System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving ambiguous health-related information input by a user; means for analyzing the received ambiguous health-related information using a generative model; means for identifying and recommending an appropriate expert based on an analysis result obtained by the analyzing means; means for collecting information after diagnosis of a patient and using the collected information as feedback for learning of the generative model; and means for automatically generating a record based on the analysis result and the collected information.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] In the past, when a patient complained of vague symptoms, it was difficult for inexperienced doctors or specialists in other fields to make an accurate diagnosis. This increased the risk of inappropriate treatment, which could ultimately have a negative impact on the patient's health. Another problem was that doctors spent a lot of time completing documentation after treatment, which reduced work efficiency. The present invention aims to solve these problems and improve diagnostic accuracy and work efficiency in medical settings. [Means for solving the problem]

[0005] The present invention solves the above problems by providing a system including: means for receiving ambiguous health-related information input by a user; analysis means using a generative model to analyze the received ambiguous information; means for identifying and recommending an appropriate specialist based on the analysis results obtained by the analysis means; means for collecting post-diagnosis information from the patient and using the collected information as feedback to train the generative model; and means for automatically generating records based on the analysis results and the collected information. This system enables appropriate diagnosis of ambiguous symptoms and quickly guides patients to the most appropriate specialist. Furthermore, by reflecting post-diagnosis feedback in the generative model, continuous improvement in diagnostic accuracy is ensured. Furthermore, the automatic generation of medical records reduces the workload of doctors.

[0006] "User" refers to a person who inputs health information and uses the system.

[0007] "Ambiguous health information" refers to information that does not specifically describe symptoms but uses vague expressions that indicate poor health.

[0008] "Generative model" refers to a machine learning algorithm or artificial intelligence model used to analyze incoming ambiguous information and identify potential diagnoses.

[0009] "Analysis means" refers to the part that has the function of analyzing ambiguous health-related information using a generative model.

[0010] "Experts" refers to doctors and medical professionals with medical expertise in a particular field who are identified based on the analysis results.

[0011] The "recommending means" refers to the part that has the function of guiding the user to an appropriate expert based on the analysis results obtained by the analysis means.

[0012] "Feedback" refers to collecting information about changes in a patient's symptoms or new symptoms after diagnosis and using it as training data for the generative model.

[0013] "Learning a generative model" refers to the process by which a generative model incorporates new data through feedback to improve its analytical accuracy.

[0014] "Means for generating records" refers to the part that has the function of automatically creating medical records and charts based on the analysis results and collected feedback.

[0015] The "system" refers to a comprehensive platform that integrates the above means to efficiently connect patients and doctors. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11]FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, not only improving the accuracy of diagnoses but also increasing the work efficiency of doctors.

[0038] System Overview

[0039] This system is primarily composed of a user terminal, a server, a generative model, and an expert recommendation means. The user terminal is responsible for sending health information entered by the user to the server. The server receives this information and analyzes it using the generative model. Based on the analysis results, the server recommends an appropriate expert and collects subsequent feedback information to train the generative model.

[0040] Program processing explanation

[0041] User behavior

[0042] 1. Symptom input: The user inputs vague symptoms related to their health into an input form on the user terminal. For example, they input vague information such as "my stomach hurts" or "I feel a bit sick."

[0043] 2. Sending information: After entering the information, the user clicks the "Send" button to send the information to the server.

[0044] Server-side operation

[0045] 1. Information reception: The server receives vague health information sent from the user terminal.

[0046] 2. Analysis: The received information is analyzed using a generative model, which uses natural language processing techniques to convert ambiguous expressions into specific diagnostic candidates.

[0047] 3. Expert recommendation: Based on the analysis results, an appropriate expert (e.g., a gastroenterologist or gastroenterologist) is identified from the database and the information is sent to the user's terminal.

[0048] Generate medical records

[0049] 1. Information storage: The server stores the input information from the user and the analysis results in a database.

[0050] 2. Medical record generation: Medical records (medical records) are automatically generated based on the analysis results and sent to recommended specialists.

[0051] Gathering feedback and learning

[0052] 1. Feedback input: After the diagnosis, the user again inputs how their condition has changed and sends it to the server. For example, information such as "symptoms have improved" or "new symptoms have appeared."

[0053] 2. Learning update: Based on the received feedback, the server updates the training dataset of the generative model to improve the accuracy of the diagnostic algorithm.

[0054] Specific examples

[0055] Example 1: Patient complaining of abdominal pain

[0056] 1. User terminal: The patient types "My stomach hurts" and sends it.

[0057] 2. Server: Receives input data and analyzes it with the generative model.

[0058] 3. Server: The generative model converts the information into specific symptoms such as "stomachache," "pain in the lower right abdomen," and "nausea."

[0059] 4. Server: Retrieves gastroenterologists and gastroenterologists from the database and sends them to the user terminal.

[0060] 5. User terminal: The patient selects a doctor from the recommended list.

[0061] 6. Server: Receives the selected information, automatically generates a medical record, and sends it to the doctor.

[0062] 7. User terminal: After diagnosis, the patient enters and submits "Symptoms are not improving."

[0063] 8. Server: Receives feedback and updates the generative model.

[0064] This concludes the description of the embodiment of the present invention. This system makes it possible to accurately analyze a patient's vague symptoms and recommend the most appropriate specialist. It also improves the work efficiency of doctors, enabling faster and more reliable medical care.

[0065] The processing flow will be explained below.

[0066] Step 1:

[0067] The user enters vague health symptoms into the input form on the user terminal and clicks the "Submit" button.

[0068] Step 2:

[0069] The device sends vague health information entered by the user to the server as an HTTP POST request.

[0070] Step 3:

[0071] The server receives vague health information from the user terminal and temporarily stores it in a database.

[0072] Step 4:

[0073] The server passes the ambiguous information it receives to the natural language processing engine and generative model, and begins the analysis process.

[0074] Step 5:

[0075] A generative model analyzes ambiguous information and identifies likely specific diagnoses (e.g., "stomach pain," "pain in the right lower abdomen," "nausea").

[0076] Step 6:

[0077] Based on the analysis results, the server retrieves relevant specialists (e.g., gastroenterologists, gastroenterologists, etc.) from the database.

[0078] Step 7:

[0079] The server ranks the acquired expert information and sends it to the user's terminal as an HTTP response.

[0080] Step 8:

[0081] The terminal displays the expert information received from the server in list form on the user interface.

[0082] Step 9:

[0083] The user selects the desired expert from the displayed list of experts and clicks the "Select" button.

[0084] Step 10:

[0085] The device sends the information of the selected expert to the server as an HTTP POST request.

[0086] Step 11:

[0087] The server receives the selections from the user and stores them in a database.

[0088] Step 12:

[0089] The server generates a medical record based on the analysis results and the selected expert information.

[0090] Step 13:

[0091] The server sends the generated medical record to the corresponding specialist's terminal as an HTTP response.

[0092] Step 14:

[0093] After the diagnosis, the user enters the progress of their symptoms and any new symptoms into the feedback form on the user's device and clicks the "Submit" button.

[0094] Step 15:

[0095] The terminal sends the input feedback data to the server as an HTTP POST request.

[0096] Step 16:

[0097] The server receives the feedback data and stores it in a database.

[0098] Step 17:

[0099] The server analyzes the received feedback data and provides it as training data for the generative model.

[0100] Step 18:

[0101] The generative model incorporates new feedback data, updates the learning algorithm, and improves diagnostic accuracy.

[0102] The above is a specific processing flow for making a diagnosis based on vague information about the user's health and improving accuracy using feedback.

[0103] Example 1

[0104] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0105] In conventional health support systems, the health information entered by users is vague, making it difficult to make accurate diagnoses or recommend appropriate specialists. Furthermore, feedback information after diagnosis is not fully utilized, making it difficult to improve the accuracy of the system. Furthermore, the creation of medical records is not automated, increasing the workload of medical professionals.

[0106] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0107] In this invention, the server includes a means for receiving ambiguous health-related information input by a user, an analysis means using a generative model to analyze the received ambiguous health-related information, and a means for identifying and recommending an appropriate specialist based on the analysis results obtained by the analysis means, thereby making it possible to convert ambiguous health-related information into specific diagnosis candidates and recommend an appropriate specialist.

[0108] A "user terminal" is a device that allows a user to input health-related information and transmit that information to a server.

[0109] The "server" is the central computer of the system that receives information sent from user terminals, analyzes it, and recommends experts.

[0110] A "generative model" is an algorithm that uses natural language processing techniques to analyze ambiguous information received and convert it into specific diagnostic candidates.

[0111] "Analysis means" refers to the process of using a generative model to analyze received ambiguous information and generate specific diagnosis candidates.

[0112] The "expert recommendation means" is a means for identifying an appropriate expert from a database based on the analysis results and providing that information.

[0113] "Information storage means" refers to the process of storing user input information and analysis results in a database.

[0114] "Medical record generation means" refers to the process of automatically generating a medical record based on the analysis results and sending it to the recommended specialist.

[0115] "Feedback input means" refers to a process in which the user re-enters information about changes in his or her condition after diagnosis and sends that information to the server.

[0116] The "learning update means" refers to the process of adding received feedback information to the learning dataset of the generative model to improve the accuracy of the model.

[0117] "Ambiguous information" refers to information entered by a user in vague terms, without clearly describing specific symptoms or conditions.

[0118] "Analysis results" refer to specific diagnosis candidates and expert recommendation information analyzed by the generative model.

[0119] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, not only improving the accuracy of diagnoses but also increasing the work efficiency of doctors.

[0120] System Overview

[0121] This system is primarily composed of a user terminal, a server, a generative model, and an expert recommendation means. The user terminal is responsible for sending health information entered by the user to the server. The server receives this information and analyzes it using the generative model. Based on the analysis results, the server recommends an appropriate expert and collects subsequent feedback information to train the generative model.

[0122] User behavior

[0123] The user enters vague symptoms related to their health into an input form on the user terminal. Specifically, they enter vague information such as "my stomach hurts" or "I feel a bit sick." Then, the user clicks the "send" button to send the information to the server.

[0124] Server-side operation

[0125] Receiving information

[0126] The server receives vague health information sent from the user's device. The server's API receives the request from the device and adds the text data to a processing queue.

[0127] Analysis means

[0128] The server analyzes the received information using a generative model. This model uses natural language processing technology to convert vague expressions into specific diagnosis candidates. For example, if an input says "my stomach hurts," the generative model converts that into specific symptoms such as "stomachache," "pain in the right lower abdomen," and "nausea."

[0129] Expert recommendation

[0130] Based on the analysis results, the server identifies an appropriate expert from the database. The expert's information is returned to the user's terminal. The server sends the analysis results and the expert's information to the terminal in response to the request.

[0131] Generate medical records

[0132] Information storage

[0133] The server stores the user's input information and analysis results in a database, which then becomes historical data that can be referenced later.

[0134] Medical record generation

[0135] The server automatically generates a medical record (chart) based on the analysis results, which is then sent to the recommended specialist.

[0136] Gathering feedback and learning

[0137] Feedback Input

[0138] After the diagnosis, the user again inputs how their condition has changed and sends it to the server. For example, they input feedback information such as "symptoms have improved" or "new symptoms have appeared."

[0139] Learning Update

[0140] The server uses the received feedback to update the training dataset of the generative model, thereby improving the accuracy of the diagnostic algorithm.

[0141] Specific examples

[0142] Example 1: Patient complaining of abdominal pain

[0143] 1. Symptom input:

[0144] The user inputs "My stomach hurts" into the user terminal and sends it.

[0145] 2. Receiving information:

[0146] The server receives this input data and analyzes it with the generative model.

[0147] 3. Analysis method:

[0148] The generative model converts the input "my stomach hurts" into specific symptoms such as "stomachache," "pain in the lower right abdomen," and "nausea."

[0149] 4. Expert recommendation:

[0150] The server retrieves gastroenterology and gastroenterology specialists from a database and transmits them to the user terminal.

[0151] 5. Expert Selection:

[0152] The user selects a doctor from a list of recommendations.

[0153] 6. Medical record generation:

[0154] The server receives the selection information, automatically generates a medical record, and sends it to the doctor.

[0155] 7. Feedback Input:

[0156] After the diagnosis, the user enters "Symptoms are not improving" and submits the message.

[0157] 8. Learning Update:

[0158] The server receives the feedback and updates the generative model.

[0159] Prompt Sentence Examples

[0160] For the input "My stomach hurts," the generative AI model generates the following analysis result:

[0161] Entered information: "My stomach hurts"

[0162] Analysis results: "Stomach pain," "pain in the lower right abdomen," "nausea"

[0163] Appropriate specialist: "Gastroenterologist" "Gastroenterologist"

[0164] This makes it possible to turn vague symptoms into specific candidate diagnoses and identify appropriate specialists.

[0165] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0166] Step 1: Enter your symptoms

[0167] The user inputs vague information about their symptoms into an input form on the terminal, such as "my stomach hurts" or "I feel a bit sick."

[0168] Specifically, the user uses the terminal keyboard to input symptoms into a text box.

[0169] Input: Vague health information entered by the user.

[0170] Output: Symptom information entered into the user's terminal.

[0171] Step 2: Send information

[0172] The user checks the information they entered and clicks the "Send" button, which causes the user's device to send the data to the server.

[0173] Specifically, the terminal sends the input text data as an HTTP request to the server's API.

[0174] Input: The symptom information the user entered and clicked submit.

[0175] Output: Symptom information sent to the server.

[0176] Step 3: Receiving information

[0177] The server receives ambiguous health information transmitted from the user terminal.

[0178] Specifically, the server's API receives a request from the terminal and adds the text data to a processing queue.

[0179] Input: Symptom information sent from the user terminal.

[0180] Output: Symptom information received.

[0181] Step 4: Analysis methods

[0182] The server analyzes the received information using a generative model, which uses natural language processing techniques to convert ambiguous expressions into specific candidate diagnoses.

[0183] Specifically, the generative AI model analyzes input data and converts vague expressions into specific diagnostic candidates such as "stomachache," "pain in the lower right abdomen," and "nausea."

[0184] Input: Received symptom information.

[0185] Output: A list of specific diagnostic candidates.

[0186] Step 5: Expert Recommendations

[0187] Based on the analysis results, the server identifies appropriate experts from the database, and the recommended expert information is sent to the user terminal.

[0188] Specifically, the server compares the analysis results with a list of experts, generates a list of appropriate experts, and returns it to the user terminal.

[0189] Input: A specific list of diagnostic candidates.

[0190] Output: A list of recommended experts.

[0191] Step 6: Save your information

[0192] The server stores the input information from the user and the analysis results in a database.

[0193] Specifically, the server stores the input information and analysis results in a database table.

[0194] Input: Information entered by the user, analysis results.

[0195] Output: Saved data.

[0196] Step 7: Create a medical record

[0197] The server automatically generates a medical record (chart) based on the analysis results and sends it to the recommended specialist.

[0198] Specifically, the server generates medical records in accordance with the medical record format and sends them to specialists via email or a dedicated system.

[0199] Input: User input information, analysis results.

[0200] Output: Generated medical chart, medical record sent to specialist.

[0201] Step 8: Enter your feedback

[0202] After the diagnosis, the user again inputs any changes in their condition and sends it to the server. For example, they can input information such as "symptoms have improved" or "new symptoms have appeared."

[0203] As a specific operation, the user inputs feedback information using the keyboard of the terminal and clicks the send button.

[0204] Input: User feedback information.

[0205] Output: Feedback information sent by the user.

[0206] Step 9: Learning Update

[0207] The server adds the received feedback information to the training dataset of the generative model and retrains the model.

[0208] Specifically, the server adds feedback information to the dataset and retrains the generative AI model to improve its accuracy.

[0209] Input: Feedback information from the user.

[0210] Output: An updated generative model.

[0211] (Application example 1)

[0212] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0213] Conventional systems for analyzing ambiguous health-related information are limited to the function of recommending experts, and lack mechanisms for effectively collecting and learning from feedback information after recommendations and improving the accuracy of generative models. Furthermore, there is a lack of systems for quickly recommending appropriate products and experts for customer support in brick-and-mortar stores, making it difficult to improve customer satisfaction and ensure appropriate health management in brick-and-mortar stores.

[0214] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0215] In this invention, the server includes: means for receiving ambiguous health-related information entered by a user; analysis means using a generative model to analyze the received ambiguous health-related information; means for identifying and recommending an appropriate expert or product based on the analysis results obtained by the analysis means; means for collecting information after the customer has used the recommended expert or product and using the collected information as feedback to train the generative model; and means for automatically generating records based on the analysis results and the collected information. This enables accurate analysis of ambiguous health information in physical stores and rapid recommendation of appropriate experts and products, thereby improving customer satisfaction and streamlining health management in stores.

[0216] "User" means a customer who uses the System to input their health information.

[0217] "Ambiguous health information" refers to information about a condition in which the user is experiencing some kind of discomfort or symptom but cannot specifically describe it.

[0218] "Means for receiving" refers to the function by which the server receives vague health-related information sent from the user terminal.

[0219] "Analysis means" refers to the function that converts received ambiguous health information into specific expert and product recommendations using a generative model.

[0220] A "generative model" is an AI model that uses machine learning algorithms to analyze input information and generate specific diagnostic candidates and recommendations.

[0221] An "expert" is someone who has knowledge and skills in a particular field and can provide appropriate advice and diagnoses to users.

[0222] "Means for identifying and recommending" refers to a function that notifies users of appropriate experts and products based on the results obtained by the analysis means.

[0223] "Feedback" refers to information entered by users about the effects or changes they have experienced after using a recommended professional or product.

[0224] "Means used for learning" refers to the function of using collected feedback information as training data for a generative model to improve the accuracy of the model.

[0225] "Means for generating records" refers to a function that automatically creates medical records and advice records based on analysis results and feedback information.

[0226] "System" refers to a comprehensive mechanism that includes all of the above means and functions and is used to analyze vague information about a user's health and recommend appropriate experts and products.

[0227] The following describes in detail an embodiment of the "Smart Health Assistant" of the present invention. The present invention is a system that identifies the most suitable specialist or product based on vague health information entered by the user, makes more accurate recommendations, and collects feedback after the recommendation to improve the accuracy of the model.

[0228] System Overview

[0229] This system consists of a user terminal, a server, a generative model, and a database. The user terminal is responsible for sending health information entered by the user to the server. The server receives the information and analyzes it using the generative model. Based on the analysis results, the server recommends appropriate experts or products, and then collects subsequent feedback information to train the generative model.

[0230] User behavior

[0231] 1. Symptom input: The user inputs vague symptoms related to their health into an input form on the user device (e.g., smart glasses). For example, they input vague information such as "I have a headache" or "My skin is itchy."

[0232] 2. Sending information: After entering the information, the user clicks the "Send" button to send the information to the server.

[0233] Server-side operation

[0234] 1. Information reception: The server receives vague health information sent from the user terminal.

[0235] 2. Analysis method: The received information is analyzed using a generative model (e.g., OpenAI GPT-4). This generative model uses natural language processing technology to convert ambiguous expressions into specific diagnosis candidates and recommendations.

[0236] 3. Expert recommendation: Based on the analysis results, the appropriate expert (pharmacist, health advisor, etc.) or product (painkiller, moisturizing cream, etc.) is identified from the database and the information is sent to the user's device.

[0237] Generate medical records

[0238] 1. Information storage: The server stores the input information from the user and the analysis results in a database.

[0239] 2. Medical record generation: A medical record (medical record) or advice record is automatically generated based on the analysis results and sent to the recommended specialist.

[0240] Gathering feedback and learning

[0241] 1. Feedback input: The user re-enters information about the effects and changes in condition after using the recommended specialist or product, and sends it to the server. For example, information such as "Symptoms improved" or "Insufficient effect."

[0242] 2. Learning update: Based on the received feedback, the server updates the training dataset of the generative model to improve the accuracy of the diagnosis and recommendation algorithm.

[0243] Specific examples

[0244] Example 1: A customer complaining of a headache

[0245] 1. Smart Glasses: The customer types "I have a headache" and submits.

[0246] 2. Server: Receives input data and analyzes it with the generative model.

[0247] 3. Server: The generative model converts the data into specific diagnosis candidates such as "tension headache" or "stress."

[0248] 4. Server: Retrieves the appropriate pharmacist or painkiller from the database and sends it to the user's smart glasses.

[0249] 5. Smart Glasses: Customers select a pharmacist or product from a list of recommendations.

[0250] 6. Server: Receives the selected information, automatically generates a medical record, and sends it to the pharmacist.

[0251] 7. Smart glasses: The customer types "my symptoms have improved" and submits.

[0252] 8. Server: Receives feedback and updates the generative model.

[0253] Prompt Sentence Examples

[0254] "Enter any vague health symptoms you have. For example, I have a headache, my skin is itchy, etc."

[0255] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0256] Step 1: Enter your symptoms

[0257] Users enter vague health symptoms into a form on the smart glasses in a physical store. Examples of information include "I have a headache" or "My skin is itchy." This input is the starting point for the system, which then identifies the user's complaint in concrete terms.

[0258] Step 2: Send information

[0259] After inputting the information, the user clicks the "Send" button. The smart glasses then send the input information to the store's server. The transmitted data includes the symptom information entered by the user. The server then passes the received information to the next analysis step.

[0260] Step 3: Receiving information

[0261] The server receives the vague health information sent by the smart glasses, stores the received data in preparation for analysis, and passes it on to the next processing step.

[0262] Step 4: Analysis

[0263] The server analyzes the received information using a generative model (e.g., OpenAI GPT-4). The input includes the symptom information submitted by the user. The generative model analyzes the ambiguous information and generates specific diagnosis candidates and recommendations. The output includes specific diagnosis candidates such as "tension headache" and "stress."

[0264] Step 5: Expert or product recommendation

[0265] The server identifies appropriate experts (pharmacists, health advisors, etc.) or products (painkillers, moisturizing creams, etc.) from a database based on the analysis results obtained by the generative model. The input includes the analysis results, and the server selects the best recommendation destination based on these results. The output includes a list of recommended experts and products.

[0266] Step 6: Submit your recommendation

[0267] The server sends the identified expert or product information to the user's smart glasses. The input includes the identified recommendation information, which is sent to notify the user. The output includes a recommendation list that is displayed on the user's device.

[0268] Step 7: Enter your feedback

[0269] After receiving expert advice or using the product, the user again enters feedback about the effects and changes in their physical condition into the smart glasses' input form. For example, feedback such as "Symptoms improved" or "Not effective enough" is entered. This input information is used for subsequent learning.

[0270] Step 8: Send your feedback

[0271] After entering the feedback information, the user clicks the "Send" button. The smart glasses then send this feedback information to the store's server. The transmitted data includes the feedback information. The server then passes the received information to the next learning process.

[0272] Step 9: Receiving feedback

[0273] The server receives the feedback information sent from the smart glasses, stores the received data in a training dataset, and passes it on to the next processing step.

[0274] Step 10: Retraining the generative model

[0275] The server retrains the generative model based on the received feedback. The input includes the feedback, which improves the accuracy of the generative model. The output includes the updated generative model.

[0276] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0277] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, improving the accuracy of diagnosis, and also recognizes the user's emotions and reflects them in the selection of the recommended specialist.

[0278] System Overview

[0279] This system is primarily composed of a user terminal, a server, a generative model, an emotion engine, and an expert recommendation means. The user terminal is responsible for sending health information and natural language emotion expressions entered by the user to the server. The server receives this information and analyzes it using the generative model and emotion engine. Based on the analysis results, the server recommends an appropriate expert and collects subsequent feedback information to train the generative model.

[0280] Program processing explanation

[0281] User behavior

[0282] 1. Symptom and emotion input: The user inputs vague symptoms related to their health and the emotions associated with them into an input form on the user terminal. For example, they input information such as "I have a stomachache and I'm worried" or "I feel a bit sick and worried."

[0283] 2. Sending information: After inputting, the user clicks the "Send" button to send the symptom and emotion information to the server.

[0284] Server-side operation

[0285] 1. Information reception: The server receives vague health information and emotion information sent from the user terminal and temporarily stores them in a database.

[0286] 2. Symptom analysis: The received information is passed to a natural language processing engine and a generative model to begin the symptom analysis process. This generative model uses natural language processing techniques to convert ambiguous expressions into specific diagnostic candidates.

[0287] 3. Emotion analysis means: At the same time, the emotion engine analyzes the received emotion information to grasp the user's mental state.

[0288] 4. Expert recommendation: Based on the results of symptom analysis and emotion analysis, relevant experts (e.g., gastroenterologists) are retrieved from a database, and the most suitable expert is identified taking into account emotion information, and this information is sent to the user's device.

[0289] Generate medical records

[0290] 1. Information storage: The server stores the input information from the user and the analysis results in a database.

[0291] 2. Medical record generation: Medical records (medical records) are automatically generated based on the analysis results and sent to recommended specialists.

[0292] Gathering feedback and learning

[0293] 1. Feedback input: After the diagnosis, the user again inputs how their condition has changed and sends it to the server. For example, they input feedback information such as "Symptoms have improved" or "New symptoms have appeared."

[0294] 2. Learning update: Based on the received feedback, the server updates the training dataset of the generative model to improve the accuracy of the diagnostic algorithm.

[0295] Specific examples

[0296] Example 1: Patient complaining of abdominal pain

[0297] 1. User terminal: The patient types, "I have a stomachache and I'm worried" and sends it.

[0298] 2. Server: Receives input data and analyzes symptom data using a generative model and emotion data using an emotion engine.

[0299] 3. Server: The generative model converts the symptoms into specific ones such as "stomachache," "pain in the lower right abdomen," and "nausea," and the emotion engine identifies emotions such as "anxiety" and "worry."

[0300] 4. Server: Retrieves gastroenterologists and gastroenterologists from a database, identifies the appropriate doctor for the patient, taking into account their emotional state, and sends the information to the user's terminal.

[0301] 5. User terminal: The patient selects a doctor from the recommended list.

[0302] 6. Server: Receives the selected information, automatically generates a medical record, and sends it to the doctor.

[0303] 7. User terminal: After diagnosis, the patient enters and submits "Symptoms are not improving."

[0304] 8. Server: Receives feedback and updates the generative model.

[0305] This concludes the description of the embodiment of the present invention. This system makes it possible to accurately analyze a patient's vague symptoms and emotions and recommend the most appropriate specialist. It also improves the work efficiency of doctors, enabling faster and more reliable medical care.

[0306] The processing flow will be explained below.

[0307] Step 1:

[0308] The user enters vague health symptoms and associated feelings into an input form on the user's device and clicks the "Submit" button.

[0309] Step 2:

[0310] The terminal sends the ambiguous health information and emotion information input by the user to the server as an HTTP POST request.

[0311] Step 3:

[0312] The server receives the information sent from the user terminal and temporarily stores it in a database.

[0313] Step 4:

[0314] The server passes the received ambiguous information to a natural language processing engine and a generative model, and begins the symptom analysis process.

[0315] Step 5:

[0316] A generative model analyzes ambiguous information and identifies likely specific diagnoses (e.g., "stomach pain," "pain in the right lower abdomen," "nausea").

[0317] Step 6:

[0318] The server passes the received emotion information to the emotion engine and starts emotion analysis processing.

[0319] Step 7:

[0320] The emotion engine analyzes the emotional information and identifies the user's mental state (e.g., "anxiety" or "worry").

[0321] Step 8:

[0322] The server retrieves relevant specialists (e.g., gastroenterologists, gastroenterologists, etc.) from a database based on the symptom analysis and emotion analysis results.

[0323] Step 9:

[0324] The server optimizes the expert information it acquires, taking into account the results of sentiment analysis, and sends it to the user's device as an HTTP response.

[0325] Step 10:

[0326] The terminal displays the expert information received from the server in list form on the user interface.

[0327] Step 11:

[0328] The user selects the desired expert from the displayed list of experts and clicks the "Select" button.

[0329] Step 12:

[0330] The device sends the information of the selected expert to the server as an HTTP POST request.

[0331] Step 13:

[0332] The server receives the selections from the user and stores them in a database.

[0333] Step 14:

[0334] The server generates a medical record based on the symptom analysis results and the selected specialist information.

[0335] Step 15:

[0336] The server sends the generated medical record to the corresponding specialist's terminal as an HTTP response.

[0337] Step 16:

[0338] After the diagnosis, the user enters the progress of their symptoms and any new symptoms into the feedback form on the user's device and clicks the "Submit" button.

[0339] Step 17:

[0340] The terminal sends the input feedback data to the server as an HTTP POST request.

[0341] Step 18:

[0342] The server receives the feedback data and stores it in a database.

[0343] Step 19:

[0344] The server analyzes the received feedback data and provides it as training data for the generative model.

[0345] Step 20:

[0346] The generative model incorporates new feedback data, updates the learning algorithm, and improves diagnostic accuracy.

[0347] The above is a specific processing flow for making a diagnosis based on vague information and emotions about the user's health and improving accuracy using feedback.

[0348] Example 2

[0349] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0350] Conventional medical systems have difficulty accurately analyzing the vague health information entered by patients and recommending appropriate specialists. Furthermore, patients' emotional state is rarely reflected in diagnoses or specialist selection, potentially reducing diagnostic accuracy and patient satisfaction. Furthermore, there is a lack of mechanisms for effectively utilizing feedback information after diagnosis, which has delayed improvements to the medical system as a whole.

[0351] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0352] In this invention, the server includes: means for receiving ambiguous health-related information and emotional expressions input by a user; analysis means using a generative model and an emotion analysis engine to analyze the received ambiguous health-related information and emotional expressions; means for identifying and recommending an appropriate specialist based on the symptom analysis results and emotion analysis results obtained by the analysis means; means for collecting feedback information from patients after diagnosis and using the collected information to train the generative model and improve the accuracy of the algorithm; and means for automatically generating medical records based on the analysis results and collected information. This makes it possible to analyze ambiguous symptoms and emotional information and recommend the most appropriate specialist, improving patient diagnostic accuracy and satisfaction and promoting overall improvement of the medical system.

[0353] "Ambiguous health information" refers to health information that is expressed in terms or conditions that do not clearly indicate a specific diagnosis or symptom.

[0354] "Emotional expressions" are words or sentences that express the feelings or psychological state that the user is currently experiencing.

[0355] "Means for receiving" refers to a mechanism for acquiring information sent by a user and storing it in a form that can be used within the system.

[0356] A "generative model" is a machine learning algorithm that uses natural language processing techniques to convert vague expressions into specific diagnostic candidates.

[0357] An "emotion analysis engine" is software or an algorithm that analyzes emotional expressions contained in natural language and identifies a user's emotional state.

[0358] "Analysis means" refers to various processing devices within the system for analyzing received ambiguous health-related information and emotional expressions using generative models and emotion analysis engines.

[0359] The "means for identifying and recommending experts" is a mechanism for selecting appropriate medical experts based on the analysis results and presenting that information to the user.

[0360] The "means for collecting feedback information" is a mechanism for collecting information such as changes in condition and new symptoms sent by the user after diagnosis and incorporating it into the system.

[0361] The "means used to train the generative model" refers to a mechanism for updating the generative model based on collected feedback information and improving the accuracy of analysis from the next time onwards.

[0362] "Means for automatically generating medical records" refers to a system for automatically creating medical records (charts) based on the analysis results and collected feedback information.

[0363] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, improving the accuracy of diagnosis, and also recognizes the user's emotions and reflects them in the selection of the recommended specialist.

[0364] System Configuration

[0365] This system is mainly composed of a user terminal, a server, a generative model, a sentiment analysis engine, and an expert recommendation means.

[0366] User terminal operation

[0367] The user device serves to transmit the health information and emotional expressions entered by the user in natural language to the server. For example, a smartphone or computer application can function as a user device. The user uses these devices to input and transmit information such as "I'm worried because my stomach hurts."

[0368] Server Operation

[0369] The server is the central hardware that receives and analyzes information sent from user devices. The server has the following functions:

[0370] 1. Data reception and storage: The server receives vague health-related information and emotional expressions sent from the user terminal and stores them in a temporary database.

[0371] 2. Symptom Analysis: The server passes the received data to a generative model, which uses natural language processing techniques to convert vague expressions into specific diagnosis candidates. This generative model uses advanced natural language processing algorithms such as GPT-3 and BERT.

[0372] 3. Sentiment analysis: At the same time, the server passes the received emotional data to an emotion analysis engine to analyze the user's mental state. For example, tools such as Watson NLU or Hume AI can be used for the emotion analysis engine.

[0373] 4. Expert recommendation: Based on the analysis results, an appropriate expert is selected and related expert information is sent to the user's device. The expert recommendation involves retrieving appropriate expert information from the database and taking into account the user's emotional state.

[0374] Medical record generation and feedback

[0375] The server performs the following processes based on the analysis results and user feedback:

[0376] 1. Medical record generation: Medical records are automatically generated based on the analysis results and sent to the recommended specialists. This process improves the efficiency of doctors' work.

[0377] 2. Feedback collection: After the diagnosis, the user re-enters any changes in their condition and sends them to the server, including information such as "symptoms have improved" or "new symptoms have appeared."

[0378] 3. Learning and updating of generative model: The server updates the generative model based on the collected feedback information to improve the accuracy of the diagnostic algorithm.

[0379] Specific examples

[0380] For example, if a user types "I'm anxious because my stomach hurts" and submits it...

[0381] 1. User device: Enter and send information on the smartphone application.

[0382] 2. Server: Receives input data, analyzes symptom data using a generative model, and emotion data using an emotion analysis engine.

[0383] 3. Server: The generative model converts the symptoms into specific ones such as "stomachache" or "pain in the lower right abdomen," and the emotion analysis engine identifies emotions such as "anxiety" or "worry."

[0384] 4. Server: Retrieves gastroenterologists and gastroenterologists from a database and recommends the appropriate doctor to the patient, taking into account their emotional state.

[0385] 5. User terminal: The patient selects a doctor from the recommended list and receives treatment.

[0386] 6. Server: After diagnosis, the user inputs "symptoms do not improve" and submits it, which receives feedback and updates the generative model.

[0387] This concludes the description of the "Smart Health Link" system, which accurately analyzes vague symptoms and emotions and recommends appropriate specialists, thereby improving the accuracy of diagnoses and patient satisfaction.

[0388] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0389] Step 1: User Input

[0390] Users input vague information about their health and emotional expressions into an input form on their device, such as "I have a stomachache and I'm worried," into a smartphone or computer application.

[0391] Input: Ambiguous health-related information and emotional expressions

[0392] Output: An object that temporarily stores the input information

[0393] Step 2: Send data from the user device to the server

[0394] The user checks the information they have entered and clicks the "Send" button, which causes the user terminal to send the input data to the server.

[0395] Input: Information entered by the user and sent (vague health information and emotional expressions)

[0396] Output: HTTP request with the information sent

[0397] Step 3: Data received and stored by the server

[0398] The server receives the information sent from the user terminal and stores it in a temporary storage database.

[0399] Input: Information sent from the user's device (vague health-related information and emotional expressions)

[0400] Output: Input data stored in a database

[0401] Step 4: Symptom analysis by server

[0402] The server passes the received data to a generative model, which uses natural language processing techniques to convert ambiguous expressions into specific diagnosis candidates. The generative model uses GPT-3, BERT, and other models.

[0403] Input: Vague health information stored in a database

[0404] Output: Results converted into specific symptoms (e.g., "stomach pain," "pain in the lower right abdomen," etc.)

[0405] Step 5: Sentiment analysis on the server

[0406] The server passes the received emotional data to an emotion analysis engine, which may use Watson NLU or Hume AI, to analyze the user's mental state.

[0407] Input: Emotions stored in the database

[0408] Output: Analyzed emotion results (e.g., "anxiety," "worry," etc.)

[0409] Step 6: Server recommends experts

[0410] The server retrieves relevant experts from a database based on the results of symptom and emotion analysis, identifies the most suitable expert based on the analysis results, and sends that information to the user's device.

[0411] Input: Symptom analysis results and emotion analysis results

[0412] Output: Recommended expert information (e.g., "Gastroenterologist")

[0413] Step 7: Generate and send medical records

[0414] The server automatically generates a medical record based on the analysis results and the user's input data and sends it to the recommended specialist.

[0415] Input: Analysis results and user input data

[0416] Output: Automatically generated medical records (charts)

[0417] Step 8: Provide feedback

[0418] After the diagnosis, the user inputs any changes in their condition again, such as feedback information such as "symptoms have improved" or "new symptoms have appeared," and sends it to the server.

[0419] Input: Feedback information after diagnosis

[0420] Output: Feedback information sent to the server

[0421] Step 9: Update learning using feedback

[0422] Based on the received feedback, the server updates the training dataset for the generative model and improves the accuracy of the analysis algorithm.

[0423] Input: Collected feedback information

[0424] Output: Updated generative model and algorithm

[0425] This is the specific program processing flow of this system. This system makes it possible to accurately analyze vague health-related information and emotions and recommend the most appropriate health specialist.

[0426] (Application example 2)

[0427] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0428] Autonomous vehicles require constant monitoring of the health status of drivers and passengers, and prompt and appropriate medical treatment when an abnormality is detected. However, current systems lack the means to analyze health and emotional data in real time and recommend the most appropriate medical specialist. As a result, there is a risk that responses to abnormalities may be delayed, hindering the provision of prompt and appropriate medical care.

[0429] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0430] In this invention, the server includes: means for receiving ambiguous health-related information input by a user; means for analyzing the ambiguous information using a generative model; means for identifying and recommending an appropriate expert based on the obtained analysis results; means for collecting health data and emotion data using sensors and cameras installed in the vehicle; means for analyzing the collected health data and emotion data in real time and recommending an optimal expert when an abnormality is detected; means for collecting post-diagnosis information from the patient and using it as feedback to train the generative model; and means for automatically generating a record based on the analysis results and the collected information and transmitting it to an in-vehicle display or a user terminal. This enables constant monitoring of the health status of passengers in an autonomous vehicle and quickly recommending an appropriate expert when an abnormality is detected.

[0431] "User" means the driver and passengers in the vehicle who input health information into the system.

[0432] "Ambiguous health information" refers to information about a health condition entered by a user that does not specify specific symptoms, and includes information expressed in natural language.

[0433] A "generative model" is a model that uses machine learning algorithms to analyze ambiguous health information and generate specific diagnostic candidates.

[0434] The "analysis means" is a system element that uses a generative model to analyze received ambiguous health information and obtain a diagnosis.

[0435] "Experts" are medical professionals recommended based on diagnostic results and who have expertise in a particular medical field.

[0436] The "means for recommending an expert" is a system element that identifies an appropriate expert based on the results obtained by the analysis means and recommends the expert to the user.

[0437] "Sensors" refer to devices installed in vehicles that collect physiological data such as heart rate and body temperature.

[0438] The "camera" is a video capture device installed inside the vehicle, which is used to analyze the user's facial expressions and behavior and collect emotional data.

[0439] "Means for collecting health and emotional data" refers to system elements that use sensors and cameras to collect information about the user's health and emotional state in real time.

[0440] "Means for real-time analysis" refers to a system element that instantly analyzes collected health and emotional data and responds quickly if an abnormality is detected.

[0441] "When an abnormality is detected" refers to when an abnormal health or emotional state is recognized based on the collected health and emotional data.

[0442] The "means used as feedback for training the generative model" is a system element that incorporates information collected from the user after the diagnosis as a training dataset for the generative model to improve the accuracy of the model.

[0443] "Means for generating records" refers to a system element that automatically generates a medical record based on the analysis results and collected information, and transmits the generated record to a display in the vehicle or to a user terminal.

[0444] This system monitors the health status of passengers in autonomous vehicles in real time and quickly recommends appropriate medical professionals when an abnormality is detected. The system consists of a user terminal, sensors, cameras, a server, a generative model, an emotion engine, and an expert recommendation method. The overall system flow is as follows:

[0445] First, when a user inputs vague health-related information, the information is sent from the user terminal to the server. The user terminal consists of a mobile device used by passengers or an in-vehicle display, and information can be input through a user interface. Examples include inputting information in natural language, such as "I have a headache and feel unwell."

[0446] The server uses a generative model and an emotion engine to analyze the received information. The generative model employs machine learning algorithms to convert vague health information into specific diagnostic candidates. The emotion engine analyzes the input emotion information to identify the user's emotional state. Based on the analysis results, the server identifies the most appropriate medical professional and sends the information to the user's device.

[0447] Next, sensors and cameras installed inside the vehicle collect passenger health and emotional data in real time. The sensors acquire physiological data such as heart rate and body temperature, and the cameras analyze the passenger's facial expression and facial expression. This data is sent to a server, and if an abnormality is detected in real time, it is analyzed and processed to enable rapid response.

[0448] The server automatically generates a medical record based on the analysis results and sends it to the vehicle's display or the user's device. This allows the user to quickly receive remote medical treatment from a recommended specialist. After the diagnosis, the user sends their feedback information back to the server, which is used as training data for the generative model, contributing to improving the model's accuracy.

[0449] As a concrete example, suppose a passenger inputs "I have a stomachache and I'm anxious." This information is sent to the server, and the generative model interprets it as "stomachache" and "anxiety." At the same time, sensors detect an increase in heart rate, and cameras recognize changes in the passenger's complexion. Based on this data, the server recommends a gastroenterologist and provides information to the passenger. The passenger undergoes a remote consultation and then provides feedback, which the server incorporates to update the generative model.

[0450] An example of a prompt might be:

[0451] "You are developing a system that analyzes passengers' health conditions based on data collected from sensors and cameras and recommends appropriate specialists. According to the collected data, Passenger A is experiencing a 'headache' and 'feeling unwell.' Which doctor would you recommend?"

[0452] In this way, the system enables real-time health monitoring and rapid response, providing an effective means of protecting passenger safety and health.

[0453] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0454] Step 1:

[0455] User enters health information

[0456] Using a user terminal, the user inputs vague information about their health (e.g., "I'm worried because my stomach hurts") into the input form and clicks the submit button. The input information is sent from the user terminal to the server.

[0457] Input: Vague health information (natural language)

[0458] Output: Health information sent to the server

[0459] Step 2:

[0460] The server receives the health information.

[0461] The server receives vague health information sent from the user's device, temporarily stores it in a database, and prepares it for the next analysis step.

[0462] Input: Health information sent from the user's device

[0463] Output: Saved health information

[0464] Step 3:

[0465] Analyzing health information with generative models

[0466] The server passes the stored health information to the generative model, which then begins analysis. The generative model uses machine learning algorithms to convert vague information into specific symptoms. For example, the model outputs information like "I have a stomachache and I'm anxious" as specific symptoms like "stomachache" and "anxiety."

[0467] Input: Saved Health Information

[0468] Output: Specific symptoms

[0469] Step 4:

[0470] Analyzing emotional information with an emotion engine

[0471] The server passes the emotional expressions contained in the stored health information to the emotion engine for emotion analysis. The emotion engine uses natural language processing technology to identify the user's emotional state (e.g., anxiety, worry).

[0472] Input: Emotional expressions contained in stored health information

[0473] Output: User's emotional state

[0474] Step 5:

[0475] Recommend an expert

[0476] The server recommends an appropriate expert based on the analysis results of the generative model and the emotion engine. It searches the database for relevant experts, identifies the expert best suited to the user (e.g., a gastroenterologist), and sends the information to the user's terminal.

[0477] Input: specific symptoms, user's emotional state

[0478] Output: Recommended expert information

[0479] Step 6:

[0480] Collecting health data in real time

[0481] Sensors and cameras installed in the vehicle collect passenger health data (e.g., heart rate, body temperature) and emotional data (e.g., facial color, facial expressions) in real time, and the collected data is sent to a server.

[0482] Input: Health and emotion data collected by sensors and cameras

[0483] Output: Real-time data sent to the server

[0484] Step 7:

[0485] Analyzing real-time data

[0486] The server analyzes the health and emotional data sent in real time and performs analytical processing to quickly respond if an abnormality is detected, such as a sudden increase in heart rate or a change in facial color.

[0487] Input: Health and emotional data transmitted in real time

[0488] Output: Anomaly detection results

[0489] Step 8:

[0490] Response when an abnormality is detected

[0491] If the server detects an abnormality, it will immediately recommend an appropriate specialist and send that information to the user's device and the vehicle's display, allowing the user to receive prompt and appropriate medical treatment.

[0492] Input: Anomaly detection result

[0493] Output: Pertinent expert information

[0494] Step 9:

[0495] Gathering feedback and learning

[0496] After the diagnosis, the user sends feedback information about changes in their symptoms to the server, which uses this information as training data for the generative model to improve its accuracy.

[0497] Input: Feedback information

[0498] Output: Updated generative model

[0499] Step 10:

[0500] Automatic recording generation and transmission

[0501] The server automatically generates a medical record based on the analysis results and collected information and sends it to the vehicle's display or user terminal, allowing the user to check the medical record and take appropriate medical action.

[0502] Input: Analysis results and collected information

[0503] Output: Generated medical record

[0504] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0505] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0506] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0507] [Second embodiment]

[0508] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0509] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0510] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0511] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0512] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0513] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0514] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0515] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0516] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0517] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0518] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0519] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0520] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, not only improving the accuracy of diagnoses but also increasing the work efficiency of doctors.

[0521] System Overview

[0522] This system is primarily composed of a user terminal, a server, a generative model, and an expert recommendation means. The user terminal is responsible for sending health information entered by the user to the server. The server receives this information and analyzes it using the generative model. Based on the analysis results, the server recommends an appropriate expert and collects subsequent feedback information to train the generative model.

[0523] Program processing explanation

[0524] User behavior

[0525] 1. Symptom input: The user inputs vague symptoms related to their health into an input form on the user terminal. For example, they input vague information such as "my stomach hurts" or "I feel a bit sick."

[0526] 2. Sending information: After entering the information, the user clicks the "Send" button to send the information to the server.

[0527] Server-side operation

[0528] 1. Information reception: The server receives vague health information sent from the user terminal.

[0529] 2. Analysis: The received information is analyzed using a generative model, which uses natural language processing techniques to convert ambiguous expressions into specific diagnostic candidates.

[0530] 3. Expert recommendation: Based on the analysis results, an appropriate expert (e.g., a gastroenterologist or gastroenterologist) is identified from the database and the information is sent to the user's terminal.

[0531] Generate medical records

[0532] 1. Information storage: The server stores the input information from the user and the analysis results in a database.

[0533] 2. Medical record generation: Medical records (medical records) are automatically generated based on the analysis results and sent to recommended specialists.

[0534] Gathering feedback and learning

[0535] 1. Feedback input: After the diagnosis, the user again inputs how their condition has changed and sends it to the server. For example, information such as "symptoms have improved" or "new symptoms have appeared."

[0536] 2. Learning update: Based on the received feedback, the server updates the training dataset of the generative model to improve the accuracy of the diagnostic algorithm.

[0537] Specific examples

[0538] Example 1: Patient complaining of abdominal pain

[0539] 1. User terminal: The patient types "My stomach hurts" and sends it.

[0540] 2. Server: Receives input data and analyzes it with the generative model.

[0541] 3. Server: The generative model converts the information into specific symptoms such as "stomachache," "pain in the lower right abdomen," and "nausea."

[0542] 4. Server: Retrieves gastroenterologists and gastroenterologists from the database and sends them to the user terminal.

[0543] 5. User terminal: The patient selects a doctor from the recommended list.

[0544] 6. Server: Receives the selected information, automatically generates a medical record, and sends it to the doctor.

[0545] 7. User terminal: After diagnosis, the patient enters and submits "Symptoms are not improving."

[0546] 8. Server: Receives feedback and updates the generative model.

[0547] This concludes the description of the embodiment of the present invention. This system makes it possible to accurately analyze a patient's vague symptoms and recommend the most appropriate specialist. It also improves the work efficiency of doctors, enabling faster and more reliable medical care.

[0548] The processing flow will be explained below.

[0549] Step 1:

[0550] The user enters vague health symptoms into the input form on the user terminal and clicks the "Submit" button.

[0551] Step 2:

[0552] The device sends vague health information entered by the user to the server as an HTTP POST request.

[0553] Step 3:

[0554] The server receives vague health information from the user terminal and temporarily stores it in a database.

[0555] Step 4:

[0556] The server passes the ambiguous information it receives to the natural language processing engine and generative model, and begins the analysis process.

[0557] Step 5:

[0558] A generative model analyzes ambiguous information and identifies likely specific diagnoses (e.g., "stomach pain," "pain in the right lower abdomen," "nausea").

[0559] Step 6:

[0560] Based on the analysis results, the server retrieves relevant specialists (e.g., gastroenterologists, gastroenterologists, etc.) from the database.

[0561] Step 7:

[0562] The server ranks the acquired expert information and sends it to the user's terminal as an HTTP response.

[0563] Step 8:

[0564] The terminal displays the expert information received from the server in list form on the user interface.

[0565] Step 9:

[0566] The user selects the desired expert from the displayed list of experts and clicks the "Select" button.

[0567] Step 10:

[0568] The device sends the information of the selected expert to the server as an HTTP POST request.

[0569] Step 11:

[0570] The server receives the selections from the user and stores them in a database.

[0571] Step 12:

[0572] The server generates a medical record based on the analysis results and the selected expert information.

[0573] Step 13:

[0574] The server sends the generated medical record to the corresponding specialist's terminal as an HTTP response.

[0575] Step 14:

[0576] After the diagnosis, the user enters the progress of their symptoms and any new symptoms into the feedback form on the user's device and clicks the "Submit" button.

[0577] Step 15:

[0578] The terminal sends the input feedback data to the server as an HTTP POST request.

[0579] Step 16:

[0580] The server receives the feedback data and stores it in a database.

[0581] Step 17:

[0582] The server analyzes the received feedback data and provides it as training data for the generative model.

[0583] Step 18:

[0584] The generative model incorporates new feedback data, updates the learning algorithm, and improves diagnostic accuracy.

[0585] The above is a specific processing flow for making a diagnosis based on vague information about the user's health and improving accuracy using feedback.

[0586] Example 1

[0587] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0588] In conventional health support systems, the health information entered by users is vague, making it difficult to make accurate diagnoses or recommend appropriate specialists. Furthermore, feedback information after diagnosis is not fully utilized, making it difficult to improve the accuracy of the system. Furthermore, the creation of medical records is not automated, increasing the workload of medical professionals.

[0589] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0590] In this invention, the server includes a means for receiving ambiguous health-related information input by a user, an analysis means using a generative model to analyze the received ambiguous health-related information, and a means for identifying and recommending an appropriate specialist based on the analysis results obtained by the analysis means, thereby making it possible to convert ambiguous health-related information into specific diagnosis candidates and recommend an appropriate specialist.

[0591] A "user terminal" is a device that allows a user to input health-related information and transmit that information to a server.

[0592] The "server" is the central computer of the system that receives information sent from user terminals, analyzes it, and recommends experts.

[0593] A "generative model" is an algorithm that uses natural language processing techniques to analyze ambiguous information received and convert it into specific diagnostic candidates.

[0594] "Analysis means" refers to the process of using a generative model to analyze received ambiguous information and generate specific diagnosis candidates.

[0595] The "expert recommendation means" is a means for identifying an appropriate expert from a database based on the analysis results and providing that information.

[0596] "Information storage means" refers to the process of storing user input information and analysis results in a database.

[0597] "Medical record generation means" refers to the process of automatically generating a medical record based on the analysis results and sending it to the recommended specialist.

[0598] "Feedback input means" refers to a process in which the user re-enters information about changes in his or her condition after diagnosis and sends that information to the server.

[0599] The "learning update means" refers to the process of adding received feedback information to the learning dataset of the generative model to improve the accuracy of the model.

[0600] "Ambiguous information" refers to information entered by a user in vague terms, without clearly describing specific symptoms or conditions.

[0601] "Analysis results" refer to specific diagnosis candidates and expert recommendation information analyzed by the generative model.

[0602] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, not only improving the accuracy of diagnoses but also increasing the work efficiency of doctors.

[0603] System Overview

[0604] This system is primarily composed of a user terminal, a server, a generative model, and an expert recommendation means. The user terminal is responsible for sending health information entered by the user to the server. The server receives this information and analyzes it using the generative model. Based on the analysis results, the server recommends an appropriate expert and collects subsequent feedback information to train the generative model.

[0605] User behavior

[0606] The user enters vague symptoms related to their health into an input form on the user terminal. Specifically, they enter vague information such as "my stomach hurts" or "I feel a bit sick." Then, the user clicks the "send" button to send the information to the server.

[0607] Server-side operation

[0608] Receiving information

[0609] The server receives vague health information sent from the user's device. The server's API receives the request from the device and adds the text data to a processing queue.

[0610] Analysis means

[0611] The server analyzes the received information using a generative model. This model uses natural language processing technology to convert vague expressions into specific diagnosis candidates. For example, if an input says "my stomach hurts," the generative model converts that into specific symptoms such as "stomachache," "pain in the right lower abdomen," and "nausea."

[0612] Expert recommendation

[0613] Based on the analysis results, the server identifies an appropriate expert from the database. The expert's information is returned to the user's terminal. The server sends the analysis results and the expert's information to the terminal in response to the request.

[0614] Generate medical records

[0615] Information storage

[0616] The server stores the user's input information and analysis results in a database, which then becomes historical data that can be referenced later.

[0617] Medical record generation

[0618] The server automatically generates a medical record (chart) based on the analysis results, which is then sent to the recommended specialist.

[0619] Gathering feedback and learning

[0620] Feedback Input

[0621] After the diagnosis, the user again inputs how their condition has changed and sends it to the server. For example, they input feedback information such as "symptoms have improved" or "new symptoms have appeared."

[0622] Learning Update

[0623] The server uses the received feedback to update the training dataset of the generative model, thereby improving the accuracy of the diagnostic algorithm.

[0624] Specific examples

[0625] Example 1: Patient complaining of abdominal pain

[0626] 1. Symptom input:

[0627] The user inputs "My stomach hurts" into the user terminal and sends it.

[0628] 2. Receiving information:

[0629] The server receives this input data and analyzes it with the generative model.

[0630] 3. Analysis method:

[0631] The generative model converts the input "my stomach hurts" into specific symptoms such as "stomachache," "pain in the lower right abdomen," and "nausea."

[0632] 4. Expert recommendation:

[0633] The server retrieves gastroenterology and gastroenterology specialists from a database and transmits them to the user terminal.

[0634] 5. Expert Selection:

[0635] The user selects a doctor from a list of recommendations.

[0636] 6. Medical record generation:

[0637] The server receives the selection information, automatically generates a medical record, and sends it to the doctor.

[0638] 7. Feedback Input:

[0639] After the diagnosis, the user enters "Symptoms are not improving" and submits the message.

[0640] 8. Learning Update:

[0641] The server receives the feedback and updates the generative model.

[0642] Prompt Sentence Examples

[0643] For the input "My stomach hurts," the generative AI model generates the following analysis result:

[0644] Entered information: "My stomach hurts"

[0645] Analysis results: "Stomach pain," "pain in the lower right abdomen," "nausea"

[0646] Appropriate specialist: "Gastroenterologist" "Gastroenterologist"

[0647] This makes it possible to turn vague symptoms into specific candidate diagnoses and identify appropriate specialists.

[0648] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0649] Step 1: Enter your symptoms

[0650] The user inputs vague information about their symptoms into an input form on the terminal, such as "my stomach hurts" or "I feel a bit sick."

[0651] Specifically, the user uses the terminal keyboard to input symptoms into a text box.

[0652] Input: Vague health information entered by the user.

[0653] Output: Symptom information entered into the user's terminal.

[0654] Step 2: Send information

[0655] The user checks the information they entered and clicks the "Send" button, which causes the user's device to send the data to the server.

[0656] Specifically, the terminal sends the input text data as an HTTP request to the server's API.

[0657] Input: The symptom information the user entered and clicked submit.

[0658] Output: Symptom information sent to the server.

[0659] Step 3: Receiving information

[0660] The server receives ambiguous health information transmitted from the user terminal.

[0661] Specifically, the server's API receives a request from the terminal and adds the text data to a processing queue.

[0662] Input: Symptom information sent from the user terminal.

[0663] Output: Symptom information received.

[0664] Step 4: Analysis methods

[0665] The server analyzes the received information using a generative model, which uses natural language processing techniques to convert ambiguous expressions into specific candidate diagnoses.

[0666] Specifically, the generative AI model analyzes input data and converts vague expressions into specific diagnostic candidates such as "stomachache," "pain in the lower right abdomen," and "nausea."

[0667] Input: Received symptom information.

[0668] Output: A list of specific diagnostic candidates.

[0669] Step 5: Expert Recommendations

[0670] Based on the analysis results, the server identifies appropriate experts from the database, and the recommended expert information is sent to the user terminal.

[0671] Specifically, the server compares the analysis results with a list of experts, generates a list of appropriate experts, and returns it to the user terminal.

[0672] Input: A specific list of diagnostic candidates.

[0673] Output: A list of recommended experts.

[0674] Step 6: Save your information

[0675] The server stores the input information from the user and the analysis results in a database.

[0676] Specifically, the server stores the input information and analysis results in a database table.

[0677] Input: Information entered by the user, analysis results.

[0678] Output: Saved data.

[0679] Step 7: Create a medical record

[0680] The server automatically generates a medical record (chart) based on the analysis results and sends it to the recommended specialist.

[0681] Specifically, the server generates medical records in accordance with the medical record format and sends them to specialists via email or a dedicated system.

[0682] Input: User input information, analysis results.

[0683] Output: Generated medical chart, medical record sent to specialist.

[0684] Step 8: Enter your feedback

[0685] After the diagnosis, the user again inputs any changes in their condition and sends it to the server. For example, they can input information such as "symptoms have improved" or "new symptoms have appeared."

[0686] As a specific operation, the user inputs feedback information using the keyboard of the terminal and clicks the send button.

[0687] Input: User feedback information.

[0688] Output: Feedback information sent by the user.

[0689] Step 9: Learning Update

[0690] The server adds the received feedback information to the training dataset of the generative model and retrains the model.

[0691] Specifically, the server adds feedback information to the dataset and retrains the generative AI model to improve its accuracy.

[0692] Input: Feedback information from the user.

[0693] Output: An updated generative model.

[0694] (Application example 1)

[0695] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0696] Conventional systems for analyzing ambiguous health-related information are limited to the function of recommending experts, and lack mechanisms for effectively collecting and learning from feedback information after recommendations and improving the accuracy of generative models. Furthermore, there is a lack of systems for quickly recommending appropriate products and experts for customer support in brick-and-mortar stores, making it difficult to improve customer satisfaction and ensure appropriate health management in brick-and-mortar stores.

[0697] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0698] In this invention, the server includes: means for receiving ambiguous health-related information entered by a user; analysis means using a generative model to analyze the received ambiguous health-related information; means for identifying and recommending an appropriate expert or product based on the analysis results obtained by the analysis means; means for collecting information after the customer has used the recommended expert or product and using the collected information as feedback to train the generative model; and means for automatically generating records based on the analysis results and the collected information. This enables accurate analysis of ambiguous health information in physical stores and rapid recommendation of appropriate experts and products, thereby improving customer satisfaction and streamlining health management in stores.

[0699] "User" means a customer who uses the System to input their health information.

[0700] "Ambiguous health information" refers to information about a condition in which the user is experiencing some kind of discomfort or symptom but cannot specifically describe it.

[0701] "Means for receiving" refers to the function by which the server receives vague health-related information sent from the user terminal.

[0702] "Analysis means" refers to the function that converts received ambiguous health information into specific expert and product recommendations using a generative model.

[0703] A "generative model" is an AI model that uses machine learning algorithms to analyze input information and generate specific diagnostic candidates and recommendations.

[0704] An "expert" is someone who has knowledge and skills in a particular field and can provide appropriate advice and diagnoses to users.

[0705] "Means for identifying and recommending" refers to a function that notifies users of appropriate experts and products based on the results obtained by the analysis means.

[0706] "Feedback" refers to information entered by users about the effects or changes they have experienced after using a recommended professional or product.

[0707] "Means used for learning" refers to the function of using collected feedback information as training data for a generative model to improve the accuracy of the model.

[0708] "Means for generating records" refers to a function that automatically creates medical records and advice records based on analysis results and feedback information.

[0709] "System" refers to a comprehensive mechanism that includes all of the above means and functions and is used to analyze vague information about a user's health and recommend appropriate experts and products.

[0710] The following describes in detail an embodiment of the "Smart Health Assistant" of the present invention. The present invention is a system that identifies the most suitable specialist or product based on vague health information entered by the user, makes more accurate recommendations, and collects feedback after the recommendation to improve the accuracy of the model.

[0711] System Overview

[0712] This system consists of a user terminal, a server, a generative model, and a database. The user terminal is responsible for sending health information entered by the user to the server. The server receives the information and analyzes it using the generative model. Based on the analysis results, the server recommends appropriate experts or products, and then collects subsequent feedback information to train the generative model.

[0713] User behavior

[0714] 1. Symptom input: The user inputs vague symptoms related to their health into an input form on the user device (e.g., smart glasses). For example, they input vague information such as "I have a headache" or "My skin is itchy."

[0715] 2. Sending information: After entering the information, the user clicks the "Send" button to send the information to the server.

[0716] Server-side operation

[0717] 1. Information reception: The server receives vague health information sent from the user terminal.

[0718] 2. Analysis method: The received information is analyzed using a generative model (e.g., OpenAI GPT-4). This generative model uses natural language processing technology to convert ambiguous expressions into specific diagnosis candidates and recommendations.

[0719] 3. Expert recommendation: Based on the analysis results, the appropriate expert (pharmacist, health advisor, etc.) or product (painkiller, moisturizing cream, etc.) is identified from the database and the information is sent to the user's device.

[0720] Generate medical records

[0721] 1. Information storage: The server stores the input information from the user and the analysis results in a database.

[0722] 2. Medical record generation: A medical record (medical record) or advice record is automatically generated based on the analysis results and sent to the recommended specialist.

[0723] Gathering feedback and learning

[0724] 1. Feedback input: The user re-enters information about the effects and changes in condition after using the recommended specialist or product, and sends it to the server. For example, information such as "Symptoms improved" or "Insufficient effect."

[0725] 2. Learning update: Based on the received feedback, the server updates the training dataset of the generative model to improve the accuracy of the diagnosis and recommendation algorithm.

[0726] Specific examples

[0727] Example 1: A customer complaining of a headache

[0728] 1. Smart Glasses: The customer types "I have a headache" and submits.

[0729] 2. Server: Receives input data and analyzes it with the generative model.

[0730] 3. Server: The generative model converts the data into specific diagnosis candidates such as "tension headache" or "stress."

[0731] 4. Server: Retrieves the appropriate pharmacist or painkiller from the database and sends it to the user's smart glasses.

[0732] 5. Smart Glasses: Customers select a pharmacist or product from a list of recommendations.

[0733] 6. Server: Receives the selected information, automatically generates a medical record, and sends it to the pharmacist.

[0734] 7. Smart glasses: The customer types "my symptoms have improved" and submits.

[0735] 8. Server: Receives feedback and updates the generative model.

[0736] Prompt Sentence Examples

[0737] "Enter any vague health symptoms you have. For example, I have a headache, my skin is itchy, etc."

[0738] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0739] Step 1: Enter your symptoms

[0740] Users enter vague health symptoms into a form on the smart glasses in a physical store. Examples of information include "I have a headache" or "My skin is itchy." This input is the starting point for the system, which then identifies the user's complaint in concrete terms.

[0741] Step 2: Send information

[0742] After inputting the information, the user clicks the "Send" button. The smart glasses then send the input information to the store's server. The transmitted data includes the symptom information entered by the user. The server then passes the received information to the next analysis step.

[0743] Step 3: Receiving information

[0744] The server receives the vague health information sent by the smart glasses, stores the received data in preparation for analysis, and passes it on to the next processing step.

[0745] Step 4: Analysis

[0746] The server analyzes the received information using a generative model (e.g., OpenAI GPT-4). The input includes the symptom information submitted by the user. The generative model analyzes the ambiguous information and generates specific diagnosis candidates and recommendations. The output includes specific diagnosis candidates such as "tension headache" and "stress."

[0747] Step 5: Expert or product recommendation

[0748] The server identifies appropriate experts (pharmacists, health advisors, etc.) or products (painkillers, moisturizing creams, etc.) from a database based on the analysis results obtained by the generative model. The input includes the analysis results, and the server selects the best recommendation destination based on these results. The output includes a list of recommended experts and products.

[0749] Step 6: Submit your recommendation

[0750] The server sends the identified expert or product information to the user's smart glasses. The input includes the identified recommendation information, which is sent to notify the user. The output includes a recommendation list that is displayed on the user's device.

[0751] Step 7: Enter your feedback

[0752] After receiving expert advice or using the product, the user again enters feedback about the effects and changes in their physical condition into the smart glasses' input form. For example, feedback such as "Symptoms improved" or "Not effective enough" is entered. This input information is used for subsequent learning.

[0753] Step 8: Send your feedback

[0754] After entering the feedback information, the user clicks the "Send" button. The smart glasses then send this feedback information to the store's server. The transmitted data includes the feedback information. The server then passes the received information to the next learning process.

[0755] Step 9: Receiving feedback

[0756] The server receives the feedback information sent from the smart glasses, stores the received data in a training dataset, and passes it on to the next processing step.

[0757] Step 10: Retraining the generative model

[0758] The server retrains the generative model based on the received feedback. The input includes the feedback, which improves the accuracy of the generative model. The output includes the updated generative model.

[0759] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0760] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, improving the accuracy of diagnosis, and also recognizes the user's emotions and reflects them in the selection of the recommended specialist.

[0761] System Overview

[0762] This system is primarily composed of a user terminal, a server, a generative model, an emotion engine, and an expert recommendation means. The user terminal is responsible for sending health information and natural language emotion expressions entered by the user to the server. The server receives this information and analyzes it using the generative model and emotion engine. Based on the analysis results, the server recommends an appropriate expert and collects subsequent feedback information to train the generative model.

[0763] Program processing explanation

[0764] User behavior

[0765] 1. Symptom and emotion input: The user inputs vague symptoms related to their health and the emotions associated with them into an input form on the user terminal. For example, they input information such as "I have a stomachache and I'm worried" or "I feel a bit sick and worried."

[0766] 2. Sending information: After inputting, the user clicks the "Send" button to send the symptom and emotion information to the server.

[0767] Server-side operation

[0768] 1. Information reception: The server receives vague health information and emotion information sent from the user terminal and temporarily stores them in a database.

[0769] 2. Symptom analysis: The received information is passed to a natural language processing engine and a generative model to begin the symptom analysis process. This generative model uses natural language processing techniques to convert ambiguous expressions into specific diagnostic candidates.

[0770] 3. Emotion analysis means: At the same time, the emotion engine analyzes the received emotion information to grasp the user's mental state.

[0771] 4. Expert recommendation: Based on the results of symptom analysis and emotion analysis, relevant experts (e.g., gastroenterologists) are retrieved from a database, and the most suitable expert is identified taking into account emotion information, and this information is sent to the user's device.

[0772] Generate medical records

[0773] 1. Information storage: The server stores the input information from the user and the analysis results in a database.

[0774] 2. Medical record generation: Medical records (medical records) are automatically generated based on the analysis results and sent to recommended specialists.

[0775] Gathering feedback and learning

[0776] 1. Feedback input: After the diagnosis, the user again inputs how their condition has changed and sends it to the server. For example, they input feedback information such as "Symptoms have improved" or "New symptoms have appeared."

[0777] 2. Learning update: Based on the received feedback, the server updates the training dataset of the generative model to improve the accuracy of the diagnostic algorithm.

[0778] Specific examples

[0779] Example 1: Patient complaining of abdominal pain

[0780] 1. User terminal: The patient types, "I have a stomachache and I'm worried" and sends it.

[0781] 2. Server: Receives input data and analyzes symptom data using a generative model and emotion data using an emotion engine.

[0782] 3. Server: The generative model converts the symptoms into specific ones such as "stomachache," "pain in the lower right abdomen," and "nausea," and the emotion engine identifies emotions such as "anxiety" and "worry."

[0783] 4. Server: Retrieves gastroenterologists and gastroenterologists from a database, identifies the appropriate doctor for the patient, taking into account their emotional state, and sends the information to the user's terminal.

[0784] 5. User terminal: The patient selects a doctor from the recommended list.

[0785] 6. Server: Receives the selected information, automatically generates a medical record, and sends it to the doctor.

[0786] 7. User terminal: After diagnosis, the patient enters and submits "Symptoms are not improving."

[0787] 8. Server: Receives feedback and updates the generative model.

[0788] This concludes the description of the embodiment of the present invention. This system makes it possible to accurately analyze a patient's vague symptoms and emotions and recommend the most appropriate specialist. It also improves the work efficiency of doctors, enabling faster and more reliable medical care.

[0789] The processing flow will be explained below.

[0790] Step 1:

[0791] The user enters vague health symptoms and associated feelings into an input form on the user's device and clicks the "Submit" button.

[0792] Step 2:

[0793] The terminal sends the ambiguous health information and emotion information input by the user to the server as an HTTP POST request.

[0794] Step 3:

[0795] The server receives the information sent from the user terminal and temporarily stores it in a database.

[0796] Step 4:

[0797] The server passes the received ambiguous information to a natural language processing engine and a generative model, and begins the symptom analysis process.

[0798] Step 5:

[0799] A generative model analyzes ambiguous information and identifies likely specific diagnoses (e.g., "stomach pain," "pain in the right lower abdomen," "nausea").

[0800] Step 6:

[0801] The server passes the received emotion information to the emotion engine and starts emotion analysis processing.

[0802] Step 7:

[0803] The emotion engine analyzes the emotional information and identifies the user's mental state (e.g., "anxiety" or "worry").

[0804] Step 8:

[0805] The server retrieves relevant specialists (e.g., gastroenterologists, gastroenterologists, etc.) from a database based on the symptom analysis and emotion analysis results.

[0806] Step 9:

[0807] The server optimizes the expert information it acquires, taking into account the results of sentiment analysis, and sends it to the user's device as an HTTP response.

[0808] Step 10:

[0809] The terminal displays the expert information received from the server in list form on the user interface.

[0810] Step 11:

[0811] The user selects the desired expert from the displayed list of experts and clicks the "Select" button.

[0812] Step 12:

[0813] The device sends the information of the selected expert to the server as an HTTP POST request.

[0814] Step 13:

[0815] The server receives the selections from the user and stores them in a database.

[0816] Step 14:

[0817] The server generates a medical record based on the symptom analysis results and the selected specialist information.

[0818] Step 15:

[0819] The server sends the generated medical record to the corresponding specialist's terminal as an HTTP response.

[0820] Step 16:

[0821] After the diagnosis, the user enters the progress of their symptoms and any new symptoms into the feedback form on the user's device and clicks the "Submit" button.

[0822] Step 17:

[0823] The terminal sends the input feedback data to the server as an HTTP POST request.

[0824] Step 18:

[0825] The server receives the feedback data and stores it in a database.

[0826] Step 19:

[0827] The server analyzes the received feedback data and provides it as training data for the generative model.

[0828] Step 20:

[0829] The generative model incorporates new feedback data, updates the learning algorithm, and improves diagnostic accuracy.

[0830] The above is a specific processing flow for making a diagnosis based on vague information and emotions about the user's health and improving accuracy using feedback.

[0831] Example 2

[0832] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0833] Conventional medical systems have difficulty accurately analyzing the vague health information entered by patients and recommending appropriate specialists. Furthermore, patients' emotional state is rarely reflected in diagnoses or specialist selection, potentially reducing diagnostic accuracy and patient satisfaction. Furthermore, there is a lack of mechanisms for effectively utilizing feedback information after diagnosis, which has delayed improvements to the medical system as a whole.

[0834] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0835] In this invention, the server includes: means for receiving ambiguous health-related information and emotional expressions input by a user; analysis means using a generative model and an emotion analysis engine to analyze the received ambiguous health-related information and emotional expressions; means for identifying and recommending an appropriate specialist based on the symptom analysis results and emotion analysis results obtained by the analysis means; means for collecting feedback information from patients after diagnosis and using the collected information to train the generative model and improve the accuracy of the algorithm; and means for automatically generating medical records based on the analysis results and collected information. This makes it possible to analyze ambiguous symptoms and emotional information and recommend the most appropriate specialist, improving patient diagnostic accuracy and satisfaction and promoting overall improvement of the medical system.

[0836] "Ambiguous health information" refers to health information that is expressed in terms or conditions that do not clearly indicate a specific diagnosis or symptom.

[0837] "Emotional expressions" are words or sentences that express the feelings or psychological state that the user is currently experiencing.

[0838] "Means for receiving" refers to a mechanism for acquiring information sent by a user and storing it in a form that can be used within the system.

[0839] A "generative model" is a machine learning algorithm that uses natural language processing techniques to convert vague expressions into specific diagnostic candidates.

[0840] An "emotion analysis engine" is software or an algorithm that analyzes emotional expressions contained in natural language and identifies a user's emotional state.

[0841] "Analysis means" refers to various processing devices within the system for analyzing received ambiguous health-related information and emotional expressions using generative models and emotion analysis engines.

[0842] The "means for identifying and recommending experts" is a mechanism for selecting appropriate medical experts based on the analysis results and presenting that information to the user.

[0843] The "means for collecting feedback information" is a mechanism for collecting information such as changes in condition and new symptoms sent by the user after diagnosis and incorporating it into the system.

[0844] The "means used to train the generative model" refers to a mechanism for updating the generative model based on collected feedback information and improving the accuracy of analysis from the next time onwards.

[0845] "Means for automatically generating medical records" refers to a system for automatically creating medical records (charts) based on the analysis results and collected feedback information.

[0846] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, improving the accuracy of diagnosis, and also recognizes the user's emotions and reflects them in the selection of the recommended specialist.

[0847] System Configuration

[0848] This system is mainly composed of a user terminal, a server, a generative model, a sentiment analysis engine, and an expert recommendation means.

[0849] User terminal operation

[0850] The user device serves to transmit the health information and emotional expressions entered by the user in natural language to the server. For example, a smartphone or computer application can function as a user device. The user uses these devices to input and transmit information such as "I'm worried because my stomach hurts."

[0851] Server Operation

[0852] The server is the central hardware that receives and analyzes information sent from user devices. The server has the following functions:

[0853] 1. Data reception and storage: The server receives vague health-related information and emotional expressions sent from the user terminal and stores them in a temporary database.

[0854] 2. Symptom Analysis: The server passes the received data to a generative model, which uses natural language processing techniques to convert vague expressions into specific diagnosis candidates. This generative model uses advanced natural language processing algorithms such as GPT-3 and BERT.

[0855] 3. Sentiment analysis: At the same time, the server passes the received emotional data to an emotion analysis engine to analyze the user's mental state. For example, tools such as Watson NLU or Hume AI can be used for the emotion analysis engine.

[0856] 4. Expert recommendation: Based on the analysis results, an appropriate expert is selected and related expert information is sent to the user's device. The expert recommendation involves retrieving appropriate expert information from the database and taking into account the user's emotional state.

[0857] Medical record generation and feedback

[0858] The server performs the following processes based on the analysis results and user feedback:

[0859] 1. Medical record generation: Medical records are automatically generated based on the analysis results and sent to the recommended specialists. This process improves the efficiency of doctors' work.

[0860] 2. Feedback collection: After the diagnosis, the user re-enters any changes in their condition and sends them to the server, including information such as "symptoms have improved" or "new symptoms have appeared."

[0861] 3. Learning and updating of generative model: The server updates the generative model based on the collected feedback information to improve the accuracy of the diagnostic algorithm.

[0862] Specific examples

[0863] For example, if a user types "I'm anxious because my stomach hurts" and submits it...

[0864] 1. User device: Enter and send information on the smartphone application.

[0865] 2. Server: Receives input data, analyzes symptom data using a generative model, and emotion data using an emotion analysis engine.

[0866] 3. Server: The generative model converts the symptoms into specific ones such as "stomachache" or "pain in the lower right abdomen," and the emotion analysis engine identifies emotions such as "anxiety" or "worry."

[0867] 4. Server: Retrieves gastroenterologists and gastroenterologists from a database and recommends the appropriate doctor to the patient, taking into account their emotional state.

[0868] 5. User terminal: The patient selects a doctor from the recommended list and receives treatment.

[0869] 6. Server: After diagnosis, the user inputs "symptoms do not improve" and submits it, which receives feedback and updates the generative model.

[0870] This concludes the description of the "Smart Health Link" system, which accurately analyzes vague symptoms and emotions and recommends appropriate specialists, thereby improving the accuracy of diagnoses and patient satisfaction.

[0871] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0872] Step 1: User Input

[0873] Users input vague information about their health and emotional expressions into an input form on their device, such as "I have a stomachache and I'm worried," into a smartphone or computer application.

[0874] Input: Ambiguous health-related information and emotional expressions

[0875] Output: An object that temporarily stores the input information

[0876] Step 2: Send data from the user device to the server

[0877] The user checks the information they have entered and clicks the "Send" button, which causes the user terminal to send the input data to the server.

[0878] Input: Information entered by the user and sent (vague health information and emotional expressions)

[0879] Output: HTTP request with the information sent

[0880] Step 3: Data received and stored by the server

[0881] The server receives the information sent from the user terminal and stores it in a temporary storage database.

[0882] Input: Information sent from the user's device (vague health-related information and emotional expressions)

[0883] Output: Input data stored in a database

[0884] Step 4: Symptom analysis by server

[0885] The server passes the received data to a generative model, which uses natural language processing techniques to convert ambiguous expressions into specific diagnosis candidates. The generative model uses GPT-3, BERT, and other models.

[0886] Input: Vague health information stored in a database

[0887] Output: Results converted into specific symptoms (e.g., "stomach pain," "pain in the lower right abdomen," etc.)

[0888] Step 5: Sentiment analysis on the server

[0889] The server passes the received emotional data to an emotion analysis engine, which may use Watson NLU or Hume AI, to analyze the user's mental state.

[0890] Input: Emotions stored in the database

[0891] Output: Analyzed emotion results (e.g., "anxiety," "worry," etc.)

[0892] Step 6: Server recommends experts

[0893] The server retrieves relevant experts from a database based on the results of symptom and emotion analysis, identifies the most suitable expert based on the analysis results, and sends that information to the user's device.

[0894] Input: Symptom analysis results and emotion analysis results

[0895] Output: Recommended expert information (e.g., "Gastroenterologist")

[0896] Step 7: Generate and send medical records

[0897] The server automatically generates a medical record based on the analysis results and the user's input data and sends it to the recommended specialist.

[0898] Input: Analysis results and user input data

[0899] Output: Automatically generated medical records (charts)

[0900] Step 8: Provide feedback

[0901] After the diagnosis, the user inputs any changes in their condition again, such as feedback information such as "symptoms have improved" or "new symptoms have appeared," and sends it to the server.

[0902] Input: Feedback information after diagnosis

[0903] Output: Feedback information sent to the server

[0904] Step 9: Update learning using feedback

[0905] Based on the received feedback, the server updates the training dataset for the generative model and improves the accuracy of the analysis algorithm.

[0906] Input: Collected feedback information

[0907] Output: Updated generative model and algorithm

[0908] This is the specific program processing flow of this system. This system makes it possible to accurately analyze vague health-related information and emotions and recommend the most appropriate health specialist.

[0909] (Application example 2)

[0910] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0911] Autonomous vehicles require constant monitoring of the health status of drivers and passengers, and prompt and appropriate medical treatment when an abnormality is detected. However, current systems lack the means to analyze health and emotional data in real time and recommend the most appropriate medical specialist. As a result, there is a risk that responses to abnormalities may be delayed, hindering the provision of prompt and appropriate medical care.

[0912] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0913] In this invention, the server includes: means for receiving ambiguous health-related information input by a user; means for analyzing the ambiguous information using a generative model; means for identifying and recommending an appropriate expert based on the obtained analysis results; means for collecting health data and emotion data using sensors and cameras installed in the vehicle; means for analyzing the collected health data and emotion data in real time and recommending an optimal expert when an abnormality is detected; means for collecting post-diagnosis information from the patient and using it as feedback to train the generative model; and means for automatically generating a record based on the analysis results and the collected information and transmitting it to an in-vehicle display or a user terminal. This enables constant monitoring of the health status of passengers in an autonomous vehicle and quickly recommending an appropriate expert when an abnormality is detected.

[0914] "User" means the driver and passengers in the vehicle who input health information into the system.

[0915] "Ambiguous health information" refers to information about a health condition entered by a user that does not specify specific symptoms, and includes information expressed in natural language.

[0916] A "generative model" is a model that uses machine learning algorithms to analyze ambiguous health information and generate specific diagnostic candidates.

[0917] The "analysis means" is a system element that uses a generative model to analyze received ambiguous health information and obtain a diagnosis.

[0918] "Experts" are medical professionals recommended based on diagnostic results and who have expertise in a particular medical field.

[0919] The "means for recommending an expert" is a system element that identifies an appropriate expert based on the results obtained by the analysis means and recommends the expert to the user.

[0920] "Sensors" refer to devices installed in vehicles that collect physiological data such as heart rate and body temperature.

[0921] The "camera" is a video capture device installed inside the vehicle, which is used to analyze the user's facial expressions and behavior and collect emotional data.

[0922] "Means for collecting health and emotional data" refers to system elements that use sensors and cameras to collect information about the user's health and emotional state in real time.

[0923] "Means for real-time analysis" refers to a system element that instantly analyzes collected health and emotional data and responds quickly if an abnormality is detected.

[0924] "When an abnormality is detected" refers to when an abnormal health or emotional state is recognized based on the collected health and emotional data.

[0925] The "means used as feedback for training the generative model" is a system element that incorporates information collected from the user after the diagnosis as a training dataset for the generative model to improve the accuracy of the model.

[0926] "Means for generating records" refers to a system element that automatically generates a medical record based on the analysis results and collected information, and transmits the generated record to a display in the vehicle or to a user terminal.

[0927] This system monitors the health status of passengers in autonomous vehicles in real time and quickly recommends appropriate medical professionals when an abnormality is detected. The system consists of a user terminal, sensors, cameras, a server, a generative model, an emotion engine, and an expert recommendation method. The overall system flow is as follows:

[0928] First, when a user inputs vague health-related information, the information is sent from the user terminal to the server. The user terminal consists of a mobile device used by passengers or an in-vehicle display, and information can be input through a user interface. Examples include inputting information in natural language, such as "I have a headache and feel unwell."

[0929] The server uses a generative model and an emotion engine to analyze the received information. The generative model employs machine learning algorithms to convert vague health information into specific diagnostic candidates. The emotion engine analyzes the input emotion information to identify the user's emotional state. Based on the analysis results, the server identifies the most appropriate medical professional and sends the information to the user's device.

[0930] Next, sensors and cameras installed inside the vehicle collect passenger health and emotional data in real time. The sensors acquire physiological data such as heart rate and body temperature, and the cameras analyze the passenger's facial expression and facial expression. This data is sent to a server, and if an abnormality is detected in real time, it is analyzed and processed to enable rapid response.

[0931] The server automatically generates a medical record based on the analysis results and sends it to the vehicle's display or the user's device. This allows the user to quickly receive remote medical treatment from a recommended specialist. After the diagnosis, the user sends their feedback information back to the server, which is used as training data for the generative model, contributing to improving the model's accuracy.

[0932] As a concrete example, suppose a passenger inputs "I have a stomachache and I'm anxious." This information is sent to the server, and the generative model interprets it as "stomachache" and "anxiety." At the same time, sensors detect an increase in heart rate, and cameras recognize changes in the passenger's complexion. Based on this data, the server recommends a gastroenterologist and provides information to the passenger. The passenger undergoes a remote consultation and then provides feedback, which the server incorporates to update the generative model.

[0933] An example of a prompt might be:

[0934] "You are developing a system that analyzes passengers' health conditions based on data collected from sensors and cameras and recommends appropriate specialists. According to the collected data, Passenger A is experiencing a 'headache' and 'feeling unwell.' Which doctor would you recommend?"

[0935] In this way, the system enables real-time health monitoring and rapid response, providing an effective means of protecting passenger safety and health.

[0936] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0937] Step 1:

[0938] User enters health information

[0939] Using a user terminal, the user inputs vague information about their health (e.g., "I'm worried because my stomach hurts") into the input form and clicks the submit button. The input information is sent from the user terminal to the server.

[0940] Input: Vague health information (natural language)

[0941] Output: Health information sent to the server

[0942] Step 2:

[0943] The server receives the health information.

[0944] The server receives vague health information sent from the user's device, temporarily stores it in a database, and prepares it for the next analysis step.

[0945] Input: Health information sent from the user's device

[0946] Output: Saved health information

[0947] Step 3:

[0948] Analyzing health information with generative models

[0949] The server passes the stored health information to the generative model, which then begins analysis. The generative model uses machine learning algorithms to convert vague information into specific symptoms. For example, the model outputs information like "I have a stomachache and I'm anxious" as specific symptoms like "stomachache" and "anxiety."

[0950] Input: Saved Health Information

[0951] Output: Specific symptoms

[0952] Step 4:

[0953] Analyzing emotional information with an emotion engine

[0954] The server passes the emotional expressions contained in the stored health information to the emotion engine for emotion analysis. The emotion engine uses natural language processing technology to identify the user's emotional state (e.g., anxiety, worry).

[0955] Input: Emotional expressions contained in stored health information

[0956] Output: User's emotional state

[0957] Step 5:

[0958] Recommend an expert

[0959] The server recommends an appropriate expert based on the analysis results of the generative model and the emotion engine. It searches the database for relevant experts, identifies the expert best suited to the user (e.g., a gastroenterologist), and sends the information to the user's terminal.

[0960] Input: specific symptoms, user's emotional state

[0961] Output: Recommended expert information

[0962] Step 6:

[0963] Collecting health data in real time

[0964] Sensors and cameras installed in the vehicle collect passenger health data (e.g., heart rate, body temperature) and emotional data (e.g., facial color, facial expressions) in real time, and the collected data is sent to a server.

[0965] Input: Health and emotion data collected by sensors and cameras

[0966] Output: Real-time data sent to the server

[0967] Step 7:

[0968] Analyzing real-time data

[0969] The server analyzes the health and emotional data sent in real time and performs analytical processing to quickly respond if an abnormality is detected, such as a sudden increase in heart rate or a change in facial color.

[0970] Input: Health and emotional data transmitted in real time

[0971] Output: Anomaly detection results

[0972] Step 8:

[0973] Response when an abnormality is detected

[0974] If the server detects an abnormality, it will immediately recommend an appropriate specialist and send that information to the user's device and the vehicle's display, allowing the user to receive prompt and appropriate medical treatment.

[0975] Input: Anomaly detection result

[0976] Output: Pertinent expert information

[0977] Step 9:

[0978] Gathering feedback and learning

[0979] After the diagnosis, the user sends feedback information about changes in their symptoms to the server, which uses this information as training data for the generative model to improve its accuracy.

[0980] Input: Feedback information

[0981] Output: Updated generative model

[0982] Step 10:

[0983] Automatic recording generation and transmission

[0984] The server automatically generates a medical record based on the analysis results and collected information and sends it to the vehicle's display or user terminal, allowing the user to check the medical record and take appropriate medical action.

[0985] Input: Analysis results and collected information

[0986] Output: Generated medical record

[0987] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0988] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0989] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0990] [Third embodiment]

[0991] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0992] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0993] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0994] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0995] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0996] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0997] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0998] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0999] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1000] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1001] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1002] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1003] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, not only improving the accuracy of diagnoses but also increasing the work efficiency of doctors.

[1004] System Overview

[1005] This system is primarily composed of a user terminal, a server, a generative model, and an expert recommendation means. The user terminal is responsible for sending health information entered by the user to the server. The server receives this information and analyzes it using the generative model. Based on the analysis results, the server recommends an appropriate expert and collects subsequent feedback information to train the generative model.

[1006] Program processing explanation

[1007] User behavior

[1008] 1. Symptom input: The user inputs vague symptoms related to their health into an input form on the user terminal. For example, they input vague information such as "my stomach hurts" or "I feel a bit sick."

[1009] 2. Sending information: After entering the information, the user clicks the "Send" button to send the information to the server.

[1010] Server-side operation

[1011] 1. Information reception: The server receives vague health information sent from the user terminal.

[1012] 2. Analysis: The received information is analyzed using a generative model, which uses natural language processing techniques to convert ambiguous expressions into specific diagnostic candidates.

[1013] 3. Expert recommendation: Based on the analysis results, an appropriate expert (e.g., a gastroenterologist or gastroenterologist) is identified from the database and the information is sent to the user's terminal.

[1014] Generate medical records

[1015] 1. Information storage: The server stores the input information from the user and the analysis results in a database.

[1016] 2. Medical record generation: Medical records (medical records) are automatically generated based on the analysis results and sent to recommended specialists.

[1017] Gathering feedback and learning

[1018] 1. Feedback input: After the diagnosis, the user again inputs how their condition has changed and sends it to the server. For example, information such as "symptoms have improved" or "new symptoms have appeared."

[1019] 2. Learning update: Based on the received feedback, the server updates the training dataset of the generative model to improve the accuracy of the diagnostic algorithm.

[1020] Specific examples

[1021] Example 1: Patient complaining of abdominal pain

[1022] 1. User terminal: The patient types "My stomach hurts" and sends it.

[1023] 2. Server: Receives input data and analyzes it with the generative model.

[1024] 3. Server: The generative model converts the information into specific symptoms such as "stomachache," "pain in the lower right abdomen," and "nausea."

[1025] 4. Server: Retrieves gastroenterologists and gastroenterologists from the database and sends them to the user terminal.

[1026] 5. User terminal: The patient selects a doctor from the recommended list.

[1027] 6. Server: Receives the selected information, automatically generates a medical record, and sends it to the doctor.

[1028] 7. User terminal: After diagnosis, the patient enters and submits "Symptoms are not improving."

[1029] 8. Server: Receives feedback and updates the generative model.

[1030] This concludes the description of the embodiment of the present invention. This system makes it possible to accurately analyze a patient's vague symptoms and recommend the most appropriate specialist. It also improves the work efficiency of doctors, enabling faster and more reliable medical care.

[1031] The processing flow will be explained below.

[1032] Step 1:

[1033] The user enters vague health symptoms into the input form on the user terminal and clicks the "Submit" button.

[1034] Step 2:

[1035] The device sends vague health information entered by the user to the server as an HTTP POST request.

[1036] Step 3:

[1037] The server receives vague health information from the user terminal and temporarily stores it in a database.

[1038] Step 4:

[1039] The server passes the ambiguous information it receives to the natural language processing engine and generative model, and begins the analysis process.

[1040] Step 5:

[1041] A generative model analyzes ambiguous information and identifies likely specific diagnoses (e.g., "stomach pain," "pain in the right lower abdomen," "nausea").

[1042] Step 6:

[1043] Based on the analysis results, the server retrieves relevant specialists (e.g., gastroenterologists, gastroenterologists, etc.) from the database.

[1044] Step 7:

[1045] The server ranks the acquired expert information and sends it to the user's terminal as an HTTP response.

[1046] Step 8:

[1047] The terminal displays the expert information received from the server in list form on the user interface.

[1048] Step 9:

[1049] The user selects the desired expert from the displayed list of experts and clicks the "Select" button.

[1050] Step 10:

[1051] The device sends the information of the selected expert to the server as an HTTP POST request.

[1052] Step 11:

[1053] The server receives the selections from the user and stores them in a database.

[1054] Step 12:

[1055] The server generates a medical record based on the analysis results and the selected expert information.

[1056] Step 13:

[1057] The server sends the generated medical record to the corresponding specialist's terminal as an HTTP response.

[1058] Step 14:

[1059] After the diagnosis, the user enters the progress of their symptoms and any new symptoms into the feedback form on the user's device and clicks the "Submit" button.

[1060] Step 15:

[1061] The terminal sends the input feedback data to the server as an HTTP POST request.

[1062] Step 16:

[1063] The server receives the feedback data and stores it in a database.

[1064] Step 17:

[1065] The server analyzes the received feedback data and provides it as training data for the generative model.

[1066] Step 18:

[1067] The generative model incorporates new feedback data, updates the learning algorithm, and improves diagnostic accuracy.

[1068] The above is a specific processing flow for making a diagnosis based on vague information about the user's health and improving accuracy using feedback.

[1069] Example 1

[1070] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1071] In conventional health support systems, the health information entered by users is vague, making it difficult to make accurate diagnoses or recommend appropriate specialists. Furthermore, feedback information after diagnosis is not fully utilized, making it difficult to improve the accuracy of the system. Furthermore, the creation of medical records is not automated, increasing the workload of medical professionals.

[1072] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1073] In this invention, the server includes a means for receiving ambiguous health-related information input by a user, an analysis means using a generative model to analyze the received ambiguous health-related information, and a means for identifying and recommending an appropriate specialist based on the analysis results obtained by the analysis means, thereby making it possible to convert ambiguous health-related information into specific diagnosis candidates and recommend an appropriate specialist.

[1074] A "user terminal" is a device that allows a user to input health-related information and transmit that information to a server.

[1075] The "server" is the central computer of the system that receives information sent from user terminals, analyzes it, and recommends experts.

[1076] A "generative model" is an algorithm that uses natural language processing techniques to analyze ambiguous information received and convert it into specific diagnostic candidates.

[1077] "Analysis means" refers to the process of using a generative model to analyze received ambiguous information and generate specific diagnosis candidates.

[1078] The "expert recommendation means" is a means for identifying an appropriate expert from a database based on the analysis results and providing that information.

[1079] "Information storage means" refers to the process of storing user input information and analysis results in a database.

[1080] "Medical record generation means" refers to the process of automatically generating a medical record based on the analysis results and sending it to the recommended specialist.

[1081] "Feedback input means" refers to a process in which the user re-enters information about changes in his or her condition after diagnosis and sends that information to the server.

[1082] The "learning update means" refers to the process of adding received feedback information to the learning dataset of the generative model to improve the accuracy of the model.

[1083] "Ambiguous information" refers to information entered by a user in vague terms, without clearly describing specific symptoms or conditions.

[1084] "Analysis results" refer to specific diagnosis candidates and expert recommendation information analyzed by the generative model.

[1085] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, not only improving the accuracy of diagnoses but also increasing the work efficiency of doctors.

[1086] System Overview

[1087] This system is primarily composed of a user terminal, a server, a generative model, and an expert recommendation means. The user terminal is responsible for sending health information entered by the user to the server. The server receives this information and analyzes it using the generative model. Based on the analysis results, the server recommends an appropriate expert and collects subsequent feedback information to train the generative model.

[1088] User behavior

[1089] The user enters vague symptoms related to their health into an input form on the user terminal. Specifically, they enter vague information such as "my stomach hurts" or "I feel a bit sick." Then, the user clicks the "send" button to send the information to the server.

[1090] Server-side operation

[1091] Receiving information

[1092] The server receives vague health information sent from the user's device. The server's API receives the request from the device and adds the text data to a processing queue.

[1093] Analysis means

[1094] The server analyzes the received information using a generative model. This model uses natural language processing technology to convert vague expressions into specific diagnosis candidates. For example, if an input says "my stomach hurts," the generative model converts that into specific symptoms such as "stomachache," "pain in the right lower abdomen," and "nausea."

[1095] Expert recommendation

[1096] Based on the analysis results, the server identifies an appropriate expert from the database. The expert's information is returned to the user's terminal. The server sends the analysis results and the expert's information to the terminal in response to the request.

[1097] Generate medical records

[1098] Information storage

[1099] The server stores the user's input information and analysis results in a database, which then becomes historical data that can be referenced later.

[1100] Medical record generation

[1101] The server automatically generates a medical record (chart) based on the analysis results, which is then sent to the recommended specialist.

[1102] Gathering feedback and learning

[1103] Feedback Input

[1104] After the diagnosis, the user again inputs how their condition has changed and sends it to the server. For example, they input feedback information such as "symptoms have improved" or "new symptoms have appeared."

[1105] Learning Update

[1106] The server uses the received feedback to update the training dataset of the generative model, thereby improving the accuracy of the diagnostic algorithm.

[1107] Specific examples

[1108] Example 1: Patient complaining of abdominal pain

[1109] 1. Symptom input:

[1110] The user inputs "My stomach hurts" into the user terminal and sends it.

[1111] 2. Receiving information:

[1112] The server receives this input data and analyzes it with the generative model.

[1113] 3. Analysis method:

[1114] The generative model converts the input "my stomach hurts" into specific symptoms such as "stomachache," "pain in the lower right abdomen," and "nausea."

[1115] 4. Expert recommendation:

[1116] The server retrieves gastroenterology and gastroenterology specialists from a database and transmits them to the user terminal.

[1117] 5. Expert Selection:

[1118] The user selects a doctor from a list of recommendations.

[1119] 6. Medical record generation:

[1120] The server receives the selection information, automatically generates a medical record, and sends it to the doctor.

[1121] 7. Feedback Input:

[1122] After the diagnosis, the user enters "Symptoms are not improving" and submits the message.

[1123] 8. Learning Update:

[1124] The server receives the feedback and updates the generative model.

[1125] Prompt Sentence Examples

[1126] For the input "My stomach hurts," the generative AI model generates the following analysis result:

[1127] Entered information: "My stomach hurts"

[1128] Analysis results: "Stomach pain," "pain in the lower right abdomen," "nausea"

[1129] Appropriate specialist: "Gastroenterologist" "Gastroenterologist"

[1130] This makes it possible to turn vague symptoms into specific candidate diagnoses and identify appropriate specialists.

[1131] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1132] Step 1: Enter your symptoms

[1133] The user inputs vague information about their symptoms into an input form on the terminal, such as "my stomach hurts" or "I feel a bit sick."

[1134] Specifically, the user uses the terminal keyboard to input symptoms into a text box.

[1135] Input: Vague health information entered by the user.

[1136] Output: Symptom information entered into the user's terminal.

[1137] Step 2: Send information

[1138] The user checks the information they entered and clicks the "Send" button, which causes the user's device to send the data to the server.

[1139] Specifically, the terminal sends the input text data as an HTTP request to the server's API.

[1140] Input: The symptom information the user entered and clicked submit.

[1141] Output: Symptom information sent to the server.

[1142] Step 3: Receiving information

[1143] The server receives ambiguous health information transmitted from the user terminal.

[1144] Specifically, the server's API receives a request from the terminal and adds the text data to a processing queue.

[1145] Input: Symptom information sent from the user terminal.

[1146] Output: Symptom information received.

[1147] Step 4: Analysis methods

[1148] The server analyzes the received information using a generative model, which uses natural language processing techniques to convert ambiguous expressions into specific candidate diagnoses.

[1149] Specifically, the generative AI model analyzes input data and converts vague expressions into specific diagnostic candidates such as "stomachache," "pain in the lower right abdomen," and "nausea."

[1150] Input: Received symptom information.

[1151] Output: A list of specific diagnostic candidates.

[1152] Step 5: Expert Recommendations

[1153] Based on the analysis results, the server identifies appropriate experts from the database, and the recommended expert information is sent to the user terminal.

[1154] Specifically, the server compares the analysis results with a list of experts, generates a list of appropriate experts, and returns it to the user terminal.

[1155] Input: A specific list of diagnostic candidates.

[1156] Output: A list of recommended experts.

[1157] Step 6: Save your information

[1158] The server stores the input information from the user and the analysis results in a database.

[1159] Specifically, the server stores the input information and analysis results in a database table.

[1160] Input: Information entered by the user, analysis results.

[1161] Output: Saved data.

[1162] Step 7: Create a medical record

[1163] The server automatically generates a medical record (chart) based on the analysis results and sends it to the recommended specialist.

[1164] Specifically, the server generates medical records in accordance with the medical record format and sends them to specialists via email or a dedicated system.

[1165] Input: User input information, analysis results.

[1166] Output: Generated medical chart, medical record sent to specialist.

[1167] Step 8: Enter your feedback

[1168] After the diagnosis, the user again inputs any changes in their condition and sends it to the server. For example, they can input information such as "symptoms have improved" or "new symptoms have appeared."

[1169] As a specific operation, the user inputs feedback information using the keyboard of the terminal and clicks the send button.

[1170] Input: User feedback information.

[1171] Output: Feedback information sent by the user.

[1172] Step 9: Learning Update

[1173] The server adds the received feedback information to the training dataset of the generative model and retrains the model.

[1174] Specifically, the server adds feedback information to the dataset and retrains the generative AI model to improve its accuracy.

[1175] Input: Feedback information from the user.

[1176] Output: An updated generative model.

[1177] (Application example 1)

[1178] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1179] Conventional systems for analyzing ambiguous health-related information are limited to the function of recommending experts, and lack mechanisms for effectively collecting and learning from feedback information after recommendations and improving the accuracy of generative models. Furthermore, there is a lack of systems for quickly recommending appropriate products and experts for customer support in brick-and-mortar stores, making it difficult to improve customer satisfaction and ensure appropriate health management in brick-and-mortar stores.

[1180] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1181] In this invention, the server includes: means for receiving ambiguous health-related information entered by a user; analysis means using a generative model to analyze the received ambiguous health-related information; means for identifying and recommending an appropriate expert or product based on the analysis results obtained by the analysis means; means for collecting information after the customer has used the recommended expert or product and using the collected information as feedback to train the generative model; and means for automatically generating records based on the analysis results and the collected information. This enables accurate analysis of ambiguous health information in physical stores and rapid recommendation of appropriate experts and products, thereby improving customer satisfaction and streamlining health management in stores.

[1182] "User" means a customer who uses the System to input their health information.

[1183] "Ambiguous health information" refers to information about a condition in which the user is experiencing some kind of discomfort or symptom but cannot specifically describe it.

[1184] "Means for receiving" refers to the function by which the server receives vague health-related information sent from the user terminal.

[1185] "Analysis means" refers to the function that converts received ambiguous health information into specific expert and product recommendations using a generative model.

[1186] A "generative model" is an AI model that uses machine learning algorithms to analyze input information and generate specific diagnostic candidates and recommendations.

[1187] An "expert" is someone who has knowledge and skills in a particular field and can provide appropriate advice and diagnoses to users.

[1188] "Means for identifying and recommending" refers to a function that notifies users of appropriate experts and products based on the results obtained by the analysis means.

[1189] "Feedback" refers to information entered by users about the effects or changes they have experienced after using a recommended professional or product.

[1190] "Means used for learning" refers to the function of using collected feedback information as training data for a generative model to improve the accuracy of the model.

[1191] "Means for generating records" refers to a function that automatically creates medical records and advice records based on analysis results and feedback information.

[1192] "System" refers to a comprehensive mechanism that includes all of the above means and functions and is used to analyze vague information about a user's health and recommend appropriate experts and products.

[1193] The following describes in detail an embodiment of the "Smart Health Assistant" of the present invention. The present invention is a system that identifies the most suitable specialist or product based on vague health information entered by the user, makes more accurate recommendations, and collects feedback after the recommendation to improve the accuracy of the model.

[1194] System Overview

[1195] This system consists of a user terminal, a server, a generative model, and a database. The user terminal is responsible for sending health information entered by the user to the server. The server receives the information and analyzes it using the generative model. Based on the analysis results, the server recommends appropriate experts or products, and then collects subsequent feedback information to train the generative model.

[1196] User behavior

[1197] 1. Symptom input: The user inputs vague symptoms related to their health into an input form on the user device (e.g., smart glasses). For example, they input vague information such as "I have a headache" or "My skin is itchy."

[1198] 2. Sending information: After entering the information, the user clicks the "Send" button to send the information to the server.

[1199] Server-side operation

[1200] 1. Information reception: The server receives vague health information sent from the user terminal.

[1201] 2. Analysis method: The received information is analyzed using a generative model (e.g., OpenAI GPT-4). This generative model uses natural language processing technology to convert ambiguous expressions into specific diagnosis candidates and recommendations.

[1202] 3. Expert recommendation: Based on the analysis results, the appropriate expert (pharmacist, health advisor, etc.) or product (painkiller, moisturizing cream, etc.) is identified from the database and the information is sent to the user's device.

[1203] Generate medical records

[1204] 1. Information storage: The server stores the input information from the user and the analysis results in a database.

[1205] 2. Medical record generation: A medical record (medical record) or advice record is automatically generated based on the analysis results and sent to the recommended specialist.

[1206] Gathering feedback and learning

[1207] 1. Feedback input: The user re-enters information about the effects and changes in condition after using the recommended specialist or product, and sends it to the server. For example, information such as "Symptoms improved" or "Insufficient effect."

[1208] 2. Learning update: Based on the received feedback, the server updates the training dataset of the generative model to improve the accuracy of the diagnosis and recommendation algorithm.

[1209] Specific examples

[1210] Example 1: A customer complaining of a headache

[1211] 1. Smart Glasses: The customer types "I have a headache" and submits.

[1212] 2. Server: Receives input data and analyzes it with the generative model.

[1213] 3. Server: The generative model converts the data into specific diagnosis candidates such as "tension headache" or "stress."

[1214] 4. Server: Retrieves the appropriate pharmacist or painkiller from the database and sends it to the user's smart glasses.

[1215] 5. Smart Glasses: Customers select a pharmacist or product from a list of recommendations.

[1216] 6. Server: Receives the selected information, automatically generates a medical record, and sends it to the pharmacist.

[1217] 7. Smart glasses: The customer types "my symptoms have improved" and submits.

[1218] 8. Server: Receives feedback and updates the generative model.

[1219] Prompt Sentence Examples

[1220] "Enter any vague health symptoms you have. For example, I have a headache, my skin is itchy, etc."

[1221] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1222] Step 1: Enter your symptoms

[1223] Users enter vague health symptoms into a form on the smart glasses in a physical store. Examples of information include "I have a headache" or "My skin is itchy." This input is the starting point for the system, which then identifies the user's complaint in concrete terms.

[1224] Step 2: Send information

[1225] After inputting the information, the user clicks the "Send" button. The smart glasses then send the input information to the store's server. The transmitted data includes the symptom information entered by the user. The server then passes the received information to the next analysis step.

[1226] Step 3: Receiving information

[1227] The server receives the vague health information sent by the smart glasses, stores the received data in preparation for analysis, and passes it on to the next processing step.

[1228] Step 4: Analysis

[1229] The server analyzes the received information using a generative model (e.g., OpenAI GPT-4). The input includes the symptom information submitted by the user. The generative model analyzes the ambiguous information and generates specific diagnosis candidates and recommendations. The output includes specific diagnosis candidates such as "tension headache" and "stress."

[1230] Step 5: Expert or product recommendation

[1231] The server identifies appropriate experts (pharmacists, health advisors, etc.) or products (painkillers, moisturizing creams, etc.) from a database based on the analysis results obtained by the generative model. The input includes the analysis results, and the server selects the best recommendation destination based on these results. The output includes a list of recommended experts and products.

[1232] Step 6: Submit your recommendation

[1233] The server sends the identified expert or product information to the user's smart glasses. The input includes the identified recommendation information, which is sent to notify the user. The output includes a recommendation list that is displayed on the user's device.

[1234] Step 7: Enter your feedback

[1235] After receiving expert advice or using the product, the user again enters feedback about the effects and changes in their physical condition into the smart glasses' input form. For example, feedback such as "Symptoms improved" or "Not effective enough" is entered. This input information is used for subsequent learning.

[1236] Step 8: Send your feedback

[1237] After entering the feedback information, the user clicks the "Send" button. The smart glasses then send this feedback information to the store's server. The transmitted data includes the feedback information. The server then passes the received information to the next learning process.

[1238] Step 9: Receiving feedback

[1239] The server receives the feedback information sent from the smart glasses, stores the received data in a training dataset, and passes it on to the next processing step.

[1240] Step 10: Retraining the generative model

[1241] The server retrains the generative model based on the received feedback. The input includes the feedback, which improves the accuracy of the generative model. The output includes the updated generative model.

[1242] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1243] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, improving the accuracy of diagnosis, and also recognizes the user's emotions and reflects them in the selection of the recommended specialist.

[1244] System Overview

[1245] This system is primarily composed of a user terminal, a server, a generative model, an emotion engine, and an expert recommendation means. The user terminal is responsible for sending health information and natural language emotion expressions entered by the user to the server. The server receives this information and analyzes it using the generative model and emotion engine. Based on the analysis results, the server recommends an appropriate expert and collects subsequent feedback information to train the generative model.

[1246] Program processing explanation

[1247] User behavior

[1248] 1. Symptom and emotion input: The user inputs vague symptoms related to their health and the emotions associated with them into an input form on the user terminal. For example, they input information such as "I have a stomachache and I'm worried" or "I feel a bit sick and worried."

[1249] 2. Sending information: After inputting, the user clicks the "Send" button to send the symptom and emotion information to the server.

[1250] Server-side operation

[1251] 1. Information reception: The server receives vague health information and emotion information sent from the user terminal and temporarily stores them in a database.

[1252] 2. Symptom analysis: The received information is passed to a natural language processing engine and a generative model to begin the symptom analysis process. This generative model uses natural language processing techniques to convert ambiguous expressions into specific diagnostic candidates.

[1253] 3. Emotion analysis means: At the same time, the emotion engine analyzes the received emotion information to grasp the user's mental state.

[1254] 4. Expert recommendation: Based on the results of symptom analysis and emotion analysis, relevant experts (e.g., gastroenterologists) are retrieved from a database, and the most suitable expert is identified taking into account emotion information, and this information is sent to the user's device.

[1255] Generate medical records

[1256] 1. Information storage: The server stores the input information from the user and the analysis results in a database.

[1257] 2. Medical record generation: Medical records (medical records) are automatically generated based on the analysis results and sent to recommended specialists.

[1258] Gathering feedback and learning

[1259] 1. Feedback input: After the diagnosis, the user again inputs how their condition has changed and sends it to the server. For example, they input feedback information such as "Symptoms have improved" or "New symptoms have appeared."

[1260] 2. Learning update: Based on the received feedback, the server updates the training dataset of the generative model to improve the accuracy of the diagnostic algorithm.

[1261] Specific examples

[1262] Example 1: Patient complaining of abdominal pain

[1263] 1. User terminal: The patient types, "I have a stomachache and I'm worried" and sends it.

[1264] 2. Server: Receives input data and analyzes symptom data using a generative model and emotion data using an emotion engine.

[1265] 3. Server: The generative model converts the symptoms into specific ones such as "stomachache," "pain in the lower right abdomen," and "nausea," and the emotion engine identifies emotions such as "anxiety" and "worry."

[1266] 4. Server: Retrieves gastroenterologists and gastroenterologists from a database, identifies the appropriate doctor for the patient, taking into account their emotional state, and sends the information to the user's terminal.

[1267] 5. User terminal: The patient selects a doctor from the recommended list.

[1268] 6. Server: Receives the selected information, automatically generates a medical record, and sends it to the doctor.

[1269] 7. User terminal: After diagnosis, the patient enters and submits "Symptoms are not improving."

[1270] 8. Server: Receives feedback and updates the generative model.

[1271] This concludes the description of the embodiment of the present invention. This system makes it possible to accurately analyze a patient's vague symptoms and emotions and recommend the most appropriate specialist. It also improves the work efficiency of doctors, enabling faster and more reliable medical care.

[1272] The processing flow will be explained below.

[1273] Step 1:

[1274] The user enters vague health symptoms and associated feelings into an input form on the user's device and clicks the "Submit" button.

[1275] Step 2:

[1276] The terminal sends the ambiguous health information and emotion information input by the user to the server as an HTTP POST request.

[1277] Step 3:

[1278] The server receives the information sent from the user terminal and temporarily stores it in a database.

[1279] Step 4:

[1280] The server passes the received ambiguous information to a natural language processing engine and a generative model, and begins the symptom analysis process.

[1281] Step 5:

[1282] A generative model analyzes ambiguous information and identifies likely specific diagnoses (e.g., "stomach pain," "pain in the right lower abdomen," "nausea").

[1283] Step 6:

[1284] The server passes the received emotion information to the emotion engine and starts emotion analysis processing.

[1285] Step 7:

[1286] The emotion engine analyzes the emotional information and identifies the user's mental state (e.g., "anxiety" or "worry").

[1287] Step 8:

[1288] The server retrieves relevant specialists (e.g., gastroenterologists, gastroenterologists, etc.) from a database based on the symptom analysis and emotion analysis results.

[1289] Step 9:

[1290] The server optimizes the expert information it acquires, taking into account the results of sentiment analysis, and sends it to the user's device as an HTTP response.

[1291] Step 10:

[1292] The terminal displays the expert information received from the server in list form on the user interface.

[1293] Step 11:

[1294] The user selects the desired expert from the displayed list of experts and clicks the "Select" button.

[1295] Step 12:

[1296] The device sends the information of the selected expert to the server as an HTTP POST request.

[1297] Step 13:

[1298] The server receives the selections from the user and stores them in a database.

[1299] Step 14:

[1300] The server generates a medical record based on the symptom analysis results and the selected specialist information.

[1301] Step 15:

[1302] The server sends the generated medical record to the corresponding specialist's terminal as an HTTP response.

[1303] Step 16:

[1304] After the diagnosis, the user enters the progress of their symptoms and any new symptoms into the feedback form on the user's device and clicks the "Submit" button.

[1305] Step 17:

[1306] The terminal sends the input feedback data to the server as an HTTP POST request.

[1307] Step 18:

[1308] The server receives the feedback data and stores it in a database.

[1309] Step 19:

[1310] The server analyzes the received feedback data and provides it as training data for the generative model.

[1311] Step 20:

[1312] The generative model incorporates new feedback data, updates the learning algorithm, and improves diagnostic accuracy.

[1313] The above is a specific processing flow for making a diagnosis based on vague information and emotions about the user's health and improving accuracy using feedback.

[1314] Example 2

[1315] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1316] Conventional medical systems have difficulty accurately analyzing the vague health information entered by patients and recommending appropriate specialists. Furthermore, patients' emotional state is rarely reflected in diagnoses or specialist selection, potentially reducing diagnostic accuracy and patient satisfaction. Furthermore, there is a lack of mechanisms for effectively utilizing feedback information after diagnosis, which has delayed improvements to the medical system as a whole.

[1317] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1318] In this invention, the server includes: means for receiving ambiguous health-related information and emotional expressions input by a user; analysis means using a generative model and an emotion analysis engine to analyze the received ambiguous health-related information and emotional expressions; means for identifying and recommending an appropriate specialist based on the symptom analysis results and emotion analysis results obtained by the analysis means; means for collecting feedback information from patients after diagnosis and using the collected information to train the generative model and improve the accuracy of the algorithm; and means for automatically generating medical records based on the analysis results and collected information. This makes it possible to analyze ambiguous symptoms and emotional information and recommend the most appropriate specialist, improving patient diagnostic accuracy and satisfaction and promoting overall improvement of the medical system.

[1319] "Ambiguous health information" refers to health information that is expressed in terms or conditions that do not clearly indicate a specific diagnosis or symptom.

[1320] "Emotional expressions" are words or sentences that express the feelings or psychological state that the user is currently experiencing.

[1321] "Means for receiving" refers to a mechanism for acquiring information sent by a user and storing it in a form that can be used within the system.

[1322] A "generative model" is a machine learning algorithm that uses natural language processing techniques to convert vague expressions into specific diagnostic candidates.

[1323] An "emotion analysis engine" is software or an algorithm that analyzes emotional expressions contained in natural language and identifies a user's emotional state.

[1324] "Analysis means" refers to various processing devices within the system for analyzing received ambiguous health-related information and emotional expressions using generative models and emotion analysis engines.

[1325] The "means for identifying and recommending experts" is a mechanism for selecting appropriate medical experts based on the analysis results and presenting that information to the user.

[1326] The "means for collecting feedback information" is a mechanism for collecting information such as changes in condition and new symptoms sent by the user after diagnosis and incorporating it into the system.

[1327] The "means used to train the generative model" refers to a mechanism for updating the generative model based on collected feedback information and improving the accuracy of analysis from the next time onwards.

[1328] "Means for automatically generating medical records" refers to a system for automatically creating medical records (charts) based on the analysis results and collected feedback information.

[1329] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, improving the accuracy of diagnosis, and also recognizes the user's emotions and reflects them in the selection of the recommended specialist.

[1330] System Configuration

[1331] This system is mainly composed of a user terminal, a server, a generative model, a sentiment analysis engine, and an expert recommendation means.

[1332] User terminal operation

[1333] The user device serves to transmit the health information and emotional expressions entered by the user in natural language to the server. For example, a smartphone or computer application can function as a user device. The user uses these devices to input and transmit information such as "I'm worried because my stomach hurts."

[1334] Server Operation

[1335] The server is the central hardware that receives and analyzes information sent from user devices. The server has the following functions:

[1336] 1. Data reception and storage: The server receives vague health-related information and emotional expressions sent from the user terminal and stores them in a temporary database.

[1337] 2. Symptom Analysis: The server passes the received data to a generative model, which uses natural language processing techniques to convert vague expressions into specific diagnosis candidates. This generative model uses advanced natural language processing algorithms such as GPT-3 and BERT.

[1338] 3. Sentiment analysis: At the same time, the server passes the received emotional data to an emotion analysis engine to analyze the user's mental state. For example, tools such as Watson NLU or Hume AI can be used for the emotion analysis engine.

[1339] 4. Expert recommendation: Based on the analysis results, an appropriate expert is selected and related expert information is sent to the user's device. The expert recommendation involves retrieving appropriate expert information from the database and taking into account the user's emotional state.

[1340] Medical record generation and feedback

[1341] The server performs the following processes based on the analysis results and user feedback:

[1342] 1. Medical record generation: Medical records are automatically generated based on the analysis results and sent to the recommended specialists. This process improves the efficiency of doctors' work.

[1343] 2. Feedback collection: After the diagnosis, the user re-enters any changes in their condition and sends them to the server, including information such as "symptoms have improved" or "new symptoms have appeared."

[1344] 3. Learning and updating of generative model: The server updates the generative model based on the collected feedback information to improve the accuracy of the diagnostic algorithm.

[1345] Specific examples

[1346] For example, if a user types "I'm anxious because my stomach hurts" and submits it...

[1347] 1. User device: Enter and send information on the smartphone application.

[1348] 2. Server: Receives input data, analyzes symptom data using a generative model, and emotion data using an emotion analysis engine.

[1349] 3. Server: The generative model converts the symptoms into specific ones such as "stomachache" or "pain in the lower right abdomen," and the emotion analysis engine identifies emotions such as "anxiety" or "worry."

[1350] 4. Server: Retrieves gastroenterologists and gastroenterologists from a database and recommends the appropriate doctor to the patient, taking into account their emotional state.

[1351] 5. User terminal: The patient selects a doctor from the recommended list and receives treatment.

[1352] 6. Server: After diagnosis, the user inputs "symptoms do not improve" and submits it, which receives feedback and updates the generative model.

[1353] This concludes the description of the "Smart Health Link" system, which accurately analyzes vague symptoms and emotions and recommends appropriate specialists, thereby improving the accuracy of diagnoses and patient satisfaction.

[1354] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1355] Step 1: User Input

[1356] Users input vague information about their health and emotional expressions into an input form on their device, such as "I have a stomachache and I'm worried," into a smartphone or computer application.

[1357] Input: Ambiguous health-related information and emotional expressions

[1358] Output: An object that temporarily stores the input information

[1359] Step 2: Send data from the user device to the server

[1360] The user checks the information they have entered and clicks the "Send" button, which causes the user terminal to send the input data to the server.

[1361] Input: Information entered by the user and sent (vague health information and emotional expressions)

[1362] Output: HTTP request with the information sent

[1363] Step 3: Data received and stored by the server

[1364] The server receives the information sent from the user terminal and stores it in a temporary storage database.

[1365] Input: Information sent from the user's device (vague health-related information and emotional expressions)

[1366] Output: Input data stored in a database

[1367] Step 4: Symptom analysis by server

[1368] The server passes the received data to a generative model, which uses natural language processing techniques to convert ambiguous expressions into specific diagnosis candidates. The generative model uses GPT-3, BERT, and other models.

[1369] Input: Vague health information stored in a database

[1370] Output: Results converted into specific symptoms (e.g., "stomach pain," "pain in the lower right abdomen," etc.)

[1371] Step 5: Sentiment analysis on the server

[1372] The server passes the received emotional data to an emotion analysis engine, which may use Watson NLU or Hume AI, to analyze the user's mental state.

[1373] Input: Emotions stored in the database

[1374] Output: Analyzed emotion results (e.g., "anxiety," "worry," etc.)

[1375] Step 6: Server recommends experts

[1376] The server retrieves relevant experts from a database based on the results of symptom and emotion analysis, identifies the most suitable expert based on the analysis results, and sends that information to the user's device.

[1377] Input: Symptom analysis results and emotion analysis results

[1378] Output: Recommended expert information (e.g., "Gastroenterologist")

[1379] Step 7: Generate and send medical records

[1380] The server automatically generates a medical record based on the analysis results and the user's input data and sends it to the recommended specialist.

[1381] Input: Analysis results and user input data

[1382] Output: Automatically generated medical records (charts)

[1383] Step 8: Provide feedback

[1384] After the diagnosis, the user inputs any changes in their condition again, such as feedback information such as "symptoms have improved" or "new symptoms have appeared," and sends it to the server.

[1385] Input: Feedback information after diagnosis

[1386] Output: Feedback information sent to the server

[1387] Step 9: Update learning using feedback

[1388] Based on the received feedback, the server updates the training dataset for the generative model and improves the accuracy of the analysis algorithm.

[1389] Input: Collected feedback information

[1390] Output: Updated generative model and algorithm

[1391] This is the specific program processing flow of this system. This system makes it possible to accurately analyze vague health-related information and emotions and recommend the most appropriate health specialist.

[1392] (Application example 2)

[1393] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1394] Autonomous vehicles require constant monitoring of the health status of drivers and passengers, and prompt and appropriate medical treatment when an abnormality is detected. However, current systems lack the means to analyze health and emotional data in real time and recommend the most appropriate medical specialist. As a result, there is a risk that responses to abnormalities may be delayed, hindering the provision of prompt and appropriate medical care.

[1395] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1396] In this invention, the server includes: means for receiving ambiguous health-related information input by a user; means for analyzing the ambiguous information using a generative model; means for identifying and recommending an appropriate expert based on the obtained analysis results; means for collecting health data and emotion data using sensors and cameras installed in the vehicle; means for analyzing the collected health data and emotion data in real time and recommending an optimal expert when an abnormality is detected; means for collecting post-diagnosis information from the patient and using it as feedback to train the generative model; and means for automatically generating a record based on the analysis results and the collected information and transmitting it to an in-vehicle display or a user terminal. This enables constant monitoring of the health status of passengers in an autonomous vehicle and quickly recommending an appropriate expert when an abnormality is detected.

[1397] "User" means the driver and passengers in the vehicle who input health information into the system.

[1398] "Ambiguous health information" refers to information about a health condition entered by a user that does not specify specific symptoms, and includes information expressed in natural language.

[1399] A "generative model" is a model that uses machine learning algorithms to analyze ambiguous health information and generate specific diagnostic candidates.

[1400] The "analysis means" is a system element that uses a generative model to analyze received ambiguous health information and obtain a diagnosis.

[1401] "Experts" are medical professionals recommended based on diagnostic results and who have expertise in a particular medical field.

[1402] The "means for recommending an expert" is a system element that identifies an appropriate expert based on the results obtained by the analysis means and recommends the expert to the user.

[1403] "Sensors" refer to devices installed in vehicles that collect physiological data such as heart rate and body temperature.

[1404] The "camera" is a video capture device installed inside the vehicle, which is used to analyze the user's facial expressions and behavior and collect emotional data.

[1405] "Means for collecting health and emotional data" refers to system elements that use sensors and cameras to collect information about the user's health and emotional state in real time.

[1406] "Means for real-time analysis" refers to a system element that instantly analyzes collected health and emotional data and responds quickly if an abnormality is detected.

[1407] "When an abnormality is detected" refers to when an abnormal health or emotional state is recognized based on the collected health and emotional data.

[1408] The "means used as feedback for training the generative model" is a system element that incorporates information collected from the user after the diagnosis as a training dataset for the generative model to improve the accuracy of the model.

[1409] "Means for generating records" refers to a system element that automatically generates a medical record based on the analysis results and collected information, and transmits the generated record to a display in the vehicle or to a user terminal.

[1410] This system monitors the health status of passengers in autonomous vehicles in real time and quickly recommends appropriate medical professionals when an abnormality is detected. The system consists of a user terminal, sensors, cameras, a server, a generative model, an emotion engine, and an expert recommendation method. The overall system flow is as follows:

[1411] First, when a user inputs vague health-related information, the information is sent from the user terminal to the server. The user terminal consists of a mobile device used by passengers or an in-vehicle display, and information can be input through a user interface. Examples include inputting information in natural language, such as "I have a headache and feel unwell."

[1412] The server uses a generative model and an emotion engine to analyze the received information. The generative model employs machine learning algorithms to convert vague health information into specific diagnostic candidates. The emotion engine analyzes the input emotion information to identify the user's emotional state. Based on the analysis results, the server identifies the most appropriate medical professional and sends the information to the user's device.

[1413] Next, sensors and cameras installed inside the vehicle collect passenger health and emotional data in real time. The sensors acquire physiological data such as heart rate and body temperature, and the cameras analyze the passenger's facial expression and facial expression. This data is sent to a server, and if an abnormality is detected in real time, it is analyzed and processed to enable rapid response.

[1414] The server automatically generates a medical record based on the analysis results and sends it to the vehicle's display or the user's device. This allows the user to quickly receive remote medical treatment from a recommended specialist. After the diagnosis, the user sends their feedback information back to the server, which is used as training data for the generative model, contributing to improving the model's accuracy.

[1415] As a concrete example, suppose a passenger inputs "I have a stomachache and I'm anxious." This information is sent to the server, and the generative model interprets it as "stomachache" and "anxiety." At the same time, sensors detect an increase in heart rate, and cameras recognize changes in the passenger's complexion. Based on this data, the server recommends a gastroenterologist and provides information to the passenger. The passenger undergoes a remote consultation and then provides feedback, which the server incorporates to update the generative model.

[1416] An example of a prompt might be:

[1417] "You are developing a system that analyzes passengers' health conditions based on data collected from sensors and cameras and recommends appropriate specialists. According to the collected data, Passenger A is experiencing a 'headache' and 'feeling unwell.' Which doctor would you recommend?"

[1418] In this way, the system enables real-time health monitoring and rapid response, providing an effective means of protecting passenger safety and health.

[1419] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1420] Step 1:

[1421] User enters health information

[1422] Using a user terminal, the user inputs vague information about their health (e.g., "I'm worried because my stomach hurts") into the input form and clicks the submit button. The input information is sent from the user terminal to the server.

[1423] Input: Vague health information (natural language)

[1424] Output: Health information sent to the server

[1425] Step 2:

[1426] The server receives the health information.

[1427] The server receives vague health information sent from the user's device, temporarily stores it in a database, and prepares it for the next analysis step.

[1428] Input: Health information sent from the user's device

[1429] Output: Saved health information

[1430] Step 3:

[1431] Analyzing health information with generative models

[1432] The server passes the stored health information to the generative model, which then begins analysis. The generative model uses machine learning algorithms to convert vague information into specific symptoms. For example, the model outputs information like "I have a stomachache and I'm anxious" as specific symptoms like "stomachache" and "anxiety."

[1433] Input: Saved Health Information

[1434] Output: Specific symptoms

[1435] Step 4:

[1436] Analyzing emotional information with an emotion engine

[1437] The server passes the emotional expressions contained in the stored health information to the emotion engine for emotion analysis. The emotion engine uses natural language processing technology to identify the user's emotional state (e.g., anxiety, worry).

[1438] Input: Emotional expressions contained in stored health information

[1439] Output: User's emotional state

[1440] Step 5:

[1441] Recommend an expert

[1442] The server recommends an appropriate expert based on the analysis results of the generative model and the emotion engine. It searches the database for relevant experts, identifies the expert best suited to the user (e.g., a gastroenterologist), and sends the information to the user's terminal.

[1443] Input: specific symptoms, user's emotional state

[1444] Output: Recommended expert information

[1445] Step 6:

[1446] Collecting health data in real time

[1447] Sensors and cameras installed in the vehicle collect passenger health data (e.g., heart rate, body temperature) and emotional data (e.g., facial color, facial expressions) in real time, and the collected data is sent to a server.

[1448] Input: Health and emotion data collected by sensors and cameras

[1449] Output: Real-time data sent to the server

[1450] Step 7:

[1451] Analyzing real-time data

[1452] The server analyzes the health and emotional data sent in real time and performs analytical processing to quickly respond if an abnormality is detected, such as a sudden increase in heart rate or a change in facial color.

[1453] Input: Health and emotional data transmitted in real time

[1454] Output: Anomaly detection results

[1455] Step 8:

[1456] Response when an abnormality is detected

[1457] If the server detects an abnormality, it will immediately recommend an appropriate specialist and send that information to the user's device and the vehicle's display, allowing the user to receive prompt and appropriate medical treatment.

[1458] Input: Anomaly detection result

[1459] Output: Pertinent expert information

[1460] Step 9:

[1461] Gathering feedback and learning

[1462] After the diagnosis, the user sends feedback information about changes in their symptoms to the server, which uses this information as training data for the generative model to improve its accuracy.

[1463] Input: Feedback information

[1464] Output: Updated generative model

[1465] Step 10:

[1466] Automatic recording generation and transmission

[1467] The server automatically generates a medical record based on the analysis results and collected information and sends it to the vehicle's display or user terminal, allowing the user to check the medical record and take appropriate medical action.

[1468] Input: Analysis results and collected information

[1469] Output: Generated medical record

[1470] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1471] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1472] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1473] [Fourth embodiment]

[1474] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1475] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1476] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1477] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1478] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1479] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1480] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1481] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1482] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1483] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1484] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1485] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1486] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1487] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, not only improving the accuracy of diagnoses but also increasing the work efficiency of doctors.

[1488] System Overview

[1489] This system is primarily composed of a user terminal, a server, a generative model, and an expert recommendation means. The user terminal is responsible for sending health information entered by the user to the server. The server receives this information and analyzes it using the generative model. Based on the analysis results, the server recommends an appropriate expert and collects subsequent feedback information to train the generative model.

[1490] Program processing explanation

[1491] User behavior

[1492] 1. Symptom input: The user inputs vague symptoms related to their health into an input form on the user terminal. For example, they input vague information such as "my stomach hurts" or "I feel a bit sick."

[1493] 2. Sending information: After entering the information, the user clicks the "Send" button to send the information to the server.

[1494] Server-side operation

[1495] 1. Information reception: The server receives vague health information sent from the user terminal.

[1496] 2. Analysis: The received information is analyzed using a generative model, which uses natural language processing techniques to convert ambiguous expressions into specific diagnostic candidates.

[1497] 3. Expert recommendation: Based on the analysis results, an appropriate expert (e.g., a gastroenterologist or gastroenterologist) is identified from the database and the information is sent to the user's terminal.

[1498] Generate medical records

[1499] 1. Information storage: The server stores the input information from the user and the analysis results in a database.

[1500] 2. Medical record generation: Medical records (medical records) are automatically generated based on the analysis results and sent to recommended specialists.

[1501] Gathering feedback and learning

[1502] 1. Feedback input: After the diagnosis, the user again inputs how their condition has changed and sends it to the server. For example, information such as "symptoms have improved" or "new symptoms have appeared."

[1503] 2. Learning update: Based on the received feedback, the server updates the training dataset of the generative model to improve the accuracy of the diagnostic algorithm.

[1504] Specific examples

[1505] Example 1: Patient complaining of abdominal pain

[1506] 1. User terminal: The patient types "My stomach hurts" and sends it.

[1507] 2. Server: Receives input data and analyzes it with the generative model.

[1508] 3. Server: The generative model converts the information into specific symptoms such as "stomachache," "pain in the lower right abdomen," and "nausea."

[1509] 4. Server: Retrieves gastroenterologists and gastroenterologists from the database and sends them to the user terminal.

[1510] 5. User terminal: The patient selects a doctor from the recommended list.

[1511] 6. Server: Receives the selected information, automatically generates a medical record, and sends it to the doctor.

[1512] 7. User terminal: After diagnosis, the patient enters and submits "Symptoms are not improving."

[1513] 8. Server: Receives feedback and updates the generative model.

[1514] This concludes the description of the embodiment of the present invention. This system makes it possible to accurately analyze a patient's vague symptoms and recommend the most appropriate specialist. It also improves the work efficiency of doctors, enabling faster and more reliable medical care.

[1515] The processing flow will be explained below.

[1516] Step 1:

[1517] The user enters vague health symptoms into the input form on the user terminal and clicks the "Submit" button.

[1518] Step 2:

[1519] The device sends vague health information entered by the user to the server as an HTTP POST request.

[1520] Step 3:

[1521] The server receives vague health information from the user terminal and temporarily stores it in a database.

[1522] Step 4:

[1523] The server passes the ambiguous information it receives to the natural language processing engine and generative model, and begins the analysis process.

[1524] Step 5:

[1525] A generative model analyzes ambiguous information and identifies likely specific diagnoses (e.g., "stomach pain," "pain in the right lower abdomen," "nausea").

[1526] Step 6:

[1527] Based on the analysis results, the server retrieves relevant specialists (e.g., gastroenterologists, gastroenterologists, etc.) from the database.

[1528] Step 7:

[1529] The server ranks the acquired expert information and sends it to the user's terminal as an HTTP response.

[1530] Step 8:

[1531] The terminal displays the expert information received from the server in list form on the user interface.

[1532] Step 9:

[1533] The user selects the desired expert from the displayed list of experts and clicks the "Select" button.

[1534] Step 10:

[1535] The device sends the information of the selected expert to the server as an HTTP POST request.

[1536] Step 11:

[1537] The server receives the selections from the user and stores them in a database.

[1538] Step 12:

[1539] The server generates a medical record based on the analysis results and the selected expert information.

[1540] Step 13:

[1541] The server sends the generated medical record to the corresponding specialist's terminal as an HTTP response.

[1542] Step 14:

[1543] After the diagnosis, the user enters the progress of their symptoms and any new symptoms into the feedback form on the user's device and clicks the "Submit" button.

[1544] Step 15:

[1545] The terminal sends the input feedback data to the server as an HTTP POST request.

[1546] Step 16:

[1547] The server receives the feedback data and stores it in a database.

[1548] Step 17:

[1549] The server analyzes the received feedback data and provides it as training data for the generative model.

[1550] Step 18:

[1551] The generative model incorporates new feedback data, updates the learning algorithm, and improves diagnostic accuracy.

[1552] The above is a specific processing flow for making a diagnosis based on vague information about the user's health and improving accuracy using feedback.

[1553] Example 1

[1554] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1555] In conventional health support systems, the health information entered by users is vague, making it difficult to make accurate diagnoses or recommend appropriate specialists. Furthermore, feedback information after diagnosis is not fully utilized, making it difficult to improve the accuracy of the system. Furthermore, the creation of medical records is not automated, increasing the workload of medical professionals.

[1556] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1557] In this invention, the server includes a means for receiving ambiguous health-related information input by a user, an analysis means using a generative model to analyze the received ambiguous health-related information, and a means for identifying and recommending an appropriate specialist based on the analysis results obtained by the analysis means, thereby making it possible to convert ambiguous health-related information into specific diagnosis candidates and recommend an appropriate specialist.

[1558] A "user terminal" is a device that allows a user to input health-related information and transmit that information to a server.

[1559] The "server" is the central computer of the system that receives information sent from user terminals, analyzes it, and recommends experts.

[1560] A "generative model" is an algorithm that uses natural language processing techniques to analyze ambiguous information received and convert it into specific diagnostic candidates.

[1561] "Analysis means" refers to the process of using a generative model to analyze received ambiguous information and generate specific diagnosis candidates.

[1562] The "expert recommendation means" is a means for identifying an appropriate expert from a database based on the analysis results and providing that information.

[1563] "Information storage means" refers to the process of storing user input information and analysis results in a database.

[1564] "Medical record generation means" refers to the process of automatically generating a medical record based on the analysis results and sending it to the recommended specialist.

[1565] "Feedback input means" refers to a process in which the user re-enters information about changes in his or her condition after diagnosis and sends that information to the server.

[1566] The "learning update means" refers to the process of adding received feedback information to the learning dataset of the generative model to improve the accuracy of the model.

[1567] "Ambiguous information" refers to information entered by a user in vague terms, without clearly describing specific symptoms or conditions.

[1568] "Analysis results" refer to specific diagnosis candidates and expert recommendation information analyzed by the generative model.

[1569] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, not only improving the accuracy of diagnoses but also increasing the work efficiency of doctors.

[1570] System Overview

[1571] This system is primarily composed of a user terminal, a server, a generative model, and an expert recommendation means. The user terminal is responsible for sending health information entered by the user to the server. The server receives this information and analyzes it using the generative model. Based on the analysis results, the server recommends an appropriate expert and collects subsequent feedback information to train the generative model.

[1572] User behavior

[1573] The user enters vague symptoms related to their health into an input form on the user terminal. Specifically, they enter vague information such as "my stomach hurts" or "I feel a bit sick." Then, the user clicks the "send" button to send the information to the server.

[1574] Server-side operation

[1575] Receiving information

[1576] The server receives vague health information sent from the user's device. The server's API receives the request from the device and adds the text data to a processing queue.

[1577] Analysis means

[1578] The server analyzes the received information using a generative model. This model uses natural language processing technology to convert vague expressions into specific diagnosis candidates. For example, if an input says "my stomach hurts," the generative model converts that into specific symptoms such as "stomachache," "pain in the right lower abdomen," and "nausea."

[1579] Expert recommendation

[1580] Based on the analysis results, the server identifies an appropriate expert from the database. The expert's information is returned to the user's terminal. The server sends the analysis results and the expert's information to the terminal in response to the request.

[1581] Generate medical records

[1582] Information storage

[1583] The server stores the user's input information and analysis results in a database, which then becomes historical data that can be referenced later.

[1584] Medical record generation

[1585] The server automatically generates a medical record (chart) based on the analysis results, which is then sent to the recommended specialist.

[1586] Gathering feedback and learning

[1587] Feedback Input

[1588] After the diagnosis, the user again inputs how their condition has changed and sends it to the server. For example, they input feedback information such as "symptoms have improved" or "new symptoms have appeared."

[1589] Learning Update

[1590] The server uses the received feedback to update the training dataset of the generative model, thereby improving the accuracy of the diagnostic algorithm.

[1591] Specific examples

[1592] Example 1: Patient complaining of abdominal pain

[1593] 1. Symptom input:

[1594] The user inputs "My stomach hurts" into the user terminal and sends it.

[1595] 2. Receiving information:

[1596] The server receives this input data and analyzes it with the generative model.

[1597] 3. Analysis method:

[1598] The generative model converts the input "my stomach hurts" into specific symptoms such as "stomachache," "pain in the lower right abdomen," and "nausea."

[1599] 4. Expert recommendation:

[1600] The server retrieves gastroenterology and gastroenterology specialists from a database and transmits them to the user terminal.

[1601] 5. Expert Selection:

[1602] The user selects a doctor from a list of recommendations.

[1603] 6. Medical record generation:

[1604] The server receives the selection information, automatically generates a medical record, and sends it to the doctor.

[1605] 7. Feedback Input:

[1606] After the diagnosis, the user enters "Symptoms are not improving" and submits the message.

[1607] 8. Learning Update:

[1608] The server receives the feedback and updates the generative model.

[1609] Prompt Sentence Examples

[1610] For the input "My stomach hurts," the generative AI model generates the following analysis result:

[1611] Entered information: "My stomach hurts"

[1612] Analysis results: "Stomach pain," "pain in the lower right abdomen," "nausea"

[1613] Appropriate specialist: "Gastroenterologist" "Gastroenterologist"

[1614] This makes it possible to turn vague symptoms into specific candidate diagnoses and identify appropriate specialists.

[1615] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1616] Step 1: Enter your symptoms

[1617] The user inputs vague information about their symptoms into an input form on the terminal, such as "my stomach hurts" or "I feel a bit sick."

[1618] Specifically, the user uses the terminal keyboard to input symptoms into a text box.

[1619] Input: Vague health information entered by the user.

[1620] Output: Symptom information entered into the user's terminal.

[1621] Step 2: Send information

[1622] The user checks the information they entered and clicks the "Send" button, which causes the user's device to send the data to the server.

[1623] Specifically, the terminal sends the input text data as an HTTP request to the server's API.

[1624] Input: The symptom information the user entered and clicked submit.

[1625] Output: Symptom information sent to the server.

[1626] Step 3: Receiving information

[1627] The server receives ambiguous health information transmitted from the user terminal.

[1628] Specifically, the server's API receives a request from the terminal and adds the text data to a processing queue.

[1629] Input: Symptom information sent from the user terminal.

[1630] Output: Symptom information received.

[1631] Step 4: Analysis methods

[1632] The server analyzes the received information using a generative model, which uses natural language processing techniques to convert ambiguous expressions into specific candidate diagnoses.

[1633] Specifically, the generative AI model analyzes input data and converts vague expressions into specific diagnostic candidates such as "stomachache," "pain in the lower right abdomen," and "nausea."

[1634] Input: Received symptom information.

[1635] Output: A list of specific diagnostic candidates.

[1636] Step 5: Expert Recommendations

[1637] Based on the analysis results, the server identifies appropriate experts from the database, and the recommended expert information is sent to the user terminal.

[1638] Specifically, the server compares the analysis results with a list of experts, generates a list of appropriate experts, and returns it to the user terminal.

[1639] Input: A specific list of diagnostic candidates.

[1640] Output: A list of recommended experts.

[1641] Step 6: Save your information

[1642] The server stores the input information from the user and the analysis results in a database.

[1643] Specifically, the server stores the input information and analysis results in a database table.

[1644] Input: Information entered by the user, analysis results.

[1645] Output: Saved data.

[1646] Step 7: Create a medical record

[1647] The server automatically generates a medical record (chart) based on the analysis results and sends it to the recommended specialist.

[1648] Specifically, the server generates medical records in accordance with the medical record format and sends them to specialists via email or a dedicated system.

[1649] Input: User input information, analysis results.

[1650] Output: Generated medical chart, medical record sent to specialist.

[1651] Step 8: Enter your feedback

[1652] After the diagnosis, the user again inputs any changes in their condition and sends it to the server. For example, they can input information such as "symptoms have improved" or "new symptoms have appeared."

[1653] As a specific operation, the user inputs feedback information using the keyboard of the terminal and clicks the send button.

[1654] Input: User feedback information.

[1655] Output: Feedback information sent by the user.

[1656] Step 9: Learning Update

[1657] The server adds the received feedback information to the training dataset of the generative model and retrains the model.

[1658] Specifically, the server adds feedback information to the dataset and retrains the generative AI model to improve its accuracy.

[1659] Input: Feedback information from the user.

[1660] Output: An updated generative model.

[1661] (Application example 1)

[1662] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1663] Conventional systems for analyzing ambiguous health-related information are limited to the function of recommending experts, and lack mechanisms for effectively collecting and learning from feedback information after recommendations and improving the accuracy of generative models. Furthermore, there is a lack of systems for quickly recommending appropriate products and experts for customer support in brick-and-mortar stores, making it difficult to improve customer satisfaction and ensure appropriate health management in brick-and-mortar stores.

[1664] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1665] In this invention, the server includes: means for receiving ambiguous health-related information entered by a user; analysis means using a generative model to analyze the received ambiguous health-related information; means for identifying and recommending an appropriate expert or product based on the analysis results obtained by the analysis means; means for collecting information after the customer has used the recommended expert or product and using the collected information as feedback to train the generative model; and means for automatically generating records based on the analysis results and the collected information. This enables accurate analysis of ambiguous health information in physical stores and rapid recommendation of appropriate experts and products, thereby improving customer satisfaction and streamlining health management in stores.

[1666] "User" means a customer who uses the System to input their health information.

[1667] "Ambiguous health information" refers to information about a condition in which the user is experiencing some kind of discomfort or symptom but cannot specifically describe it.

[1668] "Means for receiving" refers to the function by which the server receives vague health-related information sent from the user terminal.

[1669] "Analysis means" refers to the function that converts received ambiguous health information into specific expert and product recommendations using a generative model.

[1670] A "generative model" is an AI model that uses machine learning algorithms to analyze input information and generate specific diagnostic candidates and recommendations.

[1671] An "expert" is someone who has knowledge and skills in a particular field and can provide appropriate advice and diagnoses to users.

[1672] "Means for identifying and recommending" refers to a function that notifies users of appropriate experts and products based on the results obtained by the analysis means.

[1673] "Feedback" refers to information entered by users about the effects or changes they have experienced after using a recommended professional or product.

[1674] "Means used for learning" refers to the function of using collected feedback information as training data for a generative model to improve the accuracy of the model.

[1675] "Means for generating records" refers to a function that automatically creates medical records and advice records based on analysis results and feedback information.

[1676] "System" refers to a comprehensive mechanism that includes all of the above means and functions and is used to analyze vague information about a user's health and recommend appropriate experts and products.

[1677] The following describes in detail an embodiment of the "Smart Health Assistant" of the present invention. The present invention is a system that identifies the most suitable specialist or product based on vague health information entered by the user, makes more accurate recommendations, and collects feedback after the recommendation to improve the accuracy of the model.

[1678] System Overview

[1679] This system consists of a user terminal, a server, a generative model, and a database. The user terminal is responsible for sending health information entered by the user to the server. The server receives the information and analyzes it using the generative model. Based on the analysis results, the server recommends appropriate experts or products, and then collects subsequent feedback information to train the generative model.

[1680] User behavior

[1681] 1. Symptom input: The user inputs vague symptoms related to their health into an input form on the user device (e.g., smart glasses). For example, they input vague information such as "I have a headache" or "My skin is itchy."

[1682] 2. Sending information: After entering the information, the user clicks the "Send" button to send the information to the server.

[1683] Server-side operation

[1684] 1. Information reception: The server receives vague health information sent from the user terminal.

[1685] 2. Analysis method: The received information is analyzed using a generative model (e.g., OpenAI GPT-4). This generative model uses natural language processing technology to convert ambiguous expressions into specific diagnosis candidates and recommendations.

[1686] 3. Expert recommendation: Based on the analysis results, the appropriate expert (pharmacist, health advisor, etc.) or product (painkiller, moisturizing cream, etc.) is identified from the database and the information is sent to the user's device.

[1687] Generate medical records

[1688] 1. Information storage: The server stores the input information from the user and the analysis results in a database.

[1689] 2. Medical record generation: A medical record (medical record) or advice record is automatically generated based on the analysis results and sent to the recommended specialist.

[1690] Gathering feedback and learning

[1691] 1. Feedback input: The user re-enters information about the effects and changes in condition after using the recommended specialist or product, and sends it to the server. For example, information such as "Symptoms improved" or "Insufficient effect."

[1692] 2. Learning update: Based on the received feedback, the server updates the training dataset of the generative model to improve the accuracy of the diagnosis and recommendation algorithm.

[1693] Specific examples

[1694] Example 1: A customer complaining of a headache

[1695] 1. Smart Glasses: The customer types "I have a headache" and submits.

[1696] 2. Server: Receives input data and analyzes it with the generative model.

[1697] 3. Server: The generative model converts the data into specific diagnosis candidates such as "tension headache" or "stress."

[1698] 4. Server: Retrieves the appropriate pharmacist or painkiller from the database and sends it to the user's smart glasses.

[1699] 5. Smart Glasses: Customers select a pharmacist or product from a list of recommendations.

[1700] 6. Server: Receives the selected information, automatically generates a medical record, and sends it to the pharmacist.

[1701] 7. Smart glasses: The customer types "my symptoms have improved" and submits.

[1702] 8. Server: Receives feedback and updates the generative model.

[1703] Prompt Sentence Examples

[1704] "Enter any vague health symptoms you have. For example, I have a headache, my skin is itchy, etc."

[1705] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1706] Step 1: Enter your symptoms

[1707] Users enter vague health symptoms into a form on the smart glasses in a physical store. Examples of information include "I have a headache" or "My skin is itchy." This input is the starting point for the system, which then identifies the user's complaint in concrete terms.

[1708] Step 2: Send information

[1709] After inputting the information, the user clicks the "Send" button. The smart glasses then send the input information to the store's server. The transmitted data includes the symptom information entered by the user. The server then passes the received information to the next analysis step.

[1710] Step 3: Receiving information

[1711] The server receives the vague health information sent by the smart glasses, stores the received data in preparation for analysis, and passes it on to the next processing step.

[1712] Step 4: Analysis

[1713] The server analyzes the received information using a generative model (e.g., OpenAI GPT-4). The input includes the symptom information submitted by the user. The generative model analyzes the ambiguous information and generates specific diagnosis candidates and recommendations. The output includes specific diagnosis candidates such as "tension headache" and "stress."

[1714] Step 5: Expert or product recommendation

[1715] The server identifies appropriate experts (pharmacists, health advisors, etc.) or products (painkillers, moisturizing creams, etc.) from a database based on the analysis results obtained by the generative model. The input includes the analysis results, and the server selects the best recommendation destination based on these results. The output includes a list of recommended experts and products.

[1716] Step 6: Submit your recommendation

[1717] The server sends the identified expert or product information to the user's smart glasses. The input includes the identified recommendation information, which is sent to notify the user. The output includes a recommendation list that is displayed on the user's device.

[1718] Step 7: Enter your feedback

[1719] After receiving expert advice or using the product, the user again enters feedback about the effects and changes in their physical condition into the smart glasses' input form. For example, feedback such as "Symptoms improved" or "Not effective enough" is entered. This input information is used for subsequent learning.

[1720] Step 8: Send your feedback

[1721] After entering the feedback information, the user clicks the "Send" button. The smart glasses then send this feedback information to the store's server. The transmitted data includes the feedback information. The server then passes the received information to the next learning process.

[1722] Step 9: Receiving feedback

[1723] The server receives the feedback information sent from the smart glasses, stores the received data in a training dataset, and passes it on to the next processing step.

[1724] Step 10: Retraining the generative model

[1725] The server retrains the generative model based on the received feedback. The input includes the feedback, which improves the accuracy of the generative model. The output includes the updated generative model.

[1726] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1727] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, improving the accuracy of diagnosis, and also recognizes the user's emotions and reflects them in the selection of the recommended specialist.

[1728] System Overview

[1729] This system is primarily composed of a user terminal, a server, a generative model, an emotion engine, and an expert recommendation means. The user terminal is responsible for sending health information and natural language emotion expressions entered by the user to the server. The server receives this information and analyzes it using the generative model and emotion engine. Based on the analysis results, the server recommends an appropriate expert and collects subsequent feedback information to train the generative model.

[1730] Program processing explanation

[1731] User behavior

[1732] 1. Symptom and emotion input: The user inputs vague symptoms related to their health and the emotions associated with them into an input form on the user terminal. For example, they input information such as "I have a stomachache and I'm worried" or "I feel a bit sick and worried."

[1733] 2. Sending information: After inputting, the user clicks the "Send" button to send the symptom and emotion information to the server.

[1734] Server-side operation

[1735] 1. Information reception: The server receives vague health information and emotion information sent from the user terminal and temporarily stores them in a database.

[1736] 2. Symptom analysis: The received information is passed to a natural language processing engine and a generative model to begin the symptom analysis process. This generative model uses natural language processing techniques to convert ambiguous expressions into specific diagnostic candidates.

[1737] 3. Emotion analysis means: At the same time, the emotion engine analyzes the received emotion information to grasp the user's mental state.

[1738] 4. Expert recommendation: Based on the results of symptom analysis and emotion analysis, relevant experts (e.g., gastroenterologists) are retrieved from a database, and the most suitable expert is identified taking into account emotion information, and this information is sent to the user's device.

[1739] Generate medical records

[1740] 1. Information storage: The server stores the input information from the user and the analysis results in a database.

[1741] 2. Medical record generation: Medical records (medical records) are automatically generated based on the analysis results and sent to recommended specialists.

[1742] Gathering feedback and learning

[1743] 1. Feedback input: After the diagnosis, the user again inputs how their condition has changed and sends it to the server. For example, they input feedback information such as "Symptoms have improved" or "New symptoms have appeared."

[1744] 2. Learning update: Based on the received feedback, the server updates the training dataset of the generative model to improve the accuracy of the diagnostic algorithm.

[1745] Specific examples

[1746] Example 1: Patient complaining of abdominal pain

[1747] 1. User terminal: The patient types, "I have a stomachache and I'm worried" and sends it.

[1748] 2. Server: Receives input data and analyzes symptom data using a generative model and emotion data using an emotion engine.

[1749] 3. Server: The generative model converts the symptoms into specific ones such as "stomachache," "pain in the lower right abdomen," and "nausea," and the emotion engine identifies emotions such as "anxiety" and "worry."

[1750] 4. Server: Retrieves gastroenterologists and gastroenterologists from a database, identifies the appropriate doctor for the patient, taking into account their emotional state, and sends the information to the user's terminal.

[1751] 5. User terminal: The patient selects a doctor from the recommended list.

[1752] 6. Server: Receives the selected information, automatically generates a medical record, and sends it to the doctor.

[1753] 7. User terminal: After diagnosis, the patient enters and submits "Symptoms are not improving."

[1754] 8. Server: Receives feedback and updates the generative model.

[1755] This concludes the description of the embodiment of the present invention. This system makes it possible to accurately analyze a patient's vague symptoms and emotions and recommend the most appropriate specialist. It also improves the work efficiency of doctors, enabling faster and more reliable medical care.

[1756] The processing flow will be explained below.

[1757] Step 1:

[1758] The user enters vague health symptoms and associated feelings into an input form on the user's device and clicks the "Submit" button.

[1759] Step 2:

[1760] The terminal sends the ambiguous health information and emotion information input by the user to the server as an HTTP POST request.

[1761] Step 3:

[1762] The server receives the information sent from the user terminal and temporarily stores it in a database.

[1763] Step 4:

[1764] The server passes the received ambiguous information to a natural language processing engine and a generative model, and begins the symptom analysis process.

[1765] Step 5:

[1766] A generative model analyzes ambiguous information and identifies likely specific diagnoses (e.g., "stomach pain," "pain in the right lower abdomen," "nausea").

[1767] Step 6:

[1768] The server passes the received emotion information to the emotion engine and starts emotion analysis processing.

[1769] Step 7:

[1770] The emotion engine analyzes the emotional information and identifies the user's mental state (e.g., "anxiety" or "worry").

[1771] Step 8:

[1772] The server retrieves relevant specialists (e.g., gastroenterologists, gastroenterologists, etc.) from a database based on the symptom analysis and emotion analysis results.

[1773] Step 9:

[1774] The server optimizes the expert information it acquires, taking into account the results of sentiment analysis, and sends it to the user's device as an HTTP response.

[1775] Step 10:

[1776] The terminal displays the expert information received from the server in list form on the user interface.

[1777] Step 11:

[1778] The user selects the desired expert from the displayed list of experts and clicks the "Select" button.

[1779] Step 12:

[1780] The device sends the information of the selected expert to the server as an HTTP POST request.

[1781] Step 13:

[1782] The server receives the selections from the user and stores them in a database.

[1783] Step 14:

[1784] The server generates a medical record based on the symptom analysis results and the selected specialist information.

[1785] Step 15:

[1786] The server sends the generated medical record to the corresponding specialist's terminal as an HTTP response.

[1787] Step 16:

[1788] After the diagnosis, the user enters the progress of their symptoms and any new symptoms into the feedback form on the user's device and clicks the "Submit" button.

[1789] Step 17:

[1790] The terminal sends the input feedback data to the server as an HTTP POST request.

[1791] Step 18:

[1792] The server receives the feedback data and stores it in a database.

[1793] Step 19:

[1794] The server analyzes the received feedback data and provides it as training data for the generative model.

[1795] Step 20:

[1796] The generative model incorporates new feedback data, updates the learning algorithm, and improves diagnostic accuracy.

[1797] The above is a specific processing flow for making a diagnosis based on vague information and emotions about the user's health and improving accuracy using feedback.

[1798] Example 2

[1799] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1800] Conventional medical systems have difficulty accurately analyzing the vague health information entered by patients and recommending appropriate specialists. Furthermore, patients' emotional state is rarely reflected in diagnoses or specialist selection, potentially reducing diagnostic accuracy and patient satisfaction. Furthermore, there is a lack of mechanisms for effectively utilizing feedback information after diagnosis, which has delayed improvements to the medical system as a whole.

[1801] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1802] In this invention, the server includes: means for receiving ambiguous health-related information and emotional expressions input by a user; analysis means using a generative model and an emotion analysis engine to analyze the received ambiguous health-related information and emotional expressions; means for identifying and recommending an appropriate specialist based on the symptom analysis results and emotion analysis results obtained by the analysis means; means for collecting feedback information from patients after diagnosis and using the collected information to train the generative model and improve the accuracy of the algorithm; and means for automatically generating medical records based on the analysis results and collected information. This makes it possible to analyze ambiguous symptoms and emotional information and recommend the most appropriate specialist, improving patient diagnostic accuracy and satisfaction and promoting overall improvement of the medical system.

[1803] "Ambiguous health information" refers to health information that is expressed in terms or conditions that do not clearly indicate a specific diagnosis or symptom.

[1804] "Emotional expressions" are words or sentences that express the feelings or psychological state that the user is currently experiencing.

[1805] "Means for receiving" refers to a mechanism for acquiring information sent by a user and storing it in a form that can be used within the system.

[1806] A "generative model" is a machine learning algorithm that uses natural language processing techniques to convert vague expressions into specific diagnostic candidates.

[1807] An "emotion analysis engine" is software or an algorithm that analyzes emotional expressions contained in natural language and identifies a user's emotional state.

[1808] "Analysis means" refers to various processing devices within the system for analyzing received ambiguous health-related information and emotional expressions using generative models and emotion analysis engines.

[1809] The "means for identifying and recommending experts" is a mechanism for selecting appropriate medical experts based on the analysis results and presenting that information to the user.

[1810] The "means for collecting feedback information" is a mechanism for collecting information such as changes in condition and new symptoms sent by the user after diagnosis and incorporating it into the system.

[1811] The "means used to train the generative model" refers to a mechanism for updating the generative model based on collected feedback information and improving the accuracy of analysis from the next time onwards.

[1812] "Means for automatically generating medical records" refers to a system for automatically creating medical records (charts) based on the analysis results and collected feedback information.

[1813] The following describes in detail an embodiment of the present invention, "Smart Health Link." The present invention is a system that identifies the most suitable specialist based on vague health information entered by the user, improving the accuracy of diagnosis, and also recognizes the user's emotions and reflects them in the selection of the recommended specialist.

[1814] System Configuration

[1815] This system is mainly composed of a user terminal, a server, a generative model, a sentiment analysis engine, and an expert recommendation means.

[1816] User terminal operation

[1817] The user device serves to transmit the health information and emotional expressions entered by the user in natural language to the server. For example, a smartphone or computer application can function as a user device. The user uses these devices to input and transmit information such as "I'm worried because my stomach hurts."

[1818] Server Operation

[1819] The server is the central hardware that receives and analyzes information sent from user devices. The server has the following functions:

[1820] 1. Data reception and storage: The server receives vague health-related information and emotional expressions sent from the user terminal and stores them in a temporary database.

[1821] 2. Symptom Analysis: The server passes the received data to a generative model, which uses natural language processing techniques to convert vague expressions into specific diagnosis candidates. This generative model uses advanced natural language processing algorithms such as GPT-3 and BERT.

[1822] 3. Sentiment analysis: At the same time, the server passes the received emotional data to an emotion analysis engine to analyze the user's mental state. For example, tools such as Watson NLU or Hume AI can be used for the emotion analysis engine.

[1823] 4. Expert recommendation: Based on the analysis results, an appropriate expert is selected and related expert information is sent to the user's device. The expert recommendation involves retrieving appropriate expert information from the database and taking into account the user's emotional state.

[1824] Medical record generation and feedback

[1825] The server performs the following processes based on the analysis results and user feedback:

[1826] 1. Medical record generation: Medical records are automatically generated based on the analysis results and sent to the recommended specialists. This process improves the efficiency of doctors' work.

[1827] 2. Feedback collection: After the diagnosis, the user re-enters any changes in their condition and sends them to the server, including information such as "symptoms have improved" or "new symptoms have appeared."

[1828] 3. Learning and updating of generative model: The server updates the generative model based on the collected feedback information to improve the accuracy of the diagnostic algorithm.

[1829] Specific examples

[1830] For example, if a user types "I'm anxious because my stomach hurts" and submits it...

[1831] 1. User device: Enter and send information on the smartphone application.

[1832] 2. Server: Receives input data, analyzes symptom data using a generative model, and emotion data using an emotion analysis engine.

[1833] 3. Server: The generative model converts the symptoms into specific ones such as "stomachache" or "pain in the lower right abdomen," and the emotion analysis engine identifies emotions such as "anxiety" or "worry."

[1834] 4. Server: Retrieves gastroenterologists and gastroenterologists from a database and recommends the appropriate doctor to the patient, taking into account their emotional state.

[1835] 5. User terminal: The patient selects a doctor from the recommended list and receives treatment.

[1836] 6. Server: After diagnosis, the user inputs "symptoms do not improve" and submits it, which receives feedback and updates the generative model.

[1837] This concludes the description of the "Smart Health Link" system, which accurately analyzes vague symptoms and emotions and recommends appropriate specialists, thereby improving the accuracy of diagnoses and patient satisfaction.

[1838] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1839] Step 1: User Input

[1840] Users input vague information about their health and emotional expressions into an input form on their device, such as "I have a stomachache and I'm worried," into a smartphone or computer application.

[1841] Input: Ambiguous health-related information and emotional expressions

[1842] Output: An object that temporarily stores the input information

[1843] Step 2: Send data from the user device to the server

[1844] The user checks the information they have entered and clicks the "Send" button, which causes the user terminal to send the input data to the server.

[1845] Input: Information entered by the user and sent (vague health information and emotional expressions)

[1846] Output: HTTP request with the information sent

[1847] Step 3: Data received and stored by the server

[1848] The server receives the information sent from the user terminal and stores it in a temporary storage database.

[1849] Input: Information sent from the user's device (vague health-related information and emotional expressions)

[1850] Output: Input data stored in a database

[1851] Step 4: Symptom analysis by server

[1852] The server passes the received data to a generative model, which uses natural language processing techniques to convert ambiguous expressions into specific diagnosis candidates. The generative model uses GPT-3, BERT, and other models.

[1853] Input: Vague health information stored in a database

[1854] Output: Results converted into specific symptoms (e.g., "stomach pain," "pain in the lower right abdomen," etc.)

[1855] Step 5: Sentiment analysis on the server

[1856] The server passes the received emotional data to an emotion analysis engine, which may use Watson NLU or Hume AI, to analyze the user's mental state.

[1857] Input: Emotions stored in the database

[1858] Output: Analyzed emotion results (e.g., "anxiety," "worry," etc.)

[1859] Step 6: Server recommends experts

[1860] The server retrieves relevant experts from a database based on the results of symptom and emotion analysis, identifies the most suitable expert based on the analysis results, and sends that information to the user's device.

[1861] Input: Symptom analysis results and emotion analysis results

[1862] Output: Recommended expert information (e.g., "Gastroenterologist")

[1863] Step 7: Generate and send medical records

[1864] The server automatically generates a medical record based on the analysis results and the user's input data and sends it to the recommended specialist.

[1865] Input: Analysis results and user input data

[1866] Output: Automatically generated medical records (charts)

[1867] Step 8: Provide feedback

[1868] After the diagnosis, the user inputs any changes in their condition again, such as feedback information such as "symptoms have improved" or "new symptoms have appeared," and sends it to the server.

[1869] Input: Feedback information after diagnosis

[1870] Output: Feedback information sent to the server

[1871] Step 9: Update learning using feedback

[1872] Based on the received feedback, the server updates the training dataset for the generative model and improves the accuracy of the analysis algorithm.

[1873] Input: Collected feedback information

[1874] Output: Updated generative model and algorithm

[1875] This is the specific program processing flow of this system. This system makes it possible to accurately analyze vague health-related information and emotions and recommend the most appropriate health specialist.

[1876] (Application example 2)

[1877] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1878] Autonomous vehicles require constant monitoring of the health status of drivers and passengers, and prompt and appropriate medical treatment when an abnormality is detected. However, current systems lack the means to analyze health and emotional data in real time and recommend the most appropriate medical specialist. As a result, there is a risk that responses to abnormalities may be delayed, hindering the provision of prompt and appropriate medical care.

[1879] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1880] In this invention, the server includes: means for receiving ambiguous health-related information input by a user; means for analyzing the ambiguous information using a generative model; means for identifying and recommending an appropriate expert based on the obtained analysis results; means for collecting health data and emotion data using sensors and cameras installed in the vehicle; means for analyzing the collected health data and emotion data in real time and recommending an optimal expert when an abnormality is detected; means for collecting post-diagnosis information from the patient and using it as feedback to train the generative model; and means for automatically generating a record based on the analysis results and the collected information and transmitting it to an in-vehicle display or a user terminal. This enables constant monitoring of the health status of passengers in an autonomous vehicle and quickly recommending an appropriate expert when an abnormality is detected.

[1881] "User" means the driver and passengers in the vehicle who input health information into the system.

[1882] "Ambiguous health information" refers to information about a health condition entered by a user that does not specify specific symptoms, and includes information expressed in natural language.

[1883] A "generative model" is a model that uses machine learning algorithms to analyze ambiguous health information and generate specific diagnostic candidates.

[1884] The "analysis means" is a system element that uses a generative model to analyze received ambiguous health information and obtain a diagnosis.

[1885] "Experts" are medical professionals recommended based on diagnostic results and who have expertise in a particular medical field.

[1886] The "means for recommending an expert" is a system element that identifies an appropriate expert based on the results obtained by the analysis means and recommends the expert to the user.

[1887] "Sensors" refer to devices installed in vehicles that collect physiological data such as heart rate and body temperature.

[1888] The "camera" is a video capture device installed inside the vehicle, which is used to analyze the user's facial expressions and behavior and collect emotional data.

[1889] "Means for collecting health and emotional data" refers to system elements that use sensors and cameras to collect information about the user's health and emotional state in real time.

[1890] "Means for real-time analysis" refers to a system element that instantly analyzes collected health and emotional data and responds quickly if an abnormality is detected.

[1891] "When an abnormality is detected" refers to when an abnormal health or emotional state is recognized based on the collected health and emotional data.

[1892] The "means used as feedback for training the generative model" is a system element that incorporates information collected from the user after the diagnosis as a training dataset for the generative model to improve the accuracy of the model.

[1893] "Means for generating records" refers to a system element that automatically generates a medical record based on the analysis results and collected information, and transmits the generated record to a display in the vehicle or to a user terminal.

[1894] This system monitors the health status of passengers in autonomous vehicles in real time and quickly recommends appropriate medical professionals when an abnormality is detected. The system consists of a user terminal, sensors, cameras, a server, a generative model, an emotion engine, and an expert recommendation method. The overall system flow is as follows:

[1895] First, when a user inputs vague health-related information, the information is sent from the user terminal to the server. The user terminal consists of a mobile device used by passengers or an in-vehicle display, and information can be input through a user interface. Examples include inputting information in natural language, such as "I have a headache and feel unwell."

[1896] The server uses a generative model and an emotion engine to analyze the received information. The generative model employs machine learning algorithms to convert vague health information into specific diagnostic candidates. The emotion engine analyzes the input emotion information to identify the user's emotional state. Based on the analysis results, the server identifies the most appropriate medical professional and sends the information to the user's device.

[1897] Next, sensors and cameras installed inside the vehicle collect passenger health and emotional data in real time. The sensors acquire physiological data such as heart rate and body temperature, and the cameras analyze the passenger's facial expression and facial expression. This data is sent to a server, and if an abnormality is detected in real time, it is analyzed and processed to enable rapid response.

[1898] The server automatically generates a medical record based on the analysis results and sends it to the vehicle's display or the user's device. This allows the user to quickly receive remote medical treatment from a recommended specialist. After the diagnosis, the user sends their feedback information back to the server, which is used as training data for the generative model, contributing to improving the model's accuracy.

[1899] As a concrete example, suppose a passenger inputs "I have a stomachache and I'm anxious." This information is sent to the server, and the generative model interprets it as "stomachache" and "anxiety." At the same time, sensors detect an increase in heart rate, and cameras recognize changes in the passenger's complexion. Based on this data, the server recommends a gastroenterologist and provides information to the passenger. The passenger undergoes a remote consultation and then provides feedback, which the server incorporates to update the generative model.

[1900] An example of a prompt might be:

[1901] "You are developing a system that analyzes passengers' health conditions based on data collected from sensors and cameras and recommends appropriate specialists. According to the collected data, Passenger A is experiencing a 'headache' and 'feeling unwell.' Which doctor would you recommend?"

[1902] In this way, the system enables real-time health monitoring and rapid response, providing an effective means of protecting passenger safety and health.

[1903] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1904] Step 1:

[1905] User enters health information

[1906] Using a user terminal, the user inputs vague information about their health (e.g., "I'm worried because my stomach hurts") into the input form and clicks the submit button. The input information is sent from the user terminal to the server.

[1907] Input: Vague health information (natural language)

[1908] Output: Health information sent to the server

[1909] Step 2:

[1910] The server receives the health information.

[1911] The server receives vague health information sent from the user's device, temporarily stores it in a database, and prepares it for the next analysis step.

[1912] Input: Health information sent from the user's device

[1913] Output: Saved health information

[1914] Step 3:

[1915] Analyzing health information with generative models

[1916] The server passes the stored health information to the generative model, which then begins analysis. The generative model uses machine learning algorithms to convert vague information into specific symptoms. For example, the model outputs information like "I have a stomachache and I'm anxious" as specific symptoms like "stomachache" and "anxiety."

[1917] Input: Saved Health Information

[1918] Output: Specific symptoms

[1919] Step 4:

[1920] Analyzing emotional information with an emotion engine

[1921] The server passes the emotional expressions contained in the stored health information to the emotion engine for emotion analysis. The emotion engine uses natural language processing technology to identify the user's emotional state (e.g., anxiety, worry).

[1922] Input: Emotional expressions contained in stored health information

[1923] Output: User's emotional state

[1924] Step 5:

[1925] Recommend an expert

[1926] The server recommends an appropriate expert based on the analysis results of the generative model and the emotion engine. It searches the database for relevant experts, identifies the expert best suited to the user (e.g., a gastroenterologist), and sends the information to the user's terminal.

[1927] Input: specific symptoms, user's emotional state

[1928] Output: Recommended expert information

[1929] Step 6:

[1930] Collecting health data in real time

[1931] Sensors and cameras installed in the vehicle collect passenger health data (e.g., heart rate, body temperature) and emotional data (e.g., facial color, facial expressions) in real time, and the collected data is sent to a server.

[1932] Input: Health and emotion data collected by sensors and cameras

[1933] Output: Real-time data sent to the server

[1934] Step 7:

[1935] Analyzing real-time data

[1936] The server analyzes the health and emotional data sent in real time and performs analytical processing to quickly respond if an abnormality is detected, such as a sudden increase in heart rate or a change in facial color.

[1937] Input: Health and emotional data transmitted in real time

[1938] Output: Anomaly detection results

[1939] Step 8:

[1940] Response when an abnormality is detected

[1941] If the server detects an abnormality, it will immediately recommend an appropriate specialist and send that information to the user's device and the vehicle's display, allowing the user to receive prompt and appropriate medical treatment.

[1942] Input: Anomaly detection result

[1943] Output: Pertinent expert information

[1944] Step 9:

[1945] Gathering feedback and learning

[1946] After the diagnosis, the user sends feedback information about changes in their symptoms to the server, which uses this information as training data for the generative model to improve its accuracy.

[1947] Input: Feedback information

[1948] Output: Updated generative model

[1949] Step 10:

[1950] Automatic recording generation and transmission

[1951] The server automatically generates a medical record based on the analysis results and collected information and sends it to the vehicle's display or user terminal, allowing the user to check the medical record and take appropriate medical action.

[1952] Input: Analysis results and collected information

[1953] Output: Generated medical record

[1954] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1955] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1956] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1957] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1958] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1959] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1960] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1961] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1962] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1963] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1964] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1965] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1966] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1967] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1968] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1969] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1970] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1971] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1972] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1973] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1974] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1975] The following is further disclosed regarding the above embodiment.

[1976] (Claim 1)

[1977] a means for receiving user-entered vague health information;

[1978] an analysis means using a generative model to analyze the received ambiguous health information;

[1979] A means for identifying and recommending an appropriate expert based on the analysis results obtained by the analysis means;

[1980] A means for collecting post-diagnosis information from patients and using the collected information as feedback for training a generative model;

[1981] means for automatically generating a record based on the analysis results and collected information;

[1982] A system including:

[1983] (Claim 2)

[1984] 10. The system of claim 1, wherein the generative model uses a machine learning algorithm.

[1985] (Claim 3)

[1986] 10. The system of claim 1, wherein the user-entered ambiguous health information includes at least one natural language input.

[1987] (Claim 4)

[1988] 10. The system of claim 1, further comprising a user terminal that displays information about the recommended experts to the user.

[1989] (Claim 5)

[1990] 10. The system of claim 1, wherein the analysis means includes means for evaluating the accuracy of the diagnostic result.

[1991] "Example 1"

[1992] (Claim 1)

[1993] a means for receiving user-entered vague health information;

[1994] an analysis means using a generative model to analyze the received ambiguous health information;

[1995] A means for identifying and recommending an appropriate expert based on the analysis results obtained by the analysis means;

[1996] A means for collecting post-diagnosis information from patients and using the collected information as feedback for training a generative model;

[1997] means for automatically generating a record based on the analysis results and collected information;

[1998] means for transmitting information identifying and recommending an appropriate expert to a user terminal;

[1999] The analysis means uses natural language processing technology to convert ambiguous expressions into specific diagnostic candidates;

[2000] A system including:

[2001] (Claim 2)

[2002] 10. The system of claim 1, wherein the generative model uses a machine learning algorithm.

[2003] (Claim 3)

[2004] 10. The system of claim 1, wherein the user-entered ambiguous health information includes at least one natural language input.

[2005] "Application Example 1"

[2006] (Claim 1)

[2007] a means for receiving user-entered vague health information;

[2008] an analysis means using a generative model to analyze the received ambiguous health information;

[2009] A means for identifying and recommending an appropriate expert or product based on the analysis results obtained by the analysis means;

[2010] A means for collecting information after the customer has used the recommended expert or product, and using the collected information as feedback for training the generative model;

[2011] means for automatically generating a record based on the analysis results and collected information;

[2012] A system including:

[2013] (Claim 2)

[2014] 10. The system of claim 1, wherein the generative model uses a machine learning algorithm.

[2015] (Claim 3)

[2016] 10. The system of claim 1, wherein the user-entered ambiguous health information includes at least one natural language input.

[2017] "Example 2: Combining Emotion Engines"

[2018] (Claim 1)

[2019] means for receiving user-entered ambiguous health information and emotional expressions;

[2020] an analysis means using a generative model and an emotion analysis engine to analyze the received ambiguous health-related information and emotion expressions;

[2021] a means for identifying and recommending an appropriate expert based on the symptom analysis result and emotion analysis result obtained by the analysis means;

[2022] A means for collecting feedback information from patients after diagnosis and using the collected ...

Claims

1. a means for receiving user-entered vague health information; an analysis means using a generative model to analyze the received ambiguous health information; A means for identifying and recommending an appropriate expert based on the analysis results obtained by the analysis means; A means for collecting post-diagnosis information from patients and using the collected information as feedback for training a generative model; means for automatically generating a record based on the analysis results and collected information; A system including:

2. The system of claim 1 , wherein the generative model uses a machine learning algorithm.

3. 10. The system of claim 1, wherein the user-entered ambiguous health information includes at least one natural language input.

4. 2. The system of claim 1, further comprising a user terminal that displays information about the recommended experts to the user.

5. 10. The system of claim 1, wherein the analysis means includes means for evaluating the accuracy of the diagnostic results.

Images