerals from analysis of images of the face or palm, and evaluation of the body's mineral balance based on the detection results

[0085] The above processing can be performed, for example, as follows.

[0086] That is, based on the data accumulated in the data server (storage device 50), pairs of sensing data and correct labels (correct data, teacher data) corresponding thereto are created, and machine learning based on an artificial intelligence (AI) model (for example, deep learning) is performed. The machine learning is repeatedly performed, features of the data are extracted, and the health state of the user is predicted.

[0087] As described above, the information (that is, the determined health state and the generated health improvement program) in which the feature extraction by the AI model is performed and the additional information can also be input to a large language model (LLM), and further converted into information that is convenient for the user to use. The display device 10 can itself perform the generation of the health advice based on the large language model, or can also cooperate with an external computer 80 capable of performing the large language model via, for example, an application programming interface (API). Details of the large language model and the generation of the health advice based on the large language model will be described later.

[0088] Returning to the description of the display device 10. The display device 10 can prompt the converted information to the user. For example, consider display to the display 12 or reading of the text.

[0089] In addition, by causing the above AI model to learn the reaction of the user at this time, the prompting of the information of the individual with higher accuracy can be performed, and connected to the behavior change of the user.

[0090] The above analysis can be realized by executing various analysis programs stored in the memory and various biological data on the processor. Alternatively, the analysis result can also be obtained by applying various biological data to a learned model stored in the memory. In addition, appropriate preprocessing can also be performed before execution of the analysis.

[0091] Alternatively, the analysis can also be performed by the external computer 80, and the analysis result is returned to the display device 10.

[0092] The acquired image data, preprocessed data, or analysis results can be saved in the storage or can be transmitted to other mirror devices, the external computer 80, or the storage device 50 via a communication device. The communication device can be a wired communication device or a wireless communication device. As the wired communication device, for example, there are an Ethernet adapter, a modem, and as the wireless communication device, for example, there are Bluetooth (trademark), Wi-Fi, Thread, ZigBee, Matter, low power wide area wireless (LPWA). The LPWA includes, for example, LTE Category M1, NB-IoT, LoRaWAN, Sigfox, Wi-SUN, ZETA, ELTRES, and the like.

[0093] In addition, the display device 10 can also include a touch panel (including a device equipped with a non-contact type interface), other input devices such as a microphone, other output devices such as a speaker.

[0094] Therefore, the display device 10 can be effectively utilized as a mirror, and can also be effectively utilized as a biological data collection device, a display device of collected and analyzed biological data, various information and content, and a tool for further communication.

[0095] For example, the display 12 embedded in the mirror surface 11 can project an image containing an image captured by the image sensor 13 to the surface of the mirror. In addition, the display 12 can display acquired biological data. As a display method, for example, in addition to the display of characters, graphical display such as charts, illustrations, and the like can be cited.

[0096] The display device 10 can perform face recognition and personal authentication by the built-in image sensor 13 or other sensors. In addition, in the case of a built-in microphone, a temperature / humidity sensor, the display device 10 can perform sound recognition and environmental measurement.

[0097] The display device 10 can communicate with other devices 30 in the facility 70 via a built-in communication device. The other devices 30 include other display devices 10, PCs, and portable terminals disposed in the same facility 70. The mobile terminal includes, for example, a smartphone, a tablet terminal, and a wearable terminal. In addition, the display device 10 can communicate with the external computer 80 via the Internet. The external computer 80 includes a server, a PC, and a portable terminal located outside the facility 70.

[0098] The display device 10 can perform sound output in the case of a built-in speaker. In addition, the display device 10 can be operated simply by touching the surface of the mirror, for example, in the case of being equipped with an input device such as a touch panel.

[0099] Thus, the display device 10 acquires various biological data, and evaluates the health state of the user by computer analysis of the acquired biological data. However, all or a part of the computer analysis can also be performed by another computer (for example, the external computer 80), and the analysis result is returned to the display device 10.

[0100] The display device 10 can acquire biological data of the user at different times (first and second times), and obtain time-series biological data. Then, by analysis of the time-series biological data, time-series evaluation of the health state can be performed.

[0101] Alternatively, by comparing the same kind of biological data (first and second biological data) sensed at different facilities 70 (first and second facilities 70; for example, home and workplace), the health state of the user in different environments can also be accurately evaluated.

[0102] As a specific example of the latter, when the measured values of blood pressure, heartbeat, respiration rate, and perspiration of the user in the workplace are greater than the corresponding values of the user in the residence by a set value or more, it can be evaluated that the user is in an excessive state of tension in the workplace.

[0103] Alternatively, when the measured values of blood pressure, heartbeat, respiration rate, blood flow, and perspiration of the user in the exercise facility are within a range that is set in advance as an allowable range from the corresponding values of the user in the residence, it can be evaluated that the user has performed moderate exercise.

[0104] Alternatively, when the measured or estimated values of moving distance, moving speed, and metabolism of the user in the workplace are less than the corresponding values of the user in the residence by a set value or more, it can be evaluated that the user is in a state of insufficient exercise in the workplace.

[0105] Alternatively, when the measured values of blood pressure, heartbeat, and respiration rate of the user in the lodging facility and the entertainment facility are less than the corresponding values of the user in the residence by a set value or more, it can be evaluated that the user is in a state of moderate relaxation in the lodging facility and the entertainment facility.

[0106] Alternatively, when the measured values of brain waves of the user in the educational facility or the workplace are different from the corresponding values of the user in the residence, it can be evaluated that the user is in a state of lacking concentration.

[0107] 2-2. Other devices

[0108] Other devices 30 can also be provided within the facility 70. The display device 10 can be combined or linked with the other devices 30.

[0109] As the other device 30, for example, a body composition detector, a grip sensor, a pressure sensor, a motion capture device, a temperature sensor, a humidity sensor, a component analyzer, a weight sensor, a flow sensor, a position sensor, an infrared temperature camera, an ultrasonic imaging camera, an RFID sensor, a motion capture device (none of which is shown), and the like can be cited, but are not limited thereto. Furthermore, the display device 10 can also include these devices. In FIG. 2 In the example, as an example of the other device 30 described above, the display device 10 has a grip sensor that takes the form of a handle.

[0110] The other device 30 also continues to collect the biological data of the user H, and transmits the collected biological data to the storage 50 together with the collection date and time to be stored. The sensing of the biological data of the user H by the other device 30 can also be performed, for example, by measurement at regular intervals.

[0111] For example, the body composition detector is provided under the feet of the display device 10, and is capable of measuring the weight, BMI, body fat rate, visceral fat level, muscle mass, body water rate, basal metabolic amount, and estimated bone mass of the user. The body composition detector can also be embedded in the floor in such a manner that the floor surface of the facility 70 becomes the same. In addition, the body composition detector can also be assembled to the handrail. The body composition detector acquires the body composition data of the user, for example, the plantar pressure, and transmits the acquired data to the display device 10. The display device 10 predicts lower limb joint disease from the change in the plantar pressure.

[0112] The grip sensor is, for example, bar-shaped, and is provided next to the display device 10. The grip sensor can also function as a holding member or a handrail. The grip sensor detects the pressure applied to the grip sensor when the user holds the grip sensor, and transmits the pressure data to the display device 10. The display device 10 calculates the grip using the pressure data and the bone analysis based on the whole body image of the user. Then, the display device 10 predicts the physical deterioration of the user using the change in the grip and the walking speed calculated from the whole body image of the user.

[0113] The infrared temperature camera is, for example, provided in the living room, study, bedroom, and workroom, and the like of the facility 70, and is capable of measuring the expression, heartbeat, oxygen saturation, carbon dioxide exhalation amount, surface body temperature, and deep body temperature of the user H.

[0114] The dual spectrum imaging camera is provided, for example, at the entrance and its vicinity, and in addition to the personal authentication function of the user H, can perform skin protein analysis of the user H, measurement of body composition (for example, the composition ratio including fat, bone, and soft tissue excluding fat, body weight, body fat, basal metabolism, visceral fat, muscle mass, and bone mass), and determination of the autonomic nerve (for example, including the degree of fatigue of the autonomic nerve, and the balance between the sympathetic nerve and the parasympathetic nerve). Alternatively, the dual spectrum imaging camera can be provided at the dining table, the kitchen, and their vicinity, and can measure glycated hemoglobin (HbAlc).

[0115] The RFID sensor is provided on the wall surface in the room, and can measure the amount of water in the body of the user H.

[0116] The pressure sensor is provided, for example, on the floor of the corridor and the room, and can measure the foot pressure distribution of the user H.

[0117] The motion capture device is provided, for example, on the corridor in the room, and can measure the walking posture, the walking speed, the change in the joint movable area, the change in the center of gravity, and the activity amount of the user H.

[0118] The other device 30 can be provided in a manner that is hidden with respect to the human eye in a manner that the user H does not immediately notice. This is to suppress the user H from becoming nervous by noticing the existence of the other device 30, and to cause the sensed biological data to deviate from the value of the user H in the daily life. That is, it is to obtain the user H in a natural state in the daily life.

[0119] In addition, various sensors for obtaining environmental data can also be provided. That is, a thermometer for measuring the room temperature, a hygrometer for measuring the humidity of the room, an illuminometer for measuring the brightness of the room, a carbon dioxide measuring device for measuring the CO2 concentration of the room, a noise sensor for measuring the environmental sound, a light sensor for detecting the state of the light in the environment, and the like can be provided. The environmental data collected in these sensors is transmitted to the storage device 50 and stored. Furthermore, the display device 10 can also include these sensors.

[0120] In addition, the biological data of the user can also be obtained using various sensors built into wearable devices and portable information terminals such as smartphones. Such biological data is useful as an aid in determining the health status of the user.

[0121] The collected biological data, environmental data, and other data are transmitted to the storage device 50 and stored. The determination results of the health status described below, and the generated health improvement program can also be stored in the storage device 50. The various data stored in the storage device 50 can also be made available in an external computer 80 used by the doctor and medical institution in charge of the user H with the permission of the user H.

[0122] Through this medical collaboration, the diagnosis and various health records required for treatment of the user H can be shared among medical related parties in a timely manner. Of course, the data collaboration can also be targeted at doctors, dentists, pharmacists, nutritionists, or nurses, and is not limited to these medical related parties.

[0123] 2-3. Model generation apparatus

[0124] The model generation device 20 is a computer that generates various statistical models. The model generation device 20 can also perform various processes using the generated statistical models.

[0125] Hereinafter, as an example of a statistical model, a statistical model for analyzing biological data and a statistical model for generating a health promotion program will be described.

[0126] 2-3-1. Analysis model of biological data

[0127] The model generation device 20 can generate a prediction model for analyzing biological data collected by the display device 10 and the other device 30. The model generation device 20 can also utilize AI analysis for data analysis.

[0128] Specifically, the model generation device 20 generates an analysis model that sets biological data of a person as input and sets the health status of the person as output, by machine learning, using, as teacher data, biological data of the person that includes at least one of data indicating the expression, heartbeat, oxygen saturation, carbon dioxide exhalation amount, surface body temperature, deep body temperature, skin protein analysis, body composition, autonomic nerve, HbA1c, amount of water in the body, foot pressure distribution, walking posture, walking speed, joint movable area change, center of gravity change, activity amount, muscle electricity, electrocardiogram, brain wave, standing and sitting posture, body temperature, blood pressure, blood flow, heart rate, respiration, sweating, eyeball, sleep time, amount and time of excretion, blood components, urine components, saliva components, images in the oral cavity, and fecal components of the person, and the health status of the person that includes the energy consumption amount and exercise effect of the person. Then, the model generation device 20 outputs the health status of the user from the collected biological data using the generated prediction model. As an example of machine learning that can be utilized, deep learning can be cited, but is not limited thereto.

[0129] For example, the model generation device 20 can implement machine learning that uses at least one of the biological data of a person's expression, heartbeat, oxygen saturation, carbon dioxide exhalation amount, surface body temperature, deep body temperature, skin protein analysis, body composition, autonomic nerve, HbAlc, in-vivo water amount, foot pressure distribution, walking posture, walking speed, joint movable area change, center of gravity change, activity amount, and the like as an existing index as a teacher variable (teacher data), and biological data of the same kind as the collected data set as an explanation variable, to create an analysis program (learned model).

[0130] In addition, the model generation device 20 can also use basic information such as age, gender, height, weight, and the like as parameters to be used to algorithmically analyze biological information such as expression, heartbeat, oxygen saturation, carbon dioxide exhalation amount, surface body temperature, deep body temperature, skin protein analysis, body composition, autonomic nerve, HbAlc, in-vivo water amount, foot pressure distribution, joint angle during walking, and changes thereto over time, lower limb muscle strength, center of gravity, or three-dimensional movement distance of observation points related thereto, walking posture, walking speed, activity amount, and the like, and motor function information. That is, the model generation device 20 can obtain an objective evaluation of the health status of the user H including health function and motor function by applying the sensed data to the above-described model.

[0131] The model generation device 20 can transmit the generated analysis model to the display device 10 or the storage device 50. In addition, the model generation device 20 can transmit the determination result of the health status of the user H to the display device 10, the output device 40, or the storage device 50. The model generation device 20 can also transmit the determination result of the health status of the user H to the terminal of the user H and his or her family. Then, by outputting the determination result of the health status of the user H to the display device 10 or the output device 40, the user H and his or her family can early detect the illness or poor physical condition of the user H, and can be applied to the health management of the user H.

[0132] Of course, with the permission of the user H, the above-described determination result can be effectively used among the related persons including medical-related persons and nursing-related persons.

[0133] As a result, the determination result can be effectively used for the support of the user H and the cooperation with the related persons.

[0134] 2-3-2. Generation model of health promotion program

[0135] The model generation device 20 can also generate a health improvement program customized for each user H using the determination result of the health status. The following is an example of a statistical model for generating a health improvement program.

[0136] That is, the model generation device 20 extracts the biological characteristics of the user by comparing the collected biological data with the prescribed evaluation indexes. Then, the model generation device 20 generates an evaluation model that sets input as biological data and sets output as an evaluation of the health promotion program, using the prescribed evaluation indexes and the evaluation value indicating the effect of the health promotion program as teacher data, by machine learning. The generated evaluation model can also be transmitted to the storage device 50 or the display device 10.

[0137] The model generation device 20 can also use this evaluation model to output, from the collected biological data, the health promotion program to be presented to the user together with the biological characteristics of the user.

[0138] More specifically, the health promotion program is roughly divided into a program related to health function and a program related to motor function.

[0139] In the case of generating a program related to health function, the model generation device 20 compares, for example, expression, heartbeat, oxygen saturation, carbon dioxide exhalation amount, surface body temperature, deep body temperature, skin protein analysis, body composition, autonomic nerve, HbA1c, amount of water in the body, and foot pressure distribution, with evaluation indexes such as each person's normal range and epidemiology, medical guidelines, and the like, thereby specifying the biological characteristics of the user H. Then, the model generation device 20 can construct an algorithm (evaluation model) related to each person's living habits by performing AI analysis on the measured data, basic information such as age / gender / height / weight, and the user H's living environment data. Here, as an example of AI analysis, deep learning can be given, but is not limited thereto. The model generation device 20 can also present action changes and effects accompanying the same by applying the constructed evaluation model to the user's biological data and the like.

[0140] In the case of generating a program related to motor function, the model generation device 20 specifies the characteristics of the user H on dynamic analysis by comparing, for example, the position and speed of each joint, and the joint angle and angular velocity. Then, the model generation device 20 performs construction of a motion measurement algorithm (evaluation model) based on the measured data, in addition to the estimation of energy consumption amount in exercise and exercise effect. Here, as an example of AI analysis, deep learning can be given, but is not limited thereto.

[0141] The model generation device 20 can also present recommended exercise and energy consumption amount in exercise and exercise effect by applying the constructed evaluation model to the user's biological data and the like.

[0142] The model generation device 20 can transmit the constructed evaluation model to the display device 10 or the storage device 50. In addition, the model generation device 20 can transmit the generated various programs to the display device 10, the output device 40, or the storage device 50. The model generation device 20 can also transmit the generated various programs to the terminal of the user and his or her family.

[0143] Further, the model generation device 20 can also generate a large-scale language model by itself. The large-scale language model learns to accept input of an evaluation of a health state, a health promotion program, and additional information (the basic information, the biological information, the external information, and the inherent information described above) related to a certain person, and output a health recommendation for the person.

[0144] By outputting the above-described program or health recommendation to the display device 10 or the output device 40, the user H and his or her family can clearly grasp what should be done, that is, a change in action, for maintaining, restoring, and improving the health function and the motor function of the user H, and can apply it to health management of the user H.

[0145] Of course, with the permission of the user H, the above-described program or health recommendation can be shared among the related persons including the medical-related persons and the nursing-related persons.

[0146] Here, an example of a change in action is given.

[0147] For example, in the kitchen, the user operates a monitor of the kitchen or his or her own portable terminal to cause a prescribed application program to start. The application program evaluates the health state of the user with reference to the biological data of the user and the like, and prompts a menu, materials, a cooking method, nutrients, a calorie amount, and the like that match the user personally. The application program can also acquire an image of a dish prepared by the user via the image sensor 13, and calculate and display the nutrients and energy contained in the dish.

[0148] In the living room, the user operates a television or a monitor to cause an application program to start. The application program proposes a game using eye tracking in order to improve or grasp cognitive function. When the user selects any one from among a plurality of games suggested, the game starts, and the cognitive function is measured. After the game, the application program displays feedback related to the score of the game and the improvement of the cognitive function with reference to the biological data required.

[0149] In the living room, the user operates a mirror-type monitor (for example, the display device 10) to cause an application program to start. The application program evaluates the health state of the user with reference to the biological data of the user and the like, and prompts a menu of a moderate exercise load that matches the user personally.

[0150] A virtual trainer is displayed on a mirror monitor, and the user practices the exercise menu following the movements of the virtual trainer. The application calculates and displays the energy consumed after the exercise, and prompts feedback related to the exercise function.

[0151] When going out, the user operates his or her own portable terminal to cause the application to start. The application displays health advice together with the measurement results of the user's health data.

[0152] Further, the model generation device 20 can be configured to include one or a plurality of computers including an arithmetic circuit (processor) such as a central processing device (CPU) and a memory such as a random access memory (RAM) and a read-only memory (ROM) for reading. The functions of the model generation device 20 described above can be realized by reading an execution program stored in the ROM into the RAM and executing it by the CPU. The model generation device 20 can use the biological data collected continuously and regularly or at a timing considered necessary by each person to determine the health state, generate the health improvement program, and generate the health advice.

[0153] Alternatively, the above functions of the model generation device 20 can also be incorporated into the display device 10.

[0154] 2-4. Suggestion generation apparatus

[0155] The advice generation device 60 is a computer configured to input the determined health state, the generated health improvement program, and additional information related to the user into a large-scale language model, and output a health advice for the user H. That is, the advice generation device 60 stores a large-scale language model that has been learned, and can output a health advice according to an input. Alternatively, the advice generation device 60 may, for example, access an external computer 80 via an API, and use a large-scale language model that has been learned. Hereinafter, the generation of a health advice based on a large-scale language model will be described with reference to FIG. 6. FIG. 7

[0156] The large-scale language model is a language model constructed using a large amount of data sets and a deep learning technique (for example, Transformer), and is a model that models the probability of occurrence of articles and words. Since the large-scale language model can understand and generate natural language or other content, and perform a wide range of tasks, the inventors believe that by inputting the above information to the large-scale language model, a health advice that is customized for the user H and is easy for the user to use can be obtained.

[0157] ​Here, as a large-scale language model, for example, a known model such as BERT (Bidirectional Encoder Representations from Transformers), GPT (Generative Pre-trained Transformer), Bard, LLaMA, Gemini, and the like can be used. In order to use these known models, the system 1, for example, the display device 10 can use an API, or a computer within the system 1 can be used.

[0158] Alternatively, the system 1, for example, the model generation device 20 can also generate a large-scale language model that is inherently generated for the purpose of health advice by machine learning.

[0159] The items input to the large-scale language model are roughly classified into the following 5 items:

[0160] - Basic information: age, sex, nationality, and the like of the user

[0161] - Biological information: height, weight, body composition, basal metabolism, blood pressure, and the like of the user

[0162] - External information: financial and economic information, logistics, season, trend, temperature, humidity, and the like

[0163] - Inherent information: interests and preferences of the user, preferences for things and food, allergies, local features of the body (for example, larger feet relative to height, and the like), economic information of the individual or family, mood on the day, and the like

[0164] - Information obtained by feature extraction by an AI model

[0165] Here, the information input to the AI model can also be input to the large-scale language model.

[0166] In addition, the basic information, biological information, external information, and inherent information are input to the large-scale language model as additional information. These kinds of information are useful for customizing health advice to the user H.

[0167] The additional information can also include the user's reaction to the provided health advice.

[0168] For example, for the user A, consider a scenario of a health recommendation of "Mr. A's nutritional state is measured as a result of vitamin C deficiency, so it is recommended to take lemon." For such a health recommendation, the user A makes a reaction of not being able to eat lemon for such a reason of allergy or dislike. The reaction can be acquired, for example, by the user A inputting via a touch panel or a microphone of the display device 10, or action analysis on the user A's action acquired by an image sensor. That is, the result of the user A's reaction, the user A's food preference or allergy information, that is, information of "Mr. A cannot eat lemon" can be obtained. The information can be registered in the database.

[0169] The large-scale language model, when receiving the user A's food preference or allergy information as an additional input, executes the process again, and outputs an updated health recommendation. The updated health recommendation is, for example, "Then, try to take citrus or kiwi. If consumed with yogurt in the morning, it can also adjust the intestinal environment. The effect after 1 week of continuous consumption is as follows.~~~". The update of the health recommendation can be implemented at any time according to the user A's reaction, and the number of updates is not limited.

[0170] 2-5. Output apparatus

[0171] The output device 40 is a device that outputs a health recommendation including a determination result of a health state and a health promotion program. The output device 40 includes a terminal of the user H (including a smartphone, a personal computer), a display, a printer, a projector, and a speaker. The display device 10 can also function as the output device 40.

[0172] For example, in a case where the determination result of the health state indicates an abnormality of the user H's health state, that is, a departure from a normal range (for example, for data represented by a numerical value, in a case where it becomes a prescribed threshold or less or in a case where it departs from a prescribed range), the output device 40 can also output an alarm to a terminal of the user H and his or her related person. In addition, the output device 40 can also display other information related to life, such as energy, information related to a residence, and regional information.

[0173] 3. Action of the system

[0174] Reference FIG. 6 The processing steps for providing a health promotion program by the system 1 will be described.

[0175] First, in step S1, the display device 10 and the other device 30 sense the user H's biological data, which is stored in the storage device 50 in association with the sensing time. The sensing is continuously performed.

[0176] Then, in step S2, the model generation device 20 or the display device 10 analyzes the collected biological data, determines the health state of the user H, and stores the determination result in the storage device 50. For example, the model generation device 20 or the display device 10 stores a learned model, and obtains the health state of the user H by applying the biological data of the user H to the learned model. Step S2 is executed periodically or at a timing deemed necessary by each person.

[0177] When the determination result of the health state is generated, in step S3, the model generation device 20 or the display device 10 generates a health improvement program for each user H based on the determination result of the health state, and stores the program in the storage device 50.

[0178] The determination result of the health state and the health improvement program are displayed on the output device 40 or the display device 10. Thereby, the user H can grasp his or her own health state (for example, health function and exercise function), and can explicitly grasp what should be done for the maintenance, recovery, and improvement of his or her own health function and exercise function. The user H can practice the prompted health improvement program, and apply it to his or her own health management.

[0179] Alternatively, the determination result of the health state and the health improvement program can also be input to the large-scale language model together with other information about the user, and a health recommendation customized for the user is generated (step S4). The health recommendation customized for the user is convenient for the user to use, and can be easily practiced by the user.

[0180] Next, in step S5, it is determined whether the user has reacted to the prompted health recommendation. In a case where the reaction of the user is confirmed, the process returns to step S4, various information including the reaction of the user is input to the large-scale language model, and an updated health recommendation is prompted to the user. The reaction of the user can include characteristics of the user who is unknown to the system 1, and therefore, the health recommendation updated using the reaction further approaches the user.

[0181] In addition, additional information can also be input to the large-scale language model each time the reaction of the user is obtained, and an updated health recommendation is further prompted to the user. Thereby, it is expected to further improve the accuracy of the health recommendation.

[0182] Alternatively, in a case where the reaction of the user is confirmed within a prescribed period, the process of step S1 is started again.

[0183] An example of the action is described below.

[0184] Here, the user A receives various health recommendations.

[0185] When the user A stands in front of the display device 10, the display device 10 recognizes the individual by face authentication or the like, and starts a prescribed application. The application gives various instructions to the user A by displaying a virtual character on the display 12 of the display device 10. That is, the display device 10 senses the biological data while conducting a conversation between the virtual character (e.g., a virtual instructor) and the user.

[0186] When the measurement is completed, the display device 10 displays the measurement result and a health recommendation according to the selection of the user A. At this time, the display device 10 can also display a virtual character (e.g., a virtual doctor) on the display 12 and make the character give an explanation.

[0187] Further, in a case where the user A wishes to display an exercise menu, the display device 10 can also display a virtual character (e.g., a virtual trainer) on the display 12 and prompt the exercise menu for each person.

[0188] Here, an example of the measurement result and the health recommendation is given.

[0189] "Muscle mass has been observed to decrease over the past few days. A diet menu and an exercise menu that match Mr. A are suggested here. Let's work together! After achieving this, the following health effects can be expected....".

[0190] Then, the sensing is repeated and the above steps are repeated. Thus, the user A can effectively use the biological data obtained in daily life to achieve an improvement in the health function of the user A. In addition, the user A can grasp the effect of the practice of the health improvement program from the determination result of the health state next time.

[0191] Other examples are given.

[0192] Here, a scenario in which the user B receives various health recommendations is considered.

[0193] The user B extends the tongue in front of the display device 10 and takes an image of the tongue. The image of the tongue of the user B is input to an AI model for tongue diagnosis together with other information such as the age, sex, height, weight, and the like of the user B.

[0194] Here, the AI model for tongue diagnosis is a learned model obtained by machine learning with the image of the tongue as input and an evaluation of the health state as output. The model also learns to generate a health improvement program from the evaluation of the health state and the above-mentioned other information. The evaluation of the health state is, for example, a prediction of the bacterial flora in the oral cavity, the bacterial flora in the intestines, an evaluation of the risk of digestive diseases, and the like. The health improvement program is, for example, a recommendation of exercise, meals, and the amount of sleep aimed at improvement of the intestinal flora.

[0195] The information subjected to feature extraction by the AI model described above is input to a large-scale language model together with other information about user B (e.g., basic information, biological information, external information, inherent information, etc.). The large-scale language model transforms the input information into information that is convenient for the user to use, and prompts the user B.

[0196] For example, if, in addition to the basic information of user B, sample data of intestinal flora, exercise, diet, amount of sleep, and the like, health promotion programs for improving intestinal flora are input to a large-scale language model (e.g., GPT), the large-scale language model answers as follows as an example of activities for a day.

[0197] Breakfast example:

[0198] • Add Greek yogurt and banana to oatmeal (e.g., whole grain oats)

[0199] • Green tea or black coffee

[0200] Morning activity example

[0201] • Commute by walking or cycling

[0202] • Stretching, short walk during work breaks

[0203] Lunch example:

[0204] • Salad (e.g., containing leaf lettuce, tomatoes, cucumbers, avocados, and carrots). Accompanied by dressing with olive oil and vinegar

[0205] • Whole grain bread, accompanied by hummus

[0206] • Water or herbal tea

[0207] Afternoon activity example:

[0208] • Work (incorporate standing work)

[0209] • Walk for about 15 minutes after lunch

[0210] Snack example:

[0211] • Nuts (e.g., almonds or walnuts) with fruit (e.g., apples or strawberries)

[0212] Dinner example:

[0213] • Grilled salmon

[0214] • Stir-fried vegetables (e.g., containing broccoli, carrots, bell peppers, and zucchini)

[0215] • Kimchi

[0216] • Water or herbal tea

[0217] Exercise example:

[0218] • Walk for 30 minutes or jog after work

[0219] • Muscle training twice a week (20 to 30 minutes; for example, push-ups, squats, dumbbell exercises)

[0220] Examples of activities before going to bed:

[0221] • Warm herbal tea (chamomile, etc.)

[0222] • Reading, meditation, viewing of relaxing music

[0223] • Control the use of electronic devices

[0224] Sleep

[0225] • Keep the bedtime constant and ensure 7 to 8 hours of sleep.

[0226] Upon receiving the above health advice, the user B makes the following responses.

[0227] • Banana allergy, so cannot eat bananas

[0228] • Lunch wants to eat noodles

[0229] • Work is office work, and muscle exercise is difficult

[0230] • Wants to drink beer at dinner

[0231] • Before going to bed, does not read a book, and wants to consider a game or the hobby of fishing

[0232] The above responses can be made, for example, by adding a check mark in the response column corresponding to each of the examples of activities on the display device 10.

[0233] The above responses are registered in the DB as information characterizing the user B.

[0234] If the information related to the above responses is added to the input information, the large-scale language model updates the health advice, for example, as follows.

[0235] Examples of breakfast:

[0236] • Oatmeal. Add fruit (strawberries, blueberries).

[0237] • Green tea or iced coffee

[0238] Examples of activities in the morning:

[0239] • Commute on foot (for example, accelerated walking, walking stairs)

[0240] • Stretching, simple exercises for posture improvement (can be performed around a desk) during breaks at work

[0241] Lunch example:

[0242] • Chinese cold noodles (e.g., including cold noodles, cucumber, tomato, ham, egg, radish, and sesame paste)

[0243] • Maccha or water

[0244] Afternoon activity example:

[0245] • Light stretching or walking about 15 minutes after lunch

[0246] • To prevent eye strain, regularly leave the desk to rest the eyes

[0247] Snack example:

[0248] • A mixture of nuts and dried fruits (e.g., including almonds, raisins, and cranberries)

[0249] Dinner example:

[0250] • Grilled chicken or fish (e.g., salmon, etc.)

[0251] • Stir-fried vegetables (e.g., including cauliflower, carrots, bell peppers, and zucchini)

[0252] • Kimchi as a side dish

[0253] • 1 can of beer. However, by choosing low-alcohol or non-alcoholic wine

[0254] Relaxation after dinner example:

[0255] • Game, fishing preparation

[0256] • Perform light stretching or deep breathing

[0257] Sleep

[0258] • Keep the bedtime constant, and ensure 7-8 hours of sleep.

[0259] The updated health advice reflects the user B's reaction to the initial health advice, and is more in line with the user B's preferences and lifestyle.

[0260] Alternatively, if the user B has further reactions, the input to the large-scale language model and the update of the health advice can also be repeated, further improving the accuracy of the health advice.

[0261] The accumulated sensing data, the determination result of the health state, and the health improvement program are stored in the storage device 50, and are shared among medical care-related persons related to the user H with the permission of the user H. Thus, it is expected that the diagnosis and treatment by a doctor are more appropriately and promptly performed. Such an effect is expected not only in face-to-face medical care but also in a remote medical care scenario, and more substantial medical care collaboration is realized.

[0262] In addition, by sharing the health information of the user H among nursing-related persons, it is expected that more appropriate and prompt nursing assistance is provided.

[0263] Of course, the sensing data and the determination result of the health state can also be used as a care system and an abnormality detection system for the user H.

[0264] In addition, the collected biological data and the determination result of the health state are utilized in the administration and regional society with the permission of the user H, and thus it is possible to reflect the policy for forming a community in which people live more comfortably.

[0265] The display device 10 and the other device 30 are provided in the facility 70, and thus there is no trouble in attaching and detaching like a wearable sensor. Further, since the sensor group is hidden from the eyes of the user H, the user H does not recognize the presence of the sensor 10. Thus, it is possible to acquire the biological data of the user H on a daily basis, and it is possible to grasp the health state with high reliability.

[0266] The above describes representative embodiments of the present application, but the present application is not limited to these, and various design modifications can be made, which are also included in the present application.

[0267] Industrial Applicability

[0268] According to the present application, it is possible to effectively use biological data obtained in the daily life of a user to maintain and improve the health of the user. The effective use of the biological data includes support for the user, and information collaboration with medical care, nursing, and the like.

[0269] Explanation of Symbols

[0270] 1 System

[0271] 10 Display device

[0272] 20 Model generation device

[0273] 40 Output device

[0274] 50 Storage device

[0275] 60 Suggestion generation device

[0276] H User

Claims

1. A system, characterized in that, It is a system used to provide health advice to users. include: A structure, located within the facility, includes sensors that continuously collect biological data from the user; The analysis device is configured to analyze the biological data collected in the sensor and determine the health status of the user. A program generation apparatus configured to use the health status determination result to generate a health enhancement program corresponding to the user; and The suggestion generation device is configured to input the determined health status, the generated health enhancement program, and additional information related to the user into a large-scale language model to obtain health suggestions for the user.

2. The system according to claim 1, characterized in that, The additional information includes at least one of the following: basic information of the user, including at least one of age, gender, and nationality; biological information of the user, including at least one of height, weight, body composition, basal metabolic rate, and blood pressure; external information of at least one of financial and economic status, logistics, season, trend, temperature, and humidity; and intrinsic information of the user, including at least one of interests and hobbies, preferences for food and drink, allergies, physical characteristics, personal or family economic status, and mood on that day.

3. The system according to claim 1, characterized in that, The additional information includes the user's response to the health advice.

4. The system according to claim 1, characterized in that, The biological data includes at least one of the following: facial expression, heart rate, oxygen saturation, carbon dioxide exhalation, surface body temperature, deep body temperature, skin protein analysis, body composition, autonomic nervous system, HbA1c, body water content, foot pressure distribution, gait posture, gait speed, changes in joint mobility, center of gravity shift, activity level, electromyography, electrocardiography, brain waves, standing and sitting posture, body temperature, blood pressure, blood flow, heart rate, respiration, sweating, eye movements, sleep duration, amount and timing of excretion, blood composition, urine composition, saliva composition, intraoral images, and fecal composition.

5. The system according to claim 1, characterized in that, The analytical device is configured to output the user's health status from the collected biological data using a predictive model. The predictive model is obtained by using human biological data containing at least one of the following: facial expression, heart rate, oxygen saturation, carbon dioxide exhalation, surface body temperature, deep body temperature, skin protein analysis, body composition, autonomic nervous system, HbA1c, body water content, foot pressure distribution, walking posture, walking speed, changes in joint mobility, center of gravity shift, activity level, electromyography, electrocardiography, brain waves, standing and sitting posture, body temperature, blood pressure, blood flow, heart rate, respiration, sweating, eye movements, sleep duration, amount and timing of excretion, blood components, urine components, saliva components, images of the oral cavity, and fecal components, as well as the user's health status including energy consumption and exercise effects, as teacher data. This is obtained by machine learning on a learning model that takes human biological data as input and human health status as output.

6. The system according to claim 1, characterized in that, The program generation unit is configured as follows: The biological characteristics of the user are extracted by comparing the collected biological data with the prescribed evaluation indicators; Using an evaluation model, a health improvement program for the user is generated from the collected biological data. The evaluation model is obtained by machine learning a learning model that takes biological data as input and evaluates the health improvement program as output, using prescribed evaluation indicators and evaluation values ​​representing the effectiveness of the health improvement program as teacher data. The extracted biological features and the generated health-enhancing program are output.

7. A method, characterized in that, This is a method used to provide health advice to users, in which the computer performs the following steps: The user's health status is determined by analyzing the biometric data collected by sensors in structures installed within the facility. The results of the health status assessment are used to generate a health enhancement program corresponding to the user. The determined health status, the generated health enhancement program, and additional information related to the user are input into a large-scale language model to obtain health recommendations for the user.

Images