System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving an inquiry about medical care from a user; means for analyzing the received inquiry content by natural language processing; means for transmitting the analyzed inquiry content to a generation system artificial intelligence; means for transmitting an answer generated by the generation system artificial intelligence to an expert and receiving supervision; and means for transmitting an answer for which supervision is completed to the user.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] Patients and their families with medical questions or concerns often have difficulty obtaining expert advice quickly. While there are many situations where patients need appropriate advice, especially before and after consultations and during treatment, the time required for making appointments and the limited availability of medical experts can delay the resolution of such requests. There is also the risk of making incorrect self-diagnosis. This can easily lead to anxiety for patients and their families, and can lead to a decline in the quality and efficiency of medical care. This invention aims to solve these problems by providing a system that allows users to obtain prompt and appropriate medical advice, thereby reducing the anxiety of patients and their families and improving the quality and efficiency of medical care. [Means for solving the problem]

[0005] The present invention provides a system including a means for receiving a medical inquiry from a user, a means for analyzing the received inquiry using natural language processing, a means for transmitting the analyzed inquiry to a generative artificial intelligence, a means for transmitting an answer generated by the generative artificial intelligence to an expert for review, and a means for transmitting the reviewed answer to the user. This allows users to receive medical consultations 24 hours a day and quickly obtain reliable information through expert review. The system also includes a means for formatting the expert-reviewed answer for provision to the user, thereby improving the readability and understandability of the information. Furthermore, the system also includes training data for the generative artificial intelligence to generate appropriate answers to specific medical-related questions, thereby reducing the burden on experts while maintaining the accuracy of the answers.

[0006] "User" refers to an individual or family member who utilizes the system to make a medical inquiry.

[0007] "Medical inquiries" refer to all medical-related questions that users have, such as doubts and concerns about their health condition or symptoms, questions about treatment methods, and consultations about how to use medical institutions.

[0008] "Means for receiving" refers to the function of acquiring inquiries from users in digital form and storing them in a form that can be processed within the system.

[0009] "Natural language processing" is a technology for analyzing user-submitted inquiries and understanding their intent and content, and includes techniques such as automatic tokenization, POS tagging, and named entity recognition.

[0010] "Means for analysis" refers to the function of analyzing the content of the received inquiry using natural language processing technology and extracting its intent and important points.

[0011] "Generative artificial intelligence" refers to machine learning models and deep learning models that automatically generate appropriate answers to user inquiries.

[0012] "Means for generating" refers to the function that the generative artificial intelligence uses to create an appropriate answer based on a user's inquiry.

[0013] An "expert" refers to a medical professional with medical knowledge and qualifications who will oversee the content of the answers generated by generative AI and ensure that the information is accurate and reliable.

[0014] "Means of sending and receiving review" refers to the function that allows answers created by generative AI to be sent to experts, who then review and correct the answers.

[0015] "Means for sending edited answers to users" refers to the function of sending final answers that have been checked and corrected by experts to users.

[0016] "Formatting means" refers to the function of arranging the final answer, which has been supervised by experts, into a format that is easy to view and understand so that it can be provided to the user.

[0017] "Training data" refers to a dataset containing medical knowledge and information that generative artificial intelligence has learned in advance to generate appropriate answers. [Brief explanation of the drawings]

[0018] [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

[0019] 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.

[0020] First, the terms used in the following description will be explained.

[0021] 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).

[0022] 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.

[0023] 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.

[0024] 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.

[0025] 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."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 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.

[0029] 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).

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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."

[0039] The present invention is a system that enables users to quickly resolve medical-related questions and concerns, and is implemented in the following manner.

[0040] First, users can send medical questions in the form of direct messages through a dedicated app or website, and the user's device (smartphone or PC) then sends the question to the server.

[0041] Next, the server receives an inquiry message from the user. The received message is stored in a database and then analyzed by a natural language processing (NLP) engine. During the analysis, the intent of the inquiry and important keywords are extracted. For example, if the question is, "I get short of breath at night. What should I do?", the keywords "night," "short of breath," and "measures" are analyzed.

[0042] The analyzed query content is sent to a generative AI (such as GPT-4), which generates an appropriate answer based on the data sent. The answers generated here are based on medical training data and are designed to provide accurate and reliable information. For example, a generated answer might read, "Possible causes of shortness of breath include allergies, asthma, and heart problems. We recommend that you first consult a doctor. We also recommend that you try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow."

[0043] Instead of providing the answer directly to the user, the generated answer is first sent to an expert, who reviews the content of the generated answer and reviews it for accuracy and appropriateness. After this review process is complete, the server formats the answer in an easy-to-understand format before sending it to the user. The final answer is then sent to the user's device as a direct message.

[0044] Users can receive the final answer on their device and obtain accurate medical advice for their questions. This allows users to consult with a doctor at any time and quickly obtain reliable information supervised by an expert.

[0045] For example, if a user sends a specific question such as, "Recently, I've been having trouble breathing before going to bed. Is there anything I can do about this?", the server receives it and analyzes it using a natural language processing engine. A generative AI system then generates an appropriate answer, which is then reviewed by a specialist and finally sent to the user. This system allows users to receive accurate and prompt medical advice.

[0046] This system contains training data to generate appropriate answers to specific medical questions, reducing the burden on medical experts while maintaining the accuracy of the answers. It is also available 24 hours a day, allowing users to receive medical consultations at their convenience. This is expected to reduce anxiety for patients and their families and improve the quality and efficiency of medical care.

[0047] The processing flow will be explained below.

[0048] Step 1:

[0049] Users enter medical questions from their device (smartphone or PC) through a dedicated app or website and send them as direct messages.

[0050] Step 2:

[0051] The server receives the query message sent by the user and stores it in a database.

[0052] Step 3:

[0053] The server sends the received message to a natural language processing (NLP) engine, which analyzes the message content, performing tokenization (dividing sentences into words), POS tagging (identifying parts of speech), and named entity recognition (recognizing specific names, places, diseases, etc.).

[0054] Step 4:

[0055] The server normalizes the query content analyzed by the NLP engine and sends it to the generative AI (e.g., GPT-4). The normalization process converts the query content into a format that is easy for the generative AI to understand.

[0056] Step 5:

[0057] Generative AI generates appropriate answers based on the data received, and these answers are internally verified to ensure they are accurate and reliable.

[0058] Step 6:

[0059] The server sends the generated answer to an expert, who reviews the answer and supervises its accuracy and appropriateness. Corrections and supplements are made as necessary.

[0060] Step 7:

[0061] The final answer, edited by experts, is sent to the server, which formats it in a user-friendly format.

[0062] Step 8:

[0063] The server sends the final formatted response to the user as a direct message.

[0064] Step 9:

[0065] The user receives the final answer on the device and obtains the appropriate medical advice for their question. The user checks the information displayed on the device and decides on the next step if necessary.

[0066] Example 1

[0067] 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."

[0068] Conventional medical consultation systems have limitations in terms of providing users with fast and accurate answers. Medical experts often have limited time and resources to respond directly, resulting in variations in the quality and speed of responses. Furthermore, if users want to seek medical advice at their own convenience, a system that can provide fast responses 24 hours a day is required, but such systems are not widely available at present. The objective of this invention is to provide a system that allows users to receive fast and reliable medical advice at any time.

[0069] 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.

[0070] In this invention, the server includes means for receiving a medical inquiry from a user, means for analyzing the received inquiry content using natural language processing, means for transmitting the analyzed inquiry content to a generative AI, means for transmitting an answer generated by the generative AI to an expert for review, means for transmitting the reviewed answer to the user, means for transmitting a formatted final answer to the user, and means for the user to receive the final answer and obtain appropriate medical advice. This enables users to receive prompt and reliable medical advice 24 hours a day.

[0071] The "means for receiving medical inquiries from users" refers to the means by which the server receives medical questions or doubts sent by users through the dedicated application or website.

[0072] "Means for analyzing the content of received inquiries using natural language processing" refers to a means for analyzing medical inquiries received by the server using a natural language processing engine and extracting important keywords and intent.

[0073] The "means for transmitting the analyzed query content to the generative artificial intelligence" is a means for transmitting the query content analyzed by natural language processing to the generative artificial intelligence as an appropriate prompt.

[0074] "Means for sending answers generated by generative AI to experts for review" refers to means for sending answers generated by generative AI to medical experts, who then review and confirm the content.

[0075] The "means for transmitting the answer that has been supervised to the user" is a means for transmitting the answer that has been supervised by the expert to the user's terminal.

[0076] The "means for transmitting a formatted final answer to a user" refers to a means for transmitting a final answer to a user that has been formatted in a format that is easy to understand after being supervised by an expert.

[0077] The "means for the user to receive the final answer and obtain appropriate medical advice" refers to the means by which the user's terminal receives the final answer sent from the server and the user can obtain that medical advice.

[0078] The present invention is a system that allows users to quickly and accurately resolve medical questions and concerns. It is expected that the system will be implemented in accordance with the following detailed description.

[0079] First, users can use a dedicated application or website to send medical questions in the form of direct messages, with the user's device (such as a smartphone or personal computer) sending the question to a server via the Internet.

[0080] Specifically, the user enters a question into a text input form on a dedicated app or website and presses the send button. For example, suppose the question entered is, "I feel short of breath before going to bed. What should I do?" The device then sends this question to the server.

[0081] Next, the server receives the inquiry message sent by the user. The received message is saved in a database. This database can be built using MySQL or PostgreSQL, for example. After saving, the server analyzes the message using a natural language processing engine (for example, Google Natural Language API). As a result of the analysis, the intent of the question and important keywords are extracted. For example, the keywords extracted are "before going to bed," "shortness of breath," and "measures."

[0082] The parsed query content is sent from the server to a generative AI (e.g., OpenAI's GPT-4), and the prompt text is also sent to the generative AI.

[0083] Example prompt sentence:

[0084] "Healthcare question: 'I'm having trouble breathing before bed. What should I do?' Generate an answer."

[0085] Generative AI generates appropriate answers based on the submitted data and prompts. The generated answers are based on medical training data and are designed to provide accurate and reliable information. For example, a generated answer might read, "Possible causes of shortness of breath include allergies, asthma, or heart problems. We recommend consulting a doctor first. Also, try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow."

[0086] The generated answer is sent to an expert for review. The expert reviews the generated answer and checks its accuracy and appropriateness. After the review process is complete, the server formats the final answer into an easy-to-understand format before sending it to the user. For example, it may use HTML formatting or highlight important keywords. The final formatted answer is then sent to the user's device as a direct message.

[0087] Finally, the user receives the final answer on their device and obtains appropriate medical advice for their question. This allows users to receive prompt and accurate medical consultations 24 hours a day. The system uses training data to generate appropriate answers to specific medical questions, reducing the burden on experts while maintaining the accuracy of the answers. This is expected to improve the quality and efficiency of medical care.

[0088] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0089] Step 1:

[0090] The user uses a dedicated application or website to input and submit a medical question. The input data here is the user's medical question. The device sends this input data to the server. Specifically, the user inputs a question such as, "I have shortness of breath before going to bed. What should I do?" and clicks the submit button.

[0091] Step 2:

[0092] The server receives the message sent by the user. The received data is the user's question, and to store it in the database, it executes an SQL query, for example, "INSERT INTO inquiries (user_id, question, timestamp) VALUES (?, ?, ?)". The input is the message data from the user, and the output is the record stored in the database.

[0093] Step 3:

[0094] The server sends the stored message to a natural language processing engine for analysis. The input data here is the user's question retrieved from the database. The natural language processing engine (for example, Google Natural Language API) analyzes the inquiry and extracts important keywords and intent. Specifically, it extracts keywords such as "before going to bed," "shortness of breath," and "measures." The output is the analysis results.

[0095] Step 4:

[0096] The server sends the analyzed query content to the generative AI. The input data consists of keywords from the analyzed query content and the corresponding prompt. For example, a prompt such as "This is a medical question. Please generate an answer to the question, 'I feel short of breath before going to bed. What should I do?'" is sent to the generative AI. The output is the generated answer.

[0097] Step 5:

[0098] The server receives the answer generated by the generative AI and then sends it to the expert. The input data is the answer received from the generative AI. The server sends this answer to the expert, who reviews it. The output is feedback from the expert and a revised answer.

[0099] Step 6:

[0100] The server receives the expert-edited answer and formats it. The input data is the expert-edited answer, which is then formatted into an easy-to-understand format, for example, using HTML to highlight important keywords. The output is the formatted answer.

[0101] Step 7:

[0102] The server sends a formatted final answer to the user. The input data is the formatted final answer, and the output is a message sent to the user's terminal. The user receives this final answer on their terminal and obtains the required medical advice.

[0103] In this way, each processing step works in tandem, allowing users to receive prompt and accurate medical advice 24 hours a day.

[0104] (Application example 1)

[0105] 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."

[0106] In modern society, users need to be able to quickly and accurately resolve their medical questions and concerns. However, current systems require users to directly contact medical experts, which often requires time and effort. In addition, general online information can be unreliable, and there is a lack of mechanisms for providing users with accurate medical advice. Furthermore, the lack of an efficient system that utilizes mobile devices such as smartphones makes it difficult to meet the needs of today's busy people.

[0107] 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.

[0108] In this invention, the server includes means for receiving medical inquiries from users, means for analyzing the received inquiry content using natural language processing, means for sending the analyzed inquiry content to a generative artificial intelligence, means for sending an answer generated by the generative artificial intelligence to an expert for supervision, means for sending the supervised answer to the user, and means for receiving inquiries from users using a smartphone application. This enables users to receive fast and reliable medical advice via their smartphones, and the accuracy of the answers is guaranteed through expert supervision, allowing users to alleviate their concerns and take appropriate measures quickly.

[0109] A "user" is an individual who has a medical question or concern.

[0110] An "inquiry" is a question sent by a user to resolve a medical question or doubt.

[0111] The "receiving means" is a system or device for electronically receiving an inquiry sent by a user.

[0112] "Natural language processing" is a technology that allows computers to understand and analyze human language.

[0113] "Generative artificial intelligence" is an AI system that automatically generates appropriate medical advice based on the content of the inquiry it receives.

[0114] An "expert" is an individual or team with a high level of expertise and experience in the medical field.

[0115] "Supervision" refers to the methods and processes by which experts verify the accuracy and appropriateness of the answers provided by the generative AI.

[0116] A "smartphone application" is a program that runs on a smartphone and allows users to send medical inquiries.

[0117] "Formatting methods" are techniques or processes used to prepare curated responses in a format that is easy for users to understand.

[0118] "Server" refers to a computer system that manages and executes a series of processes, such as receiving inquiries, analyzing them, generating responses, sending them to experts, and reviewing them.

[0119] "Learning data" refers to training data that is used in advance by generative artificial intelligence to generate appropriate medical advice.

[0120] MODE FOR CARRYING OUT THE INVENTION

[0121] The present invention is a system that allows users to quickly and accurately resolve medical questions and concerns. Each part of the system operates according to the following procedure.

[0122] 1. Program Generation

[0123] The program of the system for realizing the present invention mainly consists of the following components.

[0124] Receiving means: An interface (such as a smartphone application) for receiving medical inquiries from users.

[0125] Natural Language Processing: An NLP engine (e.g., SomeNLPModel) for analyzing query content and extracting important keywords.

[0126] Generative artificial intelligence: An AI system that generates medical advice based on control messages (e.g., GPT-4).

[0127] Editing tool: An interface for experts to review, correct, and approve answers generated by the AI.

[0128] Formatting: The process of sending expert-approved answers in a format that is easy for users to understand.

[0129] 2. Explain the program's processing

[0130] The program's processing begins when a query is sent from the user's smartphone application to the server. The server receives this query and analyzes its contents using a natural language processing (NLP) engine. During the analysis, important keywords are extracted. An NLP engine called SomeNLPModel is used in this step.

[0131] The extracted keywords are then sent to a generative artificial intelligence (GPT-4) to generate appropriate medical advice. The generated answers are then sent to experts, who review them for accuracy and appropriateness. This process is carried out using an expert review interface.

[0132] Once the expert-edited answers are ready for the user, they are formatted. This step ensures that the answers are easy for the user to understand. The final answers are then sent to the user via a smartphone application.

[0133] 3. Add specific examples

[0134] For example, suppose a user sends a question via a smartphone application: "Recently, I've been having trouble breathing before going to bed. Is there anything I can do about this?" In this case, the server receives the question and analyzes it using an NLP engine (SomeNLPModel). This analysis extracts keywords such as "night," "shortness of breath," and "measures." A generative artificial intelligence (GPT-4) generates appropriate medical advice based on these keywords. The generated answer is sent to an expert for review. After review, the answer is formatted before being provided to the user and sent to the user in the following format:

[0135] Example prompt sentence:

[0136] Q: Lately I've been having trouble breathing before going to bed. Is there anything I can do about it?

[0137] Keywords: night, shortness of breath, measures

[0138] Generate medical advice based on this keyword.

[0139] In this way, users can receive fast and accurate medical advice.

[0140] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0141] Step 1:

[0142] A user submits a medical inquiry through a smartphone application. The input is the user's question, and the output is the action of sending this question to the server. For example, if a user enters the question, "Recently, I've been having trouble breathing before going to bed. Is there anything I can do about it?", the question is sent to the server.

[0143] Step 2:

[0144] The server receives inquiries from users. The input is the user's question, and the output is the data of the received inquiry. The server receives and stores the data sent from the smartphone app. Specifically, the inquiry is saved in a database.

[0145] Step 3:

[0146] The query received by the server is analyzed using a natural language processing (NLP) engine. The input is the query data, and the output is the extracted keywords. For example, keywords such as "night," "shortness of breath," and "measures" are extracted. Specifically, SomeNLPModel, an NLP engine, analyzes the query and extracts important keywords.

[0147] Step 4:

[0148] The server sends the keywords from the analyzed query to a generative AI (GPT-4) system, which generates appropriate medical advice. The input is the extracted keywords, and the output is the generated medical advice. For example, the generated advice might read, "Possible causes of shortness of breath include allergies, asthma, and heart problems. We recommend that you first consult a doctor. Also, try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow." Specifically, the generative AI system outputs appropriate advice based on the prompt text.

[0149] Step 5:

[0150] The generated answer is sent to an expert for review. The input is the generated medical advice, and the output is an answer that has been reviewed and approved by an expert. Specifically, the answer output by the generative AI is checked by an expert review system, and any corrections that need to be made are made, and finally, it is approved.

[0151] Step 6:

[0152] The server formats the expert-edited answers for delivery to the user. The input is the edited answers, and the output is the final formatted answers. Specifically, the answers are formatted in a user-friendly format and delivered via a smartphone app.

[0153] Step 7:

[0154] The server sends the final answer to the user. The input is the formatted final answer, and the output is the answer that is displayed on the user's smartphone. Specifically, the server sends the formatted answer to a smartphone app, and the user views the answer.

[0155] This allows users to receive prompt and accurate medical advice.

[0156] 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.

[0157] This invention is a system that allows users to quickly resolve medical-related questions and concerns, and by combining it with an emotion engine that recognizes the user's emotions, it provides more appropriate responses. It is implemented in the following steps.

[0158] First, a user sends a medical question via a dedicated app or website in the form of a direct message. The user's device (smartphone or PC) then sends the question to the emotion engine, which analyzes the user's emotions. The emotion engine then uses the text data to analyze the user's emotional state (e.g., anxiety, anger, relief, etc.).

[0159] Next, after analyzing the user's emotions, the server receives the question message sent by the user. The received message is stored in a database, and then the inquiry is analyzed by a natural language processing (NLP) engine. During the analysis, the intent of the inquiry and important keywords are extracted. For example, if the question is "I get short of breath at night. What should I do?", the keywords "night," "short of breath," and "measures" are analyzed.

[0160] Along with the analyzed query content, the user's emotional information analyzed by the emotion engine is also sent to a generative AI (such as GPT-4). The generative AI generates an appropriate answer taking into account the user's emotions. For example, if the emotion is recognized as "anxiety," the answer will include a reassuring expression. In this way, the generated answer is based on medical training data and provides accurate and reliable information.

[0161] Instead of providing the answer directly to the user, the generated answer is first sent to an expert, who reviews the content of the generated answer and reviews it for accuracy and appropriateness. After this review process is complete, the server formats the answer in an easy-to-understand format before sending it to the user. The final answer is then sent to the user's device as a direct message.

[0162] Users can receive the final answer on their device and obtain accurate medical advice for their questions. This allows users to receive medical consultations 24 hours a day and quickly obtain reliable information supervised by experts.

[0163] For example, if a user sends a specific query such as, "I've been having trouble breathing before bed recently. I'm feeling very anxious. What should I do?", the emotion engine will interpret it as "anxiety." This information and the query are received and analyzed by the server and sent to the generative AI. The generative AI then generates a response such as, "Possible causes of your shortness of breath include allergies, asthma, and heart problems. We recommend that you first consult a doctor. We also recommend that you try improving your bedroom environment. For example, use an air purifier to remove dust and allergens and use a high pillow. We understand your anxiety, but please try these measures." A specialist will then review this response and confirm its appropriateness before sending it to the user. In this way, the user can receive an emotionally sensitive response, which can help alleviate their anxiety.

[0164] This system uses an emotion engine to recognize the user's emotions and provide that information to a generative AI system, which can then generate appropriate answers based on the user's emotions. Furthermore, after being supervised by experts, the system provides users with highly reliable information, greatly improving the quality and efficiency of medical consultations.

[0165] The processing flow will be explained below.

[0166] Step 1:

[0167] Users enter medical questions from their device (smartphone or PC) through a dedicated app or website and send them as direct messages.

[0168] Step 2:

[0169] The server receives the query message sent by the user and stores it in a database.

[0170] Step 3:

[0171] The server sends the message to the emotion engine, which analyzes the user's emotion. The emotion engine analyzes the text data to recognize the user's emotional state (e.g., anxiety, anger, relief, etc.) and stores this information for further processing.

[0172] Step 4:

[0173] The server sends the received message to a natural language processing (NLP) engine, which analyzes the message content, performing tokenization (dividing sentences into words), POS tagging (identifying parts of speech), and named entity recognition (recognizing specific names, places, diseases, etc.).

[0174] Step 5:

[0175] The server normalizes the query content analyzed by the NLP engine and the user's emotional information recognized by the emotion engine, and sends them to a generative AI (e.g., GPT-4). The normalization process converts the data into a format that is easy for the generative AI to understand.

[0176] Step 6:

[0177] Generative AI generates appropriate answers based on the normalized data. The system also takes into account the user's emotional information, so the generated answers include language tones and expressions that correspond to the emotion. For example, if the user's emotion is recognized as "anxiety," the system adds expressions that convey a sense of security to the answer.

[0178] Step 7:

[0179] The server sends the generated answer to the expert, who reviews the answer and supervises its accuracy and appropriateness, correcting or supplementing it as necessary.

[0180] Step 8:

[0181] The final answer, edited by an expert, is sent to the server, which formats it for presentation to the user. The formatted answer is formatted in a way that is easy for the user to understand.

[0182] Step 9:

[0183] The server sends the final formatted response to the user as a direct message.

[0184] Step 10:

[0185] The user receives the final answer on the device and obtains the medical advice relevant to their question. The user reviews the information displayed on the device and decides on the next step if necessary, such as making further inquiries or acting on the advice.

[0186] Example 2

[0187] 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."

[0188] In modern medical consultations, it is necessary to quickly and appropriately resolve users' anxieties and doubts. However, conventional systems generate mechanical answers without considering the user's feelings, which can lead to low user satisfaction. Furthermore, to ensure the reliability of the generated answers, expert supervision is essential, but there are only a limited number of systems that can do this efficiently.

[0189] 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.

[0190] In this invention, the server includes means for receiving a medical inquiry from a user, means for analyzing the received inquiry content using natural language processing, means for analyzing the user's emotions, means for transmitting the analyzed inquiry content and emotional information to the generative AI, means for transmitting the answer generated by the generative AI to an expert for review, and means for transmitting the answer after review to the user. This makes it possible to generate an appropriate medical answer that takes the user's emotions into consideration, and provide it to the user with a high level of reliability after it has been checked by the expert.

[0191] A "user" is someone who uses the system to resolve medical questions or concerns.

[0192] "Inquiry" refers to a medical question or consultation sent by a user.

[0193] "Natural language processing" is a technical method for analyzing the content of received inquiries and extracting intent and keywords.

[0194] "Sentiment analysis" is the process of identifying an emotional state from the text data contained in a user's query.

[0195] "Generative AI" refers to AI that generates appropriate medical answers based on the content of the inquiry and emotional information it receives.

[0196] An "expert" is someone who is qualified to oversee the content of medical answers created by generative AI and verify their accuracy and appropriateness.

[0197] "Supervision" is the act of an expert reviewing the content of answers generated by generative AI and ensuring their appropriateness and accuracy.

[0198] "Formatting" is the process of shaping the answers provided to the user into an understandable form.

[0199] "Training data" refers to the training data used by generative artificial intelligence to generate appropriate answers to medical-related questions.

[0200] The present invention is a system that enables users to receive prompt and appropriate advice regarding medical questions and concerns. By incorporating an emotion engine that analyzes the user's emotions, the system provides appropriate responses that take emotions into consideration.

[0201] First, a user sends a medical question in the form of a direct message through a dedicated application or website. For example, a user might use their smartphone to send a message along the lines of, "Lately, I've been having trouble breathing before going to bed. It's really worrying. What should I do?"

[0202] The user's device sends the message to an emotion engine. A commonly used emotion analysis tool can be used as the emotion engine. As a specific example, IBM Watson's Tone Analyzer can be used. This emotion engine analyzes the user's emotional state (anxiety, anger, relief, etc.) based on the text data and assigns an emotion label. In this specific example, the emotion label "anxiety" is obtained from the part "I am very anxious."

[0203] Next, the server receives the message sent by the user and stores it in a MySQL database. The stored data includes the user ID, message content, emotion label, etc. The stored message is then analyzed by a natural language processing (NLP) engine. NLP engines such as SpaCy and NLTK can be used as examples. During this analysis process, the intent of the inquiry and important keywords are extracted. For example, from the message "Recently, I've been having trouble breathing before going to bed," keywords such as "before going to bed," "trouble breathing," and "measures" are extracted.

[0204] The analyzed inquiry content and emotional information are sent to a generative AI. A large language model such as GPT-4 is used as the generative AI. At this time, the following message is input as a prompt: "The user has recently been experiencing difficulty breathing before going to bed, which has caused them great anxiety. Please provide appropriate medical advice."

[0205] The generative AI model generates an appropriate answer based on the prompt and emotional information. For example, it might say, "Possible causes of shortness of breath include allergies, asthma, or heart problems. We recommend that you first consult a doctor. Also, try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow. We understand that you may be feeling anxious, but please try these measures."

[0206] The generated answers are sent to experts, who review the answers for accuracy and appropriateness and make any necessary corrections. Once the review is complete, the server formats the answers in a way that is easy for users to understand, such as by organizing them into paragraphs or bullet points.

[0207] The final answer will be sent to the user's device as a direct message, allowing them to receive the final answer on their smartphone or PC and receive reliable medical advice for their question.

[0208] In this way, the present invention makes it possible to quickly provide appropriate medical answers that take into account the user's feelings.

[0209] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0210] Step 1:

[0211] A user submits a medical question through a dedicated application or website.

[0212] Input: User's question (e.g., "Recently, I've been having trouble breathing before going to bed. It's really worrying me. What should I do?")

[0213] Output: A query message is sent from the user's terminal to the server.

[0214] Step 2:

[0215] The user's terminal transmits the received question message to the emotion engine.

[0216] Input: User's question message

[0217] Data processing: The emotion engine analyzes the text data and assigns an emotion label (e.g., "anxiety").

[0218] Output: Question message with emotion label (e.g., "Anxious" from "I'm very anxious")

[0219] Step 3:

[0220] The server receives the query message sent from the user's terminal and stores it in a database.

[0221] Input: A question message with an emotion label.

[0222] Data processing: Received messages are stored in a MySQL database. The stored data includes user ID, message content, emotion label, etc.

[0223] Output: The question message stored in the database

[0224] Step 4:

[0225] The server sends the query message stored in the database to a natural language processing (NLP) engine, which analyzes the query.

[0226] Input: Question message stored in the database

[0227] Data calculation: Text analysis is performed by an NLP engine (e.g., SpaCy) to extract intent and important keywords (e.g., "before going to bed," "difficulty breathing," "measures").

[0228] Output: Parsed query content

[0229] Step 5:

[0230] The server sends the analyzed inquiry content and emotional information to the generative artificial intelligence.

[0231] Input: Parsed query content and sentiment label

[0232] Data processing: Generate prompt sentences (e.g., "The user has recently been experiencing difficulty breathing before going to bed, which has caused him great anxiety. Please provide appropriate medical advice.")

[0233] Output: The prompt sent to the generative AI model

[0234] Step 6:

[0235] Generative AI generates appropriate answers based on prompts and emotional information.

[0236] Input: A prompt sent to the generative AI model

[0237] Data computation: A generative AI model (e.g., GPT-4) generates an appropriate answer based on a prompt (e.g., "Possible causes of shortness of breath include allergies, asthma, heart problems, etc...").

[0238] Output: The generated answer

[0239] Step 7:

[0240] The server sends the generated answers to experts, who then review the content.

[0241] Input: Generated Answer

[0242] Data processing: Experts check the accuracy and appropriateness of the answers and make corrections if necessary (expert feedback)

[0243] Output: Edited answer

[0244] Step 8:

[0245] The server formats the edited answers and converts them into a user-friendly format.

[0246] Input: Edited answer

[0247] Data processing: Formatting responses, such as organizing them into paragraphs or bullet points

[0248] Output: Formatted final answer

[0249] Step 9:

[0250] The server sends the final response to the user's device as a direct message.

[0251] Input: Final formatted answer

[0252] Output: Final answer EX sent to the user's device (received by the user on their smartphone or PC)

[0253] (Application example 2)

[0254] 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."

[0255] Conventional systems have difficulty providing responses that take users' emotions into consideration, and often do not provide adequate support, especially to users who feel anxious. Furthermore, the generated answers require time-consuming supervision by experts, making it difficult to provide immediate responses. This has led to issues that make it difficult for users to obtain fast and reliable information.

[0256] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the received inquiry content using natural language processing, means for sending the analyzed inquiry content to the generative artificial intelligence, means for sending an answer generated by the generative artificial intelligence to an expert for supervision, means for sending the answer after supervision to the user, emotion analysis means for recognizing the user's emotional state, and means for providing the generative artificial intelligence with analysis results including emotion-analyzed information. This makes it possible to quickly provide a response that takes the user's emotions into consideration, and to provide highly reliable information that has been supervised by an expert.

[0257] "User inquiries" refer to questions or doubts that users send to the system.

[0258] "Natural language processing" is a technology that allows computers to understand and analyze human language.

[0259] "Generative AI" refers to AI technology that automatically generates appropriate answers based on input data.

[0260] An "expert" is someone who has advanced knowledge and experience in a particular field.

[0261] "Supervision" is the process in which experts check the accuracy and appropriateness of the generated answers.

[0262] "Emotion analysis means" is a technology that analyzes a user's text data and identifies the user's emotional state.

[0263] "Formatting means" refers to a technique that provides expert-edited answers in a format that is easy for users to understand.

[0264] "Learning data" is training data that generative artificial intelligence uses to generate appropriate answers.

[0265] A system for implementing the present invention has the following configuration.

[0266] 1. The server first receives an inquiry from the user. The user enters security questions or concerns through a dedicated application or website. For example, a user could use a device such as a smartphone or PC to send an inquiry such as, "I've been feeling uneasy about online banking lately. What can I do to make it safer?"

[0267] 2. When a query arrives at the server, the server analyzes it using a natural language processing (NLP) engine to extract important keywords and intent from the query. Libraries such as TextBlob and SentimentAnalysis are used here.

[0268] 3. Next, the server uses emotion analysis to recognize the user's emotions. Tools such as EmotionEngine provide technology that analyzes the user's text data and identifies their emotional state (e.g., anxiety, relief, etc.).

[0269] 4. The results of the sentiment analysis and the analysis results from the NLP engine are sent to a generative AI model (such as GPT-4). The generative AI model uses this data to generate an appropriate response that takes the user's emotions into consideration.

[0270] 5. The generated answers are sent to experts, who review them for accuracy and appropriateness. For example, a security expert might review an answer about how to use online banking safely.

[0271] 6. After the expert has verified the answer, the server formats the final answer for delivery to the user. This formatted answer is provided in a format that is easy for the user to understand.

[0272] 7. Finally, the server sends the final formatted response to the user's terminal.

[0273] As a specific example, if a user sends a query such as, "I've been feeling uneasy about online banking lately. What can I do to make it safer?", the emotion engine will interpret it as "uneasy." This information and the content of the query are received and analyzed by the server, and then sent to the generative AI. The generative AI will then generate a response such as, "To increase the security of online banking, it is important to use a strong password and set up two-step authentication. Also, be careful not to click on suspicious links." This is then reviewed by security experts, and the final response is sent to the user.

[0274] Examples of prompts used by generative AI models include:

[0275] "Users are worried and have asked about how to use online banking safely. Please reassure them and provide helpful advice."

[0276] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0277] Step 1:

[0278] A user uses a device such as a smartphone or PC to send a security inquiry to a server through a dedicated application or website. The input is the user's inquiry text, and the output is the inquiry data sent to the server. Specifically, the text sent might be something like, "I've been feeling uneasy about online banking lately. What can I do to make it safer?"

[0279] Step 2:

[0280] The server analyzes the received inquiry using a natural language processing (NLP) engine. The input is the inquiry text, and the output is the analyzed keywords and intent. Specifically, TextBlob and SentimentAnalysis are used to extract important keywords such as "online banking," "anxiety," and "safety."

[0281] Step 3:

[0282] The server recognizes the user's emotional state using an emotion analysis means. The input is the user's query text, and the output is the recognized emotional information. Specifically, the server uses the Emotion Engine to analyze the user's emotion (e.g., anxiety) and obtains it as data.

[0283] Step 4:

[0284] The server sends the analyzed keywords and emotional information to a generative artificial intelligence (generative AI model). The input is the analyzed keywords and emotional information, and the output is a generated answer. Specifically, it uses the OpenAI API (e.g., GPT-4) to generate an answer such as, "To increase the security of your online banking, it is important to use a strong password and set up two-factor authentication. Also, be careful not to click on suspicious links."

[0285] Step 5:

[0286] The generated answer is sent from the server to an expert for review. The input is the generated answer text, and the output is the answer that has been checked and corrected by the expert. Specifically, the security expert checks the accuracy and appropriateness of the answer content and makes corrections as necessary.

[0287] Step 6:

[0288] The server formats the expert-edited answers for delivery to the user. The input is the edited answer text, and the output is the formatted answer. Specifically, the server converts the answer content into a format that is easy for the user to understand.

[0289] Step 7:

[0290] Finally, the server sends the formatted answer to the user's device. The input is the formatted answer text, and the output is the answer received on the user's device. Specifically, the intuitive and easy-to-read answer is displayed on the user's smartphone or PC.

[0291] Through the above processing steps, users can receive prompt and reliable security advice that takes their emotions into consideration.

[0292] 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.

[0293] 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.

[0294] 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.

[0295] [Second embodiment]

[0296] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0297] 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.

[0298] 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).

[0299] 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.

[0300] 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.

[0301] 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).

[0302] 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.

[0303] 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.

[0304] 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.

[0305] 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.

[0306] 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.

[0307] 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."

[0308] The present invention is a system that enables users to quickly resolve medical-related questions and concerns, and is implemented in the following manner.

[0309] First, users can send medical questions in the form of direct messages through a dedicated app or website, and the user's device (smartphone or PC) then sends the question to the server.

[0310] Next, the server receives an inquiry message from the user. The received message is stored in a database and then analyzed by a natural language processing (NLP) engine. During the analysis, the intent of the inquiry and important keywords are extracted. For example, if the question is, "I get short of breath at night. What should I do?", the keywords "night," "short of breath," and "measures" are analyzed.

[0311] The analyzed query content is sent to a generative AI (such as GPT-4), which generates an appropriate answer based on the data sent. The answers generated here are based on medical training data and are designed to provide accurate and reliable information. For example, a generated answer might read, "Possible causes of shortness of breath include allergies, asthma, and heart problems. We recommend that you first consult a doctor. We also recommend that you try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow."

[0312] Instead of providing the answer directly to the user, the generated answer is first sent to an expert, who reviews the content of the generated answer and reviews it for accuracy and appropriateness. After this review process is complete, the server formats the answer in an easy-to-understand format before sending it to the user. The final answer is then sent to the user's device as a direct message.

[0313] Users can receive the final answer on their device and obtain accurate medical advice for their questions. This allows users to consult with a doctor at any time and quickly obtain reliable information supervised by an expert.

[0314] For example, if a user sends a specific question such as, "Recently, I've been having trouble breathing before going to bed. Is there anything I can do about this?", the server receives it and analyzes it using a natural language processing engine. A generative AI system then generates an appropriate answer, which is then reviewed by a specialist and finally sent to the user. This system allows users to receive accurate and prompt medical advice.

[0315] This system contains training data to generate appropriate answers to specific medical questions, reducing the burden on medical experts while maintaining the accuracy of the answers. It is also available 24 hours a day, allowing users to receive medical consultations at their convenience. This is expected to reduce anxiety for patients and their families and improve the quality and efficiency of medical care.

[0316] The processing flow will be explained below.

[0317] Step 1:

[0318] Users enter medical questions from their device (smartphone or PC) through a dedicated app or website and send them as direct messages.

[0319] Step 2:

[0320] The server receives the query message sent by the user and stores it in a database.

[0321] Step 3:

[0322] The server sends the received message to a natural language processing (NLP) engine, which analyzes the message content, performing tokenization (dividing sentences into words), POS tagging (identifying parts of speech), and named entity recognition (recognizing specific names, places, diseases, etc.).

[0323] Step 4:

[0324] The server normalizes the query content analyzed by the NLP engine and sends it to the generative AI (e.g., GPT-4). The normalization process converts the query content into a format that is easy for the generative AI to understand.

[0325] Step 5:

[0326] Generative AI generates appropriate answers based on the data received, and these answers are internally verified to ensure they are accurate and reliable.

[0327] Step 6:

[0328] The server sends the generated answer to an expert, who reviews the answer and supervises its accuracy and appropriateness. Corrections and supplements are made as necessary.

[0329] Step 7:

[0330] The final answer, edited by experts, is sent to the server, which formats it in a user-friendly format.

[0331] Step 8:

[0332] The server sends the final formatted response to the user as a direct message.

[0333] Step 9:

[0334] The user receives the final answer on the device and obtains the appropriate medical advice for their question. The user checks the information displayed on the device and decides on the next step if necessary.

[0335] Example 1

[0336] 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."

[0337] Conventional medical consultation systems have limitations in terms of providing users with fast and accurate answers. Medical experts often have limited time and resources to respond directly, resulting in variations in the quality and speed of responses. Furthermore, if users want to seek medical advice at their own convenience, a system that can provide fast responses 24 hours a day is required, but such systems are not widely available at present. The objective of this invention is to provide a system that allows users to receive fast and reliable medical advice at any time.

[0338] 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.

[0339] In this invention, the server includes means for receiving a medical inquiry from a user, means for analyzing the received inquiry content using natural language processing, means for transmitting the analyzed inquiry content to a generative AI, means for transmitting an answer generated by the generative AI to an expert for review, means for transmitting the reviewed answer to the user, means for transmitting a formatted final answer to the user, and means for the user to receive the final answer and obtain appropriate medical advice. This enables users to receive prompt and reliable medical advice 24 hours a day.

[0340] The "means for receiving medical inquiries from users" refers to the means by which the server receives medical questions or doubts sent by users through the dedicated application or website.

[0341] "Means for analyzing the content of received inquiries using natural language processing" refers to a means for analyzing medical inquiries received by the server using a natural language processing engine and extracting important keywords and intent.

[0342] The "means for transmitting the analyzed query content to the generative artificial intelligence" is a means for transmitting the query content analyzed by natural language processing to the generative artificial intelligence as an appropriate prompt.

[0343] "Means for sending answers generated by generative AI to experts for review" refers to means for sending answers generated by generative AI to medical experts, who then review and confirm the content.

[0344] The "means for transmitting the answer that has been supervised to the user" is a means for transmitting the answer that has been supervised by the expert to the user's terminal.

[0345] The "means for transmitting a formatted final answer to a user" refers to a means for transmitting a final answer to a user that has been formatted in a format that is easy to understand after being supervised by an expert.

[0346] The "means for the user to receive the final answer and obtain appropriate medical advice" refers to the means by which the user's terminal receives the final answer sent from the server and the user can obtain that medical advice.

[0347] The present invention is a system that allows users to quickly and accurately resolve medical questions and concerns. It is expected that the system will be implemented in accordance with the following detailed description.

[0348] First, users can use a dedicated application or website to send medical questions in the form of direct messages, with the user's device (such as a smartphone or personal computer) sending the question to a server via the Internet.

[0349] Specifically, the user enters a question into a text input form on a dedicated app or website and presses the send button. For example, suppose the question entered is, "I feel short of breath before going to bed. What should I do?" The device then sends this question to the server.

[0350] Next, the server receives the inquiry message sent by the user. The received message is saved in a database. This database can be built using MySQL or PostgreSQL, for example. After saving, the server analyzes the message using a natural language processing engine (for example, Google Natural Language API). As a result of the analysis, the intent of the question and important keywords are extracted. For example, the keywords extracted are "before going to bed," "shortness of breath," and "measures."

[0351] The parsed query content is sent from the server to a generative AI (e.g., OpenAI's GPT-4), and the prompt text is also sent to the generative AI.

[0352] Example prompt sentence:

[0353] "Healthcare question: 'I'm having trouble breathing before bed. What should I do?' Generate an answer."

[0354] Generative AI generates appropriate answers based on the submitted data and prompts. The generated answers are based on medical training data and are designed to provide accurate and reliable information. For example, a generated answer might read, "Possible causes of shortness of breath include allergies, asthma, or heart problems. We recommend consulting a doctor first. Also, try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow."

[0355] The generated answer is sent to an expert for review. The expert reviews the generated answer and checks its accuracy and appropriateness. After the review process is complete, the server formats the final answer into an easy-to-understand format before sending it to the user. For example, it may use HTML formatting or highlight important keywords. The final formatted answer is then sent to the user's device as a direct message.

[0356] Finally, the user receives the final answer on their device and obtains appropriate medical advice for their question. This allows users to receive prompt and accurate medical consultations 24 hours a day. The system uses training data to generate appropriate answers to specific medical questions, reducing the burden on experts while maintaining the accuracy of the answers. This is expected to improve the quality and efficiency of medical care.

[0357] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0358] Step 1:

[0359] The user uses a dedicated application or website to input and submit a medical question. The input data here is the user's medical question. The device sends this input data to the server. Specifically, the user inputs a question such as, "I have shortness of breath before going to bed. What should I do?" and clicks the submit button.

[0360] Step 2:

[0361] The server receives the message sent by the user. The received data is the user's question, and to store it in the database, it executes an SQL query, for example, "INSERT INTO inquiries (user_id, question, timestamp) VALUES (?, ?, ?)". The input is the message data from the user, and the output is the record stored in the database.

[0362] Step 3:

[0363] The server sends the stored message to a natural language processing engine for analysis. The input data here is the user's question retrieved from the database. The natural language processing engine (for example, Google Natural Language API) analyzes the inquiry and extracts important keywords and intent. Specifically, it extracts keywords such as "before going to bed," "shortness of breath," and "measures." The output is the analysis results.

[0364] Step 4:

[0365] The server sends the analyzed query content to the generative AI. The input data consists of keywords from the analyzed query content and the corresponding prompt. For example, a prompt such as "This is a medical question. Please generate an answer to the question, 'I feel short of breath before going to bed. What should I do?'" is sent to the generative AI. The output is the generated answer.

[0366] Step 5:

[0367] The server receives the answer generated by the generative AI and then sends it to the expert. The input data is the answer received from the generative AI. The server sends this answer to the expert, who reviews it. The output is feedback from the expert and a revised answer.

[0368] Step 6:

[0369] The server receives the expert-edited answer and formats it. The input data is the expert-edited answer, which is then formatted into an easy-to-understand format, for example, using HTML to highlight important keywords. The output is the formatted answer.

[0370] Step 7:

[0371] The server sends a formatted final answer to the user. The input data is the formatted final answer, and the output is a message sent to the user's terminal. The user receives this final answer on their terminal and obtains the required medical advice.

[0372] In this way, each processing step works in tandem, allowing users to receive prompt and accurate medical advice 24 hours a day.

[0373] (Application example 1)

[0374] 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."

[0375] In modern society, users need to be able to quickly and accurately resolve their medical questions and concerns. However, current systems require users to directly contact medical experts, which often requires time and effort. In addition, general online information can be unreliable, and there is a lack of mechanisms for providing users with accurate medical advice. Furthermore, the lack of an efficient system that utilizes mobile devices such as smartphones makes it difficult to meet the needs of today's busy people.

[0376] 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.

[0377] In this invention, the server includes means for receiving medical inquiries from users, means for analyzing the received inquiry content using natural language processing, means for sending the analyzed inquiry content to a generative artificial intelligence, means for sending an answer generated by the generative artificial intelligence to an expert for supervision, means for sending the supervised answer to the user, and means for receiving inquiries from users using a smartphone application. This enables users to receive fast and reliable medical advice via their smartphones, and the accuracy of the answers is guaranteed through expert supervision, allowing users to alleviate their concerns and take appropriate measures quickly.

[0378] A "user" is an individual who has a medical question or concern.

[0379] An "inquiry" is a question sent by a user to resolve a medical question or doubt.

[0380] The "receiving means" is a system or device for electronically receiving an inquiry sent by a user.

[0381] "Natural language processing" is a technology that allows computers to understand and analyze human language.

[0382] "Generative artificial intelligence" is an AI system that automatically generates appropriate medical advice based on the content of the inquiry it receives.

[0383] An "expert" is an individual or team with a high level of expertise and experience in the medical field.

[0384] "Supervision" refers to the methods and processes by which experts verify the accuracy and appropriateness of the answers provided by the generative AI.

[0385] A "smartphone application" is a program that runs on a smartphone and allows users to send medical inquiries.

[0386] "Formatting methods" are techniques or processes used to prepare curated responses in a format that is easy for users to understand.

[0387] "Server" refers to a computer system that manages and executes a series of processes, such as receiving inquiries, analyzing them, generating responses, sending them to experts, and reviewing them.

[0388] "Learning data" refers to training data that is used in advance by generative artificial intelligence to generate appropriate medical advice.

[0389] MODE FOR CARRYING OUT THE INVENTION

[0390] The present invention is a system that allows users to quickly and accurately resolve medical questions and concerns. Each part of the system operates according to the following procedure.

[0391] 1. Program Generation

[0392] The program of the system for realizing the present invention mainly consists of the following components.

[0393] Receiving means: An interface (such as a smartphone application) for receiving medical inquiries from users.

[0394] Natural Language Processing: An NLP engine (e.g., SomeNLPModel) for analyzing query content and extracting important keywords.

[0395] Generative artificial intelligence: An AI system that generates medical advice based on control messages (e.g., GPT-4).

[0396] Editing tool: An interface for experts to review, correct, and approve answers generated by the AI.

[0397] Formatting: The process of sending expert-approved answers in a format that is easy for users to understand.

[0398] 2. Explain the program's processing

[0399] The program's processing begins when a query is sent from the user's smartphone application to the server. The server receives this query and analyzes its contents using a natural language processing (NLP) engine. During the analysis, important keywords are extracted. An NLP engine called SomeNLPModel is used in this step.

[0400] The extracted keywords are then sent to a generative artificial intelligence (GPT-4) to generate appropriate medical advice. The generated answers are then sent to experts, who review them for accuracy and appropriateness. This process is carried out using an expert review interface.

[0401] Once the expert-edited answers are ready for the user, they are formatted. This step ensures that the answers are easy for the user to understand. The final answers are then sent to the user via a smartphone application.

[0402] 3. Add specific examples

[0403] For example, suppose a user sends a question via a smartphone application: "Recently, I've been having trouble breathing before going to bed. Is there anything I can do about this?" In this case, the server receives the question and analyzes it using an NLP engine (SomeNLPModel). This analysis extracts keywords such as "night," "shortness of breath," and "measures." A generative artificial intelligence (GPT-4) generates appropriate medical advice based on these keywords. The generated answer is sent to an expert for review. After review, the answer is formatted before being provided to the user and sent to the user in the following format:

[0404] Example prompt sentence:

[0405] Q: Lately I've been having trouble breathing before going to bed. Is there anything I can do about it?

[0406] Keywords: night, shortness of breath, measures

[0407] Generate medical advice based on this keyword.

[0408] In this way, users can receive fast and accurate medical advice.

[0409] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0410] Step 1:

[0411] A user submits a medical inquiry through a smartphone application. The input is the user's question, and the output is the action of sending this question to the server. For example, if a user enters the question, "Recently, I've been having trouble breathing before going to bed. Is there anything I can do about it?", the question is sent to the server.

[0412] Step 2:

[0413] The server receives inquiries from users. The input is the user's question, and the output is the data of the received inquiry. The server receives and stores the data sent from the smartphone app. Specifically, the inquiry is saved in a database.

[0414] Step 3:

[0415] The query received by the server is analyzed using a natural language processing (NLP) engine. The input is the query data, and the output is the extracted keywords. For example, keywords such as "night," "shortness of breath," and "measures" are extracted. Specifically, SomeNLPModel, an NLP engine, analyzes the query and extracts important keywords.

[0416] Step 4:

[0417] The server sends the keywords from the analyzed query to a generative AI (GPT-4) system, which generates appropriate medical advice. The input is the extracted keywords, and the output is the generated medical advice. For example, the generated advice might read, "Possible causes of shortness of breath include allergies, asthma, and heart problems. We recommend that you first consult a doctor. Also, try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow." Specifically, the generative AI system outputs appropriate advice based on the prompt text.

[0418] Step 5:

[0419] The generated answer is sent to an expert for review. The input is the generated medical advice, and the output is an answer that has been reviewed and approved by an expert. Specifically, the answer output by the generative AI is checked by an expert review system, and any corrections that need to be made are made, and finally, it is approved.

[0420] Step 6:

[0421] The server formats the expert-edited answers for delivery to the user. The input is the edited answers, and the output is the final formatted answers. Specifically, the answers are formatted in a user-friendly format and delivered via a smartphone app.

[0422] Step 7:

[0423] The server sends the final answer to the user. The input is the formatted final answer, and the output is the answer that is displayed on the user's smartphone. Specifically, the server sends the formatted answer to a smartphone app, and the user views the answer.

[0424] This allows users to receive prompt and accurate medical advice.

[0425] 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.

[0426] This invention is a system that allows users to quickly resolve medical-related questions and concerns, and by combining it with an emotion engine that recognizes the user's emotions, it provides more appropriate responses. It is implemented in the following steps.

[0427] First, a user sends a medical question via a dedicated app or website in the form of a direct message. The user's device (smartphone or PC) then sends the question to the emotion engine, which analyzes the user's emotions. The emotion engine then uses the text data to analyze the user's emotional state (e.g., anxiety, anger, relief, etc.).

[0428] Next, after analyzing the user's emotions, the server receives the question message sent by the user. The received message is stored in a database, and then the inquiry is analyzed by a natural language processing (NLP) engine. During the analysis, the intent of the inquiry and important keywords are extracted. For example, if the question is "I get short of breath at night. What should I do?", the keywords "night," "short of breath," and "measures" are analyzed.

[0429] Along with the analyzed query content, the user's emotional information analyzed by the emotion engine is also sent to a generative AI (such as GPT-4). The generative AI generates an appropriate answer taking into account the user's emotions. For example, if the emotion is recognized as "anxiety," the answer will include a reassuring expression. In this way, the generated answer is based on medical training data and provides accurate and reliable information.

[0430] Instead of providing the answer directly to the user, the generated answer is first sent to an expert, who reviews the content of the generated answer and reviews it for accuracy and appropriateness. After this review process is complete, the server formats the answer in an easy-to-understand format before sending it to the user. The final answer is then sent to the user's device as a direct message.

[0431] Users can receive the final answer on their device and obtain accurate medical advice for their questions. This allows users to receive medical consultations 24 hours a day and quickly obtain reliable information supervised by experts.

[0432] For example, if a user sends a specific query such as, "I've been having trouble breathing before bed recently. I'm feeling very anxious. What should I do?", the emotion engine will interpret it as "anxiety." This information and the query are received and analyzed by the server and sent to the generative AI. The generative AI then generates a response such as, "Possible causes of your shortness of breath include allergies, asthma, and heart problems. We recommend that you first consult a doctor. We also recommend that you try improving your bedroom environment. For example, use an air purifier to remove dust and allergens and use a high pillow. We understand your anxiety, but please try these measures." A specialist will then review this response and confirm its appropriateness before sending it to the user. In this way, the user can receive an emotionally sensitive response, which can help alleviate their anxiety.

[0433] This system uses an emotion engine to recognize the user's emotions and provide that information to a generative AI system, which can then generate appropriate answers based on the user's emotions. Furthermore, after being supervised by experts, the system provides users with highly reliable information, greatly improving the quality and efficiency of medical consultations.

[0434] The processing flow will be explained below.

[0435] Step 1:

[0436] Users enter medical questions from their device (smartphone or PC) through a dedicated app or website and send them as direct messages.

[0437] Step 2:

[0438] The server receives the query message sent by the user and stores it in a database.

[0439] Step 3:

[0440] The server sends the message to the emotion engine, which analyzes the user's emotion. The emotion engine analyzes the text data to recognize the user's emotional state (e.g., anxiety, anger, relief, etc.) and stores this information for further processing.

[0441] Step 4:

[0442] The server sends the received message to a natural language processing (NLP) engine, which analyzes the message content, performing tokenization (dividing sentences into words), POS tagging (identifying parts of speech), and named entity recognition (recognizing specific names, places, diseases, etc.).

[0443] Step 5:

[0444] The server normalizes the query content analyzed by the NLP engine and the user's emotional information recognized by the emotion engine, and sends them to a generative AI (e.g., GPT-4). The normalization process converts the data into a format that is easy for the generative AI to understand.

[0445] Step 6:

[0446] Generative AI generates appropriate answers based on the normalized data. The system also takes into account the user's emotional information, so the generated answers include language tones and expressions that correspond to the emotion. For example, if the user's emotion is recognized as "anxiety," the system adds expressions that convey a sense of security to the answer.

[0447] Step 7:

[0448] The server sends the generated answer to the expert, who reviews the answer and supervises its accuracy and appropriateness, correcting or supplementing it as necessary.

[0449] Step 8:

[0450] The final answer, edited by an expert, is sent to the server, which formats it for presentation to the user. The formatted answer is formatted in a way that is easy for the user to understand.

[0451] Step 9:

[0452] The server sends the final formatted response to the user as a direct message.

[0453] Step 10:

[0454] The user receives the final answer on the device and obtains the medical advice relevant to their question. The user reviews the information displayed on the device and decides on the next step if necessary, such as making further inquiries or acting on the advice.

[0455] Example 2

[0456] 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."

[0457] In modern medical consultations, it is necessary to quickly and appropriately resolve users' anxieties and doubts. However, conventional systems generate mechanical answers without considering the user's feelings, which can lead to low user satisfaction. Furthermore, to ensure the reliability of the generated answers, expert supervision is essential, but there are only a limited number of systems that can do this efficiently.

[0458] 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.

[0459] In this invention, the server includes means for receiving a medical inquiry from a user, means for analyzing the received inquiry content using natural language processing, means for analyzing the user's emotions, means for transmitting the analyzed inquiry content and emotional information to the generative AI, means for transmitting the answer generated by the generative AI to an expert for review, and means for transmitting the answer after review to the user. This makes it possible to generate an appropriate medical answer that takes the user's emotions into consideration, and provide it to the user with a high level of reliability after it has been checked by the expert.

[0460] A "user" is someone who uses the system to resolve medical questions or concerns.

[0461] "Inquiry" refers to a medical question or consultation sent by a user.

[0462] "Natural language processing" is a technical method for analyzing the content of received inquiries and extracting intent and keywords.

[0463] "Sentiment analysis" is the process of identifying an emotional state from the text data contained in a user's query.

[0464] "Generative AI" refers to AI that generates appropriate medical answers based on the content of the inquiry and emotional information it receives.

[0465] An "expert" is someone who is qualified to oversee the content of medical answers created by generative AI and verify their accuracy and appropriateness.

[0466] "Supervision" is the act of an expert reviewing the content of answers generated by generative AI and ensuring their appropriateness and accuracy.

[0467] "Formatting" is the process of shaping the answers provided to the user into an understandable form.

[0468] "Training data" refers to the training data used by generative artificial intelligence to generate appropriate answers to medical-related questions.

[0469] The present invention is a system that enables users to receive prompt and appropriate advice regarding medical questions and concerns. By incorporating an emotion engine that analyzes the user's emotions, the system provides appropriate responses that take emotions into consideration.

[0470] First, a user sends a medical question in the form of a direct message through a dedicated application or website. For example, a user might use their smartphone to send a message along the lines of, "Lately, I've been having trouble breathing before going to bed. It's really worrying. What should I do?"

[0471] The user's device sends the message to an emotion engine. A commonly used emotion analysis tool can be used as the emotion engine. As a specific example, IBM Watson's Tone Analyzer can be used. This emotion engine analyzes the user's emotional state (anxiety, anger, relief, etc.) based on the text data and assigns an emotion label. In this specific example, the emotion label "anxiety" is obtained from the part "I am very anxious."

[0472] Next, the server receives the message sent by the user and stores it in a MySQL database. The stored data includes the user ID, message content, emotion label, etc. The stored message is then analyzed by a natural language processing (NLP) engine. NLP engines such as SpaCy and NLTK can be used as examples. During this analysis process, the intent of the inquiry and important keywords are extracted. For example, from the message "Recently, I've been having trouble breathing before going to bed," keywords such as "before going to bed," "trouble breathing," and "measures" are extracted.

[0473] The analyzed inquiry content and emotional information are sent to a generative AI. A large language model such as GPT-4 is used as the generative AI. At this time, the following message is input as a prompt: "The user has recently been experiencing difficulty breathing before going to bed, which has caused them great anxiety. Please provide appropriate medical advice."

[0474] The generative AI model generates an appropriate answer based on the prompt and emotional information. For example, it might say, "Possible causes of shortness of breath include allergies, asthma, or heart problems. We recommend that you first consult a doctor. Also, try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow. We understand that you may be feeling anxious, but please try these measures."

[0475] The generated answers are sent to experts, who review the answers for accuracy and appropriateness and make any necessary corrections. Once the review is complete, the server formats the answers in a way that is easy for users to understand, such as by organizing them into paragraphs or bullet points.

[0476] The final answer will be sent to the user's device as a direct message, allowing them to receive the final answer on their smartphone or PC and receive reliable medical advice for their question.

[0477] In this way, the present invention makes it possible to quickly provide appropriate medical answers that take into account the user's feelings.

[0478] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0479] Step 1:

[0480] A user submits a medical question through a dedicated application or website.

[0481] Input: User's question (e.g., "Recently, I've been having trouble breathing before going to bed. It's really worrying me. What should I do?")

[0482] Output: A query message is sent from the user's terminal to the server.

[0483] Step 2:

[0484] The user's terminal transmits the received question message to the emotion engine.

[0485] Input: User's question message

[0486] Data processing: The emotion engine analyzes the text data and assigns an emotion label (e.g., "anxiety").

[0487] Output: Question message with emotion label (e.g., "Anxious" from "I'm very anxious")

[0488] Step 3:

[0489] The server receives the query message sent from the user's terminal and stores it in a database.

[0490] Input: A question message with an emotion label.

[0491] Data processing: Received messages are stored in a MySQL database. The stored data includes user ID, message content, emotion label, etc.

[0492] Output: The question message stored in the database

[0493] Step 4:

[0494] The server sends the query message stored in the database to a natural language processing (NLP) engine, which analyzes the query.

[0495] Input: Question message stored in the database

[0496] Data calculation: Text analysis is performed by an NLP engine (e.g., SpaCy) to extract intent and important keywords (e.g., "before going to bed," "difficulty breathing," "measures").

[0497] Output: Parsed query content

[0498] Step 5:

[0499] The server sends the analyzed inquiry content and emotional information to the generative artificial intelligence.

[0500] Input: Parsed query content and sentiment label

[0501] Data processing: Generate prompt sentences (e.g., "The user has recently been experiencing difficulty breathing before going to bed, which has caused him great anxiety. Please provide appropriate medical advice.")

[0502] Output: The prompt sent to the generative AI model

[0503] Step 6:

[0504] Generative AI generates appropriate answers based on prompts and emotional information.

[0505] Input: A prompt sent to the generative AI model

[0506] Data computation: A generative AI model (e.g., GPT-4) generates an appropriate answer based on a prompt (e.g., "Possible causes of shortness of breath include allergies, asthma, heart problems, etc...").

[0507] Output: The generated answer

[0508] Step 7:

[0509] The server sends the generated answers to experts, who then review the content.

[0510] Input: Generated Answer

[0511] Data processing: Experts check the accuracy and appropriateness of the answers and make corrections if necessary (expert feedback)

[0512] Output: Edited answer

[0513] Step 8:

[0514] The server formats the edited answers and converts them into a user-friendly format.

[0515] Input: Edited answer

[0516] Data processing: Formatting responses, such as organizing them into paragraphs or bullet points

[0517] Output: Formatted final answer

[0518] Step 9:

[0519] The server sends the final response to the user's device as a direct message.

[0520] Input: Final formatted answer

[0521] Output: Final answer EX sent to the user's device (received by the user on their smartphone or PC)

[0522] (Application example 2)

[0523] 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."

[0524] Conventional systems have difficulty providing responses that take users' emotions into consideration, and often do not provide adequate support, especially to users who feel anxious. Furthermore, the generated answers require time-consuming supervision by experts, making it difficult to provide immediate responses. This has led to issues that make it difficult for users to obtain fast and reliable information.

[0525] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the received inquiry content using natural language processing, means for sending the analyzed inquiry content to the generative artificial intelligence, means for sending an answer generated by the generative artificial intelligence to an expert for supervision, means for sending the answer after supervision to the user, emotion analysis means for recognizing the user's emotional state, and means for providing the generative artificial intelligence with analysis results including emotion-analyzed information. This makes it possible to quickly provide a response that takes the user's emotions into consideration, and to provide highly reliable information that has been supervised by an expert.

[0526] "User inquiries" refer to questions or doubts that users send to the system.

[0527] "Natural language processing" is a technology that allows computers to understand and analyze human language.

[0528] "Generative AI" refers to AI technology that automatically generates appropriate answers based on input data.

[0529] An "expert" is someone who has advanced knowledge and experience in a particular field.

[0530] "Supervision" is the process in which experts check the accuracy and appropriateness of the generated answers.

[0531] "Emotion analysis means" is a technology that analyzes a user's text data and identifies the user's emotional state.

[0532] "Formatting means" refers to a technique that provides expert-edited answers in a format that is easy for users to understand.

[0533] "Learning data" is training data that generative artificial intelligence uses to generate appropriate answers.

[0534] A system for implementing the present invention has the following configuration.

[0535] 1. The server first receives an inquiry from the user. The user enters security questions or concerns through a dedicated application or website. For example, a user could use a device such as a smartphone or PC to send an inquiry such as, "I've been feeling uneasy about online banking lately. What can I do to make it safer?"

[0536] 2. When a query arrives at the server, the server analyzes it using a natural language processing (NLP) engine to extract important keywords and intent from the query. Libraries such as TextBlob and SentimentAnalysis are used here.

[0537] 3. Next, the server uses emotion analysis to recognize the user's emotions. Tools such as EmotionEngine provide technology that analyzes the user's text data and identifies their emotional state (e.g., anxiety, relief, etc.).

[0538] 4. The results of the sentiment analysis and the analysis results from the NLP engine are sent to a generative AI model (such as GPT-4). The generative AI model uses this data to generate an appropriate response that takes the user's emotions into consideration.

[0539] 5. The generated answers are sent to experts, who review them for accuracy and appropriateness. For example, a security expert might review an answer about how to use online banking safely.

[0540] 6. After the expert has verified the answer, the server formats the final answer for delivery to the user. This formatted answer is provided in a format that is easy for the user to understand.

[0541] 7. Finally, the server sends the final formatted response to the user's terminal.

[0542] As a specific example, if a user sends a query such as, "I've been feeling uneasy about online banking lately. What can I do to make it safer?", the emotion engine will interpret it as "uneasy." This information and the content of the query are received and analyzed by the server, and then sent to the generative AI. The generative AI will then generate a response such as, "To increase the security of online banking, it is important to use a strong password and set up two-step authentication. Also, be careful not to click on suspicious links." This is then reviewed by security experts, and the final response is sent to the user.

[0543] Examples of prompts used by generative AI models include:

[0544] "Users are worried and have asked about how to use online banking safely. Please reassure them and provide helpful advice."

[0545] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0546] Step 1:

[0547] A user uses a device such as a smartphone or PC to send a security inquiry to a server through a dedicated application or website. The input is the user's inquiry text, and the output is the inquiry data sent to the server. Specifically, the text sent might be something like, "I've been feeling uneasy about online banking lately. What can I do to make it safer?"

[0548] Step 2:

[0549] The server analyzes the received inquiry using a natural language processing (NLP) engine. The input is the inquiry text, and the output is the analyzed keywords and intent. Specifically, TextBlob and SentimentAnalysis are used to extract important keywords such as "online banking," "anxiety," and "safety."

[0550] Step 3:

[0551] The server recognizes the user's emotional state using an emotion analysis means. The input is the user's query text, and the output is the recognized emotional information. Specifically, the server uses the Emotion Engine to analyze the user's emotion (e.g., anxiety) and obtains it as data.

[0552] Step 4:

[0553] The server sends the analyzed keywords and emotional information to a generative artificial intelligence (generative AI model). The input is the analyzed keywords and emotional information, and the output is a generated answer. Specifically, it uses the OpenAI API (e.g., GPT-4) to generate an answer such as, "To increase the security of your online banking, it is important to use a strong password and set up two-factor authentication. Also, be careful not to click on suspicious links."

[0554] Step 5:

[0555] The generated answer is sent from the server to an expert for review. The input is the generated answer text, and the output is the answer that has been checked and corrected by the expert. Specifically, the security expert checks the accuracy and appropriateness of the answer content and makes corrections as necessary.

[0556] Step 6:

[0557] The server formats the expert-edited answers for delivery to the user. The input is the edited answer text, and the output is the formatted answer. Specifically, the server converts the answer content into a format that is easy for the user to understand.

[0558] Step 7:

[0559] Finally, the server sends the formatted answer to the user's device. The input is the formatted answer text, and the output is the answer received on the user's device. Specifically, the intuitive and easy-to-read answer is displayed on the user's smartphone or PC.

[0560] Through the above processing steps, users can receive prompt and reliable security advice that takes their emotions into consideration.

[0561] 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.

[0562] 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.

[0563] 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.

[0564] [Third embodiment]

[0565] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0566] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0567] 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).

[0568] 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.

[0569] 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.

[0570] 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).

[0571] 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.

[0572] 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.

[0573] 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.

[0574] 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.

[0575] 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.

[0576] 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."

[0577] The present invention is a system that enables users to quickly resolve medical-related questions and concerns, and is implemented in the following manner.

[0578] First, users can send medical questions in the form of direct messages through a dedicated app or website, and the user's device (smartphone or PC) then sends the question to the server.

[0579] Next, the server receives an inquiry message from the user. The received message is stored in a database and then analyzed by a natural language processing (NLP) engine. During the analysis, the intent of the inquiry and important keywords are extracted. For example, if the question is, "I get short of breath at night. What should I do?", the keywords "night," "short of breath," and "measures" are analyzed.

[0580] The analyzed query content is sent to a generative AI (such as GPT-4), which generates an appropriate answer based on the data sent. The answers generated here are based on medical training data and are designed to provide accurate and reliable information. For example, a generated answer might read, "Possible causes of shortness of breath include allergies, asthma, and heart problems. We recommend that you first consult a doctor. We also recommend that you try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow."

[0581] Instead of providing the answer directly to the user, the generated answer is first sent to an expert, who reviews the content of the generated answer and reviews it for accuracy and appropriateness. After this review process is complete, the server formats the answer in an easy-to-understand format before sending it to the user. The final answer is then sent to the user's device as a direct message.

[0582] Users can receive the final answer on their device and obtain accurate medical advice for their questions. This allows users to consult with a doctor at any time and quickly obtain reliable information supervised by an expert.

[0583] For example, if a user sends a specific question such as, "Recently, I've been having trouble breathing before going to bed. Is there anything I can do about this?", the server receives it and analyzes it using a natural language processing engine. A generative AI system then generates an appropriate answer, which is then reviewed by a specialist and finally sent to the user. This system allows users to receive accurate and prompt medical advice.

[0584] This system contains training data to generate appropriate answers to specific medical questions, reducing the burden on medical experts while maintaining the accuracy of the answers. It is also available 24 hours a day, allowing users to receive medical consultations at their convenience. This is expected to reduce anxiety for patients and their families and improve the quality and efficiency of medical care.

[0585] The processing flow will be explained below.

[0586] Step 1:

[0587] Users enter medical questions from their device (smartphone or PC) through a dedicated app or website and send them as direct messages.

[0588] Step 2:

[0589] The server receives the query message sent by the user and stores it in a database.

[0590] Step 3:

[0591] The server sends the received message to a natural language processing (NLP) engine, which analyzes the message content, performing tokenization (dividing sentences into words), POS tagging (identifying parts of speech), and named entity recognition (recognizing specific names, places, diseases, etc.).

[0592] Step 4:

[0593] The server normalizes the query content analyzed by the NLP engine and sends it to the generative AI (e.g., GPT-4). The normalization process converts the query content into a format that is easy for the generative AI to understand.

[0594] Step 5:

[0595] Generative AI generates appropriate answers based on the data received, and these answers are internally verified to ensure they are accurate and reliable.

[0596] Step 6:

[0597] The server sends the generated answer to an expert, who reviews the answer and supervises its accuracy and appropriateness. Corrections and supplements are made as necessary.

[0598] Step 7:

[0599] The final answer, edited by experts, is sent to the server, which formats it in a user-friendly format.

[0600] Step 8:

[0601] The server sends the final formatted response to the user as a direct message.

[0602] Step 9:

[0603] The user receives the final answer on the device and obtains the appropriate medical advice for their question. The user checks the information displayed on the device and decides on the next step if necessary.

[0604] Example 1

[0605] 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."

[0606] Conventional medical consultation systems have limitations in terms of providing users with fast and accurate answers. Medical experts often have limited time and resources to respond directly, resulting in variations in the quality and speed of responses. Furthermore, if users want to seek medical advice at their own convenience, a system that can provide fast responses 24 hours a day is required, but such systems are not widely available at present. The objective of this invention is to provide a system that allows users to receive fast and reliable medical advice at any time.

[0607] 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.

[0608] In this invention, the server includes means for receiving a medical inquiry from a user, means for analyzing the received inquiry content using natural language processing, means for transmitting the analyzed inquiry content to a generative AI, means for transmitting an answer generated by the generative AI to an expert for review, means for transmitting the reviewed answer to the user, means for transmitting a formatted final answer to the user, and means for the user to receive the final answer and obtain appropriate medical advice. This enables users to receive prompt and reliable medical advice 24 hours a day.

[0609] The "means for receiving medical inquiries from users" refers to the means by which the server receives medical questions or doubts sent by users through the dedicated application or website.

[0610] "Means for analyzing the content of received inquiries using natural language processing" refers to a means for analyzing medical inquiries received by the server using a natural language processing engine and extracting important keywords and intent.

[0611] The "means for transmitting the analyzed query content to the generative artificial intelligence" is a means for transmitting the query content analyzed by natural language processing to the generative artificial intelligence as an appropriate prompt.

[0612] "Means for sending answers generated by generative AI to experts for review" refers to means for sending answers generated by generative AI to medical experts, who then review and confirm the content.

[0613] The "means for transmitting the answer that has been supervised to the user" is a means for transmitting the answer that has been supervised by the expert to the user's terminal.

[0614] The "means for transmitting a formatted final answer to a user" refers to a means for transmitting a final answer to a user that has been formatted in a format that is easy to understand after being supervised by an expert.

[0615] The "means for the user to receive the final answer and obtain appropriate medical advice" refers to the means by which the user's terminal receives the final answer sent from the server and the user can obtain that medical advice.

[0616] The present invention is a system that allows users to quickly and accurately resolve medical questions and concerns. It is expected that the system will be implemented in accordance with the following detailed description.

[0617] First, users can use a dedicated application or website to send medical questions in the form of direct messages, with the user's device (such as a smartphone or personal computer) sending the question to a server via the Internet.

[0618] Specifically, the user enters a question into a text input form on a dedicated app or website and presses the send button. For example, suppose the question entered is, "I feel short of breath before going to bed. What should I do?" The device then sends this question to the server.

[0619] Next, the server receives the inquiry message sent by the user. The received message is saved in a database. This database can be built using MySQL or PostgreSQL, for example. After saving, the server analyzes the message using a natural language processing engine (for example, Google Natural Language API). As a result of the analysis, the intent of the question and important keywords are extracted. For example, the keywords extracted are "before going to bed," "shortness of breath," and "measures."

[0620] The parsed query content is sent from the server to a generative AI (e.g., OpenAI's GPT-4), and the prompt text is also sent to the generative AI.

[0621] Example prompt sentence:

[0622] "Healthcare question: 'I'm having trouble breathing before bed. What should I do?' Generate an answer."

[0623] Generative AI generates appropriate answers based on the submitted data and prompts. The generated answers are based on medical training data and are designed to provide accurate and reliable information. For example, a generated answer might read, "Possible causes of shortness of breath include allergies, asthma, or heart problems. We recommend consulting a doctor first. Also, try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow."

[0624] The generated answer is sent to an expert for review. The expert reviews the generated answer and checks its accuracy and appropriateness. After the review process is complete, the server formats the final answer into an easy-to-understand format before sending it to the user. For example, it may use HTML formatting or highlight important keywords. The final formatted answer is then sent to the user's device as a direct message.

[0625] Finally, the user receives the final answer on their device and obtains appropriate medical advice for their question. This allows users to receive prompt and accurate medical consultations 24 hours a day. The system uses training data to generate appropriate answers to specific medical questions, reducing the burden on experts while maintaining the accuracy of the answers. This is expected to improve the quality and efficiency of medical care.

[0626] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0627] Step 1:

[0628] The user uses a dedicated application or website to input and submit a medical question. The input data here is the user's medical question. The device sends this input data to the server. Specifically, the user inputs a question such as, "I have shortness of breath before going to bed. What should I do?" and clicks the submit button.

[0629] Step 2:

[0630] The server receives the message sent by the user. The received data is the user's question, and to store it in the database, it executes an SQL query, for example, "INSERT INTO inquiries (user_id, question, timestamp) VALUES (?, ?, ?)". The input is the message data from the user, and the output is the record stored in the database.

[0631] Step 3:

[0632] The server sends the stored message to a natural language processing engine for analysis. The input data here is the user's question retrieved from the database. The natural language processing engine (for example, Google Natural Language API) analyzes the inquiry and extracts important keywords and intent. Specifically, it extracts keywords such as "before going to bed," "shortness of breath," and "measures." The output is the analysis results.

[0633] Step 4:

[0634] The server sends the analyzed query content to the generative AI. The input data consists of keywords from the analyzed query content and the corresponding prompt. For example, a prompt such as "This is a medical question. Please generate an answer to the question, 'I feel short of breath before going to bed. What should I do?'" is sent to the generative AI. The output is the generated answer.

[0635] Step 5:

[0636] The server receives the answer generated by the generative AI and then sends it to the expert. The input data is the answer received from the generative AI. The server sends this answer to the expert, who reviews it. The output is feedback from the expert and a revised answer.

[0637] Step 6:

[0638] The server receives the expert-edited answer and formats it. The input data is the expert-edited answer, which is then formatted into an easy-to-understand format, for example, using HTML to highlight important keywords. The output is the formatted answer.

[0639] Step 7:

[0640] The server sends a formatted final answer to the user. The input data is the formatted final answer, and the output is a message sent to the user's terminal. The user receives this final answer on their terminal and obtains the required medical advice.

[0641] In this way, each processing step works in tandem, allowing users to receive prompt and accurate medical advice 24 hours a day.

[0642] (Application example 1)

[0643] 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."

[0644] In modern society, users need to be able to quickly and accurately resolve their medical questions and concerns. However, current systems require users to directly contact medical experts, which often requires time and effort. In addition, general online information can be unreliable, and there is a lack of mechanisms for providing users with accurate medical advice. Furthermore, the lack of an efficient system that utilizes mobile devices such as smartphones makes it difficult to meet the needs of today's busy people.

[0645] 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.

[0646] In this invention, the server includes means for receiving medical inquiries from users, means for analyzing the received inquiry content using natural language processing, means for sending the analyzed inquiry content to a generative artificial intelligence, means for sending an answer generated by the generative artificial intelligence to an expert for supervision, means for sending the supervised answer to the user, and means for receiving inquiries from users using a smartphone application. This enables users to receive fast and reliable medical advice via their smartphones, and the accuracy of the answers is guaranteed through expert supervision, allowing users to alleviate their concerns and take appropriate measures quickly.

[0647] A "user" is an individual who has a medical question or concern.

[0648] An "inquiry" is a question sent by a user to resolve a medical question or doubt.

[0649] The "receiving means" is a system or device for electronically receiving an inquiry sent by a user.

[0650] "Natural language processing" is a technology that allows computers to understand and analyze human language.

[0651] "Generative artificial intelligence" is an AI system that automatically generates appropriate medical advice based on the content of the inquiry it receives.

[0652] An "expert" is an individual or team with a high level of expertise and experience in the medical field.

[0653] "Supervision" refers to the methods and processes by which experts verify the accuracy and appropriateness of the answers provided by the generative AI.

[0654] A "smartphone application" is a program that runs on a smartphone and allows users to send medical inquiries.

[0655] "Formatting methods" are techniques or processes used to prepare curated responses in a format that is easy for users to understand.

[0656] "Server" refers to a computer system that manages and executes a series of processes, such as receiving inquiries, analyzing them, generating responses, sending them to experts, and reviewing them.

[0657] "Learning data" refers to training data that is used in advance by generative artificial intelligence to generate appropriate medical advice.

[0658] MODE FOR CARRYING OUT THE INVENTION

[0659] The present invention is a system that allows users to quickly and accurately resolve medical questions and concerns. Each part of the system operates according to the following procedure.

[0660] 1. Program Generation

[0661] The program of the system for realizing the present invention mainly consists of the following components.

[0662] Receiving means: An interface (such as a smartphone application) for receiving medical inquiries from users.

[0663] Natural Language Processing: An NLP engine (e.g., SomeNLPModel) for analyzing query content and extracting important keywords.

[0664] Generative artificial intelligence: An AI system that generates medical advice based on control messages (e.g., GPT-4).

[0665] Editing tool: An interface for experts to review, correct, and approve answers generated by the AI.

[0666] Formatting: The process of sending expert-approved answers in a format that is easy for users to understand.

[0667] 2. Explain the program's processing

[0668] The program's processing begins when a query is sent from the user's smartphone application to the server. The server receives this query and analyzes its contents using a natural language processing (NLP) engine. During the analysis, important keywords are extracted. An NLP engine called SomeNLPModel is used in this step.

[0669] The extracted keywords are then sent to a generative artificial intelligence (GPT-4) to generate appropriate medical advice. The generated answers are then sent to experts, who review them for accuracy and appropriateness. This process is carried out using an expert review interface.

[0670] Once the expert-edited answers are ready for the user, they are formatted. This step ensures that the answers are easy for the user to understand. The final answers are then sent to the user via a smartphone application.

[0671] 3. Add specific examples

[0672] For example, suppose a user sends a question via a smartphone application: "Recently, I've been having trouble breathing before going to bed. Is there anything I can do about this?" In this case, the server receives the question and analyzes it using an NLP engine (SomeNLPModel). This analysis extracts keywords such as "night," "shortness of breath," and "measures." A generative artificial intelligence (GPT-4) generates appropriate medical advice based on these keywords. The generated answer is sent to an expert for review. After review, the answer is formatted before being provided to the user and sent to the user in the following format:

[0673] Example prompt sentence:

[0674] Q: Lately I've been having trouble breathing before going to bed. Is there anything I can do about it?

[0675] Keywords: night, shortness of breath, measures

[0676] Generate medical advice based on this keyword.

[0677] In this way, users can receive fast and accurate medical advice.

[0678] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0679] Step 1:

[0680] A user submits a medical inquiry through a smartphone application. The input is the user's question, and the output is the action of sending this question to the server. For example, if a user enters the question, "Recently, I've been having trouble breathing before going to bed. Is there anything I can do about it?", the question is sent to the server.

[0681] Step 2:

[0682] The server receives inquiries from users. The input is the user's question, and the output is the data of the received inquiry. The server receives and stores the data sent from the smartphone app. Specifically, the inquiry is saved in a database.

[0683] Step 3:

[0684] The query received by the server is analyzed using a natural language processing (NLP) engine. The input is the query data, and the output is the extracted keywords. For example, keywords such as "night," "shortness of breath," and "measures" are extracted. Specifically, SomeNLPModel, an NLP engine, analyzes the query and extracts important keywords.

[0685] Step 4:

[0686] The server sends the keywords from the analyzed query to a generative AI (GPT-4) system, which generates appropriate medical advice. The input is the extracted keywords, and the output is the generated medical advice. For example, the generated advice might read, "Possible causes of shortness of breath include allergies, asthma, and heart problems. We recommend that you first consult a doctor. Also, try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow." Specifically, the generative AI system outputs appropriate advice based on the prompt text.

[0687] Step 5:

[0688] The generated answer is sent to an expert for review. The input is the generated medical advice, and the output is an answer that has been reviewed and approved by an expert. Specifically, the answer output by the generative AI is checked by an expert review system, and any corrections that need to be made are made, and finally, it is approved.

[0689] Step 6:

[0690] The server formats the expert-edited answers for delivery to the user. The input is the edited answers, and the output is the final formatted answers. Specifically, the answers are formatted in a user-friendly format and delivered via a smartphone app.

[0691] Step 7:

[0692] The server sends the final answer to the user. The input is the formatted final answer, and the output is the answer that is displayed on the user's smartphone. Specifically, the server sends the formatted answer to a smartphone app, and the user views the answer.

[0693] This allows users to receive prompt and accurate medical advice.

[0694] 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.

[0695] This invention is a system that allows users to quickly resolve medical-related questions and concerns, and by combining it with an emotion engine that recognizes the user's emotions, it provides more appropriate responses. It is implemented in the following steps.

[0696] First, a user sends a medical question via a dedicated app or website in the form of a direct message. The user's device (smartphone or PC) then sends the question to the emotion engine, which analyzes the user's emotions. The emotion engine then uses the text data to analyze the user's emotional state (e.g., anxiety, anger, relief, etc.).

[0697] Next, after analyzing the user's emotions, the server receives the question message sent by the user. The received message is stored in a database, and then the inquiry is analyzed by a natural language processing (NLP) engine. During the analysis, the intent of the inquiry and important keywords are extracted. For example, if the question is "I get short of breath at night. What should I do?", the keywords "night," "short of breath," and "measures" are analyzed.

[0698] Along with the analyzed query content, the user's emotional information analyzed by the emotion engine is also sent to a generative AI (such as GPT-4). The generative AI generates an appropriate answer taking into account the user's emotions. For example, if the emotion is recognized as "anxiety," the answer will include a reassuring expression. In this way, the generated answer is based on medical training data and provides accurate and reliable information.

[0699] Instead of providing the answer directly to the user, the generated answer is first sent to an expert, who reviews the content of the generated answer and reviews it for accuracy and appropriateness. After this review process is complete, the server formats the answer in an easy-to-understand format before sending it to the user. The final answer is then sent to the user's device as a direct message.

[0700] Users can receive the final answer on their device and obtain accurate medical advice for their questions. This allows users to receive medical consultations 24 hours a day and quickly obtain reliable information supervised by experts.

[0701] For example, if a user sends a specific query such as, "I've been having trouble breathing before bed recently. I'm feeling very anxious. What should I do?", the emotion engine will interpret it as "anxiety." This information and the query are received and analyzed by the server and sent to the generative AI. The generative AI then generates a response such as, "Possible causes of your shortness of breath include allergies, asthma, and heart problems. We recommend that you first consult a doctor. We also recommend that you try improving your bedroom environment. For example, use an air purifier to remove dust and allergens and use a high pillow. We understand your anxiety, but please try these measures." A specialist will then review this response and confirm its appropriateness before sending it to the user. In this way, the user can receive an emotionally sensitive response, which can help alleviate their anxiety.

[0702] This system uses an emotion engine to recognize the user's emotions and provide that information to a generative AI system, which can then generate appropriate answers based on the user's emotions. Furthermore, after being supervised by experts, the system provides users with highly reliable information, greatly improving the quality and efficiency of medical consultations.

[0703] The processing flow will be explained below.

[0704] Step 1:

[0705] Users enter medical questions from their device (smartphone or PC) through a dedicated app or website and send them as direct messages.

[0706] Step 2:

[0707] The server receives the query message sent by the user and stores it in a database.

[0708] Step 3:

[0709] The server sends the message to the emotion engine, which analyzes the user's emotion. The emotion engine analyzes the text data to recognize the user's emotional state (e.g., anxiety, anger, relief, etc.) and stores this information for further processing.

[0710] Step 4:

[0711] The server sends the received message to a natural language processing (NLP) engine, which analyzes the message content, performing tokenization (dividing sentences into words), POS tagging (identifying parts of speech), and named entity recognition (recognizing specific names, places, diseases, etc.).

[0712] Step 5:

[0713] The server normalizes the query content analyzed by the NLP engine and the user's emotional information recognized by the emotion engine, and sends them to a generative AI (e.g., GPT-4). The normalization process converts the data into a format that is easy for the generative AI to understand.

[0714] Step 6:

[0715] Generative AI generates appropriate answers based on the normalized data. The system also takes into account the user's emotional information, so the generated answers include language tones and expressions that correspond to the emotion. For example, if the user's emotion is recognized as "anxiety," the system adds expressions that convey a sense of security to the answer.

[0716] Step 7:

[0717] The server sends the generated answer to the expert, who reviews the answer and supervises its accuracy and appropriateness, correcting or supplementing it as necessary.

[0718] Step 8:

[0719] The final answer, edited by an expert, is sent to the server, which formats it for presentation to the user. The formatted answer is formatted in a way that is easy for the user to understand.

[0720] Step 9:

[0721] The server sends the final formatted response to the user as a direct message.

[0722] Step 10:

[0723] The user receives the final answer on the device and obtains the medical advice relevant to their question. The user reviews the information displayed on the device and decides on the next step if necessary, such as making further inquiries or acting on the advice.

[0724] Example 2

[0725] 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."

[0726] In modern medical consultations, it is necessary to quickly and appropriately resolve users' anxieties and doubts. However, conventional systems generate mechanical answers without considering the user's feelings, which can lead to low user satisfaction. Furthermore, to ensure the reliability of the generated answers, expert supervision is essential, but there are only a limited number of systems that can do this efficiently.

[0727] 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.

[0728] In this invention, the server includes means for receiving a medical inquiry from a user, means for analyzing the received inquiry content using natural language processing, means for analyzing the user's emotions, means for transmitting the analyzed inquiry content and emotional information to the generative AI, means for transmitting the answer generated by the generative AI to an expert for review, and means for transmitting the answer after review to the user. This makes it possible to generate an appropriate medical answer that takes the user's emotions into consideration, and provide it to the user with a high level of reliability after it has been checked by the expert.

[0729] A "user" is someone who uses the system to resolve medical questions or concerns.

[0730] "Inquiry" refers to a medical question or consultation sent by a user.

[0731] "Natural language processing" is a technical method for analyzing the content of received inquiries and extracting intent and keywords.

[0732] "Sentiment analysis" is the process of identifying an emotional state from the text data contained in a user's query.

[0733] "Generative AI" refers to AI that generates appropriate medical answers based on the content of the inquiry and emotional information it receives.

[0734] An "expert" is someone who is qualified to oversee the content of medical answers created by generative AI and verify their accuracy and appropriateness.

[0735] "Supervision" is the act of an expert reviewing the content of answers generated by generative AI and ensuring their appropriateness and accuracy.

[0736] "Formatting" is the process of shaping the answers provided to the user into an understandable form.

[0737] "Training data" refers to the training data used by generative artificial intelligence to generate appropriate answers to medical-related questions.

[0738] The present invention is a system that enables users to receive prompt and appropriate advice regarding medical questions and concerns. By incorporating an emotion engine that analyzes the user's emotions, the system provides appropriate responses that take emotions into consideration.

[0739] First, a user sends a medical question in the form of a direct message through a dedicated application or website. For example, a user might use their smartphone to send a message along the lines of, "Lately, I've been having trouble breathing before going to bed. It's really worrying. What should I do?"

[0740] The user's device sends the message to an emotion engine. A commonly used emotion analysis tool can be used as the emotion engine. As a specific example, IBM Watson's Tone Analyzer can be used. This emotion engine analyzes the user's emotional state (anxiety, anger, relief, etc.) based on the text data and assigns an emotion label. In this specific example, the emotion label "anxiety" is obtained from the part "I am very anxious."

[0741] Next, the server receives the message sent by the user and stores it in a MySQL database. The stored data includes the user ID, message content, emotion label, etc. The stored message is then analyzed by a natural language processing (NLP) engine. NLP engines such as SpaCy and NLTK can be used as examples. During this analysis process, the intent of the inquiry and important keywords are extracted. For example, from the message "Recently, I've been having trouble breathing before going to bed," keywords such as "before going to bed," "trouble breathing," and "measures" are extracted.

[0742] The analyzed inquiry content and emotional information are sent to a generative AI. A large language model such as GPT-4 is used as the generative AI. At this time, the following message is input as a prompt: "The user has recently been experiencing difficulty breathing before going to bed, which has caused them great anxiety. Please provide appropriate medical advice."

[0743] The generative AI model generates an appropriate answer based on the prompt and emotional information. For example, it might say, "Possible causes of shortness of breath include allergies, asthma, or heart problems. We recommend that you first consult a doctor. Also, try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow. We understand that you may be feeling anxious, but please try these measures."

[0744] The generated answers are sent to experts, who review the answers for accuracy and appropriateness and make any necessary corrections. Once the review is complete, the server formats the answers in a way that is easy for users to understand, such as by organizing them into paragraphs or bullet points.

[0745] The final answer will be sent to the user's device as a direct message, allowing them to receive the final answer on their smartphone or PC and receive reliable medical advice for their question.

[0746] In this way, the present invention makes it possible to quickly provide appropriate medical answers that take into account the user's feelings.

[0747] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0748] Step 1:

[0749] A user submits a medical question through a dedicated application or website.

[0750] Input: User's question (e.g., "Recently, I've been having trouble breathing before going to bed. It's really worrying me. What should I do?")

[0751] Output: A query message is sent from the user's terminal to the server.

[0752] Step 2:

[0753] The user's terminal transmits the received question message to the emotion engine.

[0754] Input: User's question message

[0755] Data processing: The emotion engine analyzes the text data and assigns an emotion label (e.g., "anxiety").

[0756] Output: Question message with emotion label (e.g., "Anxious" from "I'm very anxious")

[0757] Step 3:

[0758] The server receives the query message sent from the user's terminal and stores it in a database.

[0759] Input: A question message with an emotion label.

[0760] Data processing: Received messages are stored in a MySQL database. The stored data includes user ID, message content, emotion label, etc.

[0761] Output: The question message stored in the database

[0762] Step 4:

[0763] The server sends the query message stored in the database to a natural language processing (NLP) engine, which analyzes the query.

[0764] Input: Question message stored in the database

[0765] Data calculation: Text analysis is performed by an NLP engine (e.g., SpaCy) to extract intent and important keywords (e.g., "before going to bed," "difficulty breathing," "measures").

[0766] Output: Parsed query content

[0767] Step 5:

[0768] The server sends the analyzed inquiry content and emotional information to the generative artificial intelligence.

[0769] Input: Parsed query content and sentiment label

[0770] Data processing: Generate prompt sentences (e.g., "The user has recently been experiencing difficulty breathing before going to bed, which has caused him great anxiety. Please provide appropriate medical advice.")

[0771] Output: The prompt sent to the generative AI model

[0772] Step 6:

[0773] Generative AI generates appropriate answers based on prompts and emotional information.

[0774] Input: A prompt sent to the generative AI model

[0775] Data computation: A generative AI model (e.g., GPT-4) generates an appropriate answer based on a prompt (e.g., "Possible causes of shortness of breath include allergies, asthma, heart problems, etc...").

[0776] Output: The generated answer

[0777] Step 7:

[0778] The server sends the generated answers to experts, who then review the content.

[0779] Input: Generated Answer

[0780] Data processing: Experts check the accuracy and appropriateness of the answers and make corrections if necessary (expert feedback)

[0781] Output: Edited answer

[0782] Step 8:

[0783] The server formats the edited answers and converts them into a user-friendly format.

[0784] Input: Edited answer

[0785] Data processing: Formatting responses, such as organizing them into paragraphs or bullet points

[0786] Output: Formatted final answer

[0787] Step 9:

[0788] The server sends the final response to the user's device as a direct message.

[0789] Input: Final formatted answer

[0790] Output: Final answer EX sent to the user's device (received by the user on their smartphone or PC)

[0791] (Application example 2)

[0792] 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."

[0793] Conventional systems have difficulty providing responses that take users' emotions into consideration, and often do not provide adequate support, especially to users who feel anxious. Furthermore, the generated answers require time-consuming supervision by experts, making it difficult to provide immediate responses. This has led to issues that make it difficult for users to obtain fast and reliable information.

[0794] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the received inquiry content using natural language processing, means for sending the analyzed inquiry content to the generative artificial intelligence, means for sending an answer generated by the generative artificial intelligence to an expert for supervision, means for sending the answer after supervision to the user, emotion analysis means for recognizing the user's emotional state, and means for providing the generative artificial intelligence with analysis results including emotion-analyzed information. This makes it possible to quickly provide a response that takes the user's emotions into consideration, and to provide highly reliable information that has been supervised by an expert.

[0795] "User inquiries" refer to questions or doubts that users send to the system.

[0796] "Natural language processing" is a technology that allows computers to understand and analyze human language.

[0797] "Generative AI" refers to AI technology that automatically generates appropriate answers based on input data.

[0798] An "expert" is someone who has advanced knowledge and experience in a particular field.

[0799] "Supervision" is the process in which experts check the accuracy and appropriateness of the generated answers.

[0800] "Emotion analysis means" is a technology that analyzes a user's text data and identifies the user's emotional state.

[0801] "Formatting means" refers to a technique that provides expert-edited answers in a format that is easy for users to understand.

[0802] "Learning data" is training data that generative artificial intelligence uses to generate appropriate answers.

[0803] A system for implementing the present invention has the following configuration.

[0804] 1. The server first receives an inquiry from the user. The user enters security questions or concerns through a dedicated application or website. For example, a user could use a device such as a smartphone or PC to send an inquiry such as, "I've been feeling uneasy about online banking lately. What can I do to make it safer?"

[0805] 2. When a query arrives at the server, the server analyzes it using a natural language processing (NLP) engine to extract important keywords and intent from the query. Libraries such as TextBlob and SentimentAnalysis are used here.

[0806] 3. Next, the server uses emotion analysis to recognize the user's emotions. Tools such as EmotionEngine provide technology that analyzes the user's text data and identifies their emotional state (e.g., anxiety, relief, etc.).

[0807] 4. The results of the sentiment analysis and the analysis results from the NLP engine are sent to a generative AI model (such as GPT-4). The generative AI model uses this data to generate an appropriate response that takes the user's emotions into consideration.

[0808] 5. The generated answers are sent to experts, who review them for accuracy and appropriateness. For example, a security expert might review an answer about how to use online banking safely.

[0809] 6. After the expert has verified the answer, the server formats the final answer for delivery to the user. This formatted answer is provided in a format that is easy for the user to understand.

[0810] 7. Finally, the server sends the final formatted response to the user's terminal.

[0811] As a specific example, if a user sends a query such as, "I've been feeling uneasy about online banking lately. What can I do to make it safer?", the emotion engine will interpret it as "uneasy." This information and the content of the query are received and analyzed by the server, and then sent to the generative AI. The generative AI will then generate a response such as, "To increase the security of online banking, it is important to use a strong password and set up two-step authentication. Also, be careful not to click on suspicious links." This is then reviewed by security experts, and the final response is sent to the user.

[0812] Examples of prompts used by generative AI models include:

[0813] "Users are worried and have asked about how to use online banking safely. Please reassure them and provide helpful advice."

[0814] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0815] Step 1:

[0816] A user uses a device such as a smartphone or PC to send a security inquiry to a server through a dedicated application or website. The input is the user's inquiry text, and the output is the inquiry data sent to the server. Specifically, the text sent might be something like, "I've been feeling uneasy about online banking lately. What can I do to make it safer?"

[0817] Step 2:

[0818] The server analyzes the received inquiry using a natural language processing (NLP) engine. The input is the inquiry text, and the output is the analyzed keywords and intent. Specifically, TextBlob and SentimentAnalysis are used to extract important keywords such as "online banking," "anxiety," and "safety."

[0819] Step 3:

[0820] The server recognizes the user's emotional state using an emotion analysis means. The input is the user's query text, and the output is the recognized emotional information. Specifically, the server uses the Emotion Engine to analyze the user's emotion (e.g., anxiety) and obtains it as data.

[0821] Step 4:

[0822] The server sends the analyzed keywords and emotional information to a generative artificial intelligence (generative AI model). The input is the analyzed keywords and emotional information, and the output is a generated answer. Specifically, it uses the OpenAI API (e.g., GPT-4) to generate an answer such as, "To increase the security of your online banking, it is important to use a strong password and set up two-factor authentication. Also, be careful not to click on suspicious links."

[0823] Step 5:

[0824] The generated answer is sent from the server to an expert for review. The input is the generated answer text, and the output is the answer that has been checked and corrected by the expert. Specifically, the security expert checks the accuracy and appropriateness of the answer content and makes corrections as necessary.

[0825] Step 6:

[0826] The server formats the expert-edited answers for delivery to the user. The input is the edited answer text, and the output is the formatted answer. Specifically, the server converts the answer content into a format that is easy for the user to understand.

[0827] Step 7:

[0828] Finally, the server sends the formatted answer to the user's device. The input is the formatted answer text, and the output is the answer received on the user's device. Specifically, the intuitive and easy-to-read answer is displayed on the user's smartphone or PC.

[0829] Through the above processing steps, users can receive prompt and reliable security advice that takes their emotions into consideration.

[0830] 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.

[0831] 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.

[0832] 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.

[0833] [Fourth embodiment]

[0834] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0835] 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.

[0836] 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).

[0837] 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.

[0838] 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.

[0839] 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).

[0840] 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.

[0841] 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.

[0842] 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.

[0843] 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.

[0844] 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.

[0845] 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.

[0846] 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."

[0847] The present invention is a system that enables users to quickly resolve medical-related questions and concerns, and is implemented in the following manner.

[0848] First, users can send medical questions in the form of direct messages through a dedicated app or website, and the user's device (smartphone or PC) then sends the question to the server.

[0849] Next, the server receives an inquiry message from the user. The received message is stored in a database and then analyzed by a natural language processing (NLP) engine. During the analysis, the intent of the inquiry and important keywords are extracted. For example, if the question is, "I get short of breath at night. What should I do?", the keywords "night," "short of breath," and "measures" are analyzed.

[0850] The analyzed query content is sent to a generative AI (such as GPT-4), which generates an appropriate answer based on the data sent. The answers generated here are based on medical training data and are designed to provide accurate and reliable information. For example, a generated answer might read, "Possible causes of shortness of breath include allergies, asthma, and heart problems. We recommend that you first consult a doctor. We also recommend that you try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow."

[0851] Instead of providing the answer directly to the user, the generated answer is first sent to an expert, who reviews the content of the generated answer and reviews it for accuracy and appropriateness. After this review process is complete, the server formats the answer in an easy-to-understand format before sending it to the user. The final answer is then sent to the user's device as a direct message.

[0852] Users can receive the final answer on their device and obtain accurate medical advice for their questions. This allows users to consult with a doctor at any time and quickly obtain reliable information supervised by an expert.

[0853] For example, if a user sends a specific question such as, "Recently, I've been having trouble breathing before going to bed. Is there anything I can do about this?", the server receives it and analyzes it using a natural language processing engine. A generative AI system then generates an appropriate answer, which is then reviewed by a specialist and finally sent to the user. This system allows users to receive accurate and prompt medical advice.

[0854] This system contains training data to generate appropriate answers to specific medical questions, reducing the burden on medical experts while maintaining the accuracy of the answers. It is also available 24 hours a day, allowing users to receive medical consultations at their convenience. This is expected to reduce anxiety for patients and their families and improve the quality and efficiency of medical care.

[0855] The processing flow will be explained below.

[0856] Step 1:

[0857] Users enter medical questions from their device (smartphone or PC) through a dedicated app or website and send them as direct messages.

[0858] Step 2:

[0859] The server receives the query message sent by the user and stores it in a database.

[0860] Step 3:

[0861] The server sends the received message to a natural language processing (NLP) engine, which analyzes the message content, performing tokenization (dividing sentences into words), POS tagging (identifying parts of speech), and named entity recognition (recognizing specific names, places, diseases, etc.).

[0862] Step 4:

[0863] The server normalizes the query content analyzed by the NLP engine and sends it to the generative AI (e.g., GPT-4). The normalization process converts the query content into a format that is easy for the generative AI to understand.

[0864] Step 5:

[0865] Generative AI generates appropriate answers based on the data received, and these answers are internally verified to ensure they are accurate and reliable.

[0866] Step 6:

[0867] The server sends the generated answer to an expert, who reviews the answer and supervises its accuracy and appropriateness. Corrections and supplements are made as necessary.

[0868] Step 7:

[0869] The final answer, edited by experts, is sent to the server, which formats it in a user-friendly format.

[0870] Step 8:

[0871] The server sends the final formatted response to the user as a direct message.

[0872] Step 9:

[0873] The user receives the final answer on the device and obtains the appropriate medical advice for their question. The user checks the information displayed on the device and decides on the next step if necessary.

[0874] Example 1

[0875] 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."

[0876] Conventional medical consultation systems have limitations in terms of providing users with fast and accurate answers. Medical experts often have limited time and resources to respond directly, resulting in variations in the quality and speed of responses. Furthermore, if users want to seek medical advice at their own convenience, a system that can provide fast responses 24 hours a day is required, but such systems are not widely available at present. The objective of this invention is to provide a system that allows users to receive fast and reliable medical advice at any time.

[0877] 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.

[0878] In this invention, the server includes means for receiving a medical inquiry from a user, means for analyzing the received inquiry content using natural language processing, means for transmitting the analyzed inquiry content to a generative AI, means for transmitting an answer generated by the generative AI to an expert for review, means for transmitting the reviewed answer to the user, means for transmitting a formatted final answer to the user, and means for the user to receive the final answer and obtain appropriate medical advice. This enables users to receive prompt and reliable medical advice 24 hours a day.

[0879] The "means for receiving medical inquiries from users" refers to the means by which the server receives medical questions or doubts sent by users through the dedicated application or website.

[0880] "Means for analyzing the content of received inquiries using natural language processing" refers to a means for analyzing medical inquiries received by the server using a natural language processing engine and extracting important keywords and intent.

[0881] The "means for transmitting the analyzed query content to the generative artificial intelligence" is a means for transmitting the query content analyzed by natural language processing to the generative artificial intelligence as an appropriate prompt.

[0882] "Means for sending answers generated by generative AI to experts for review" refers to means for sending answers generated by generative AI to medical experts, who then review and confirm the content.

[0883] The "means for transmitting the answer that has been supervised to the user" is a means for transmitting the answer that has been supervised by the expert to the user's terminal.

[0884] The "means for transmitting a formatted final answer to a user" refers to a means for transmitting a final answer to a user that has been formatted in a format that is easy to understand after being supervised by an expert.

[0885] The "means for the user to receive the final answer and obtain appropriate medical advice" refers to the means by which the user's terminal receives the final answer sent from the server and the user can obtain that medical advice.

[0886] The present invention is a system that allows users to quickly and accurately resolve medical questions and concerns. It is expected that the system will be implemented in accordance with the following detailed description.

[0887] First, users can use a dedicated application or website to send medical questions in the form of direct messages, with the user's device (such as a smartphone or personal computer) sending the question to a server via the Internet.

[0888] Specifically, the user enters a question into a text input form on a dedicated app or website and presses the send button. For example, suppose the question entered is, "I feel short of breath before going to bed. What should I do?" The device then sends this question to the server.

[0889] Next, the server receives the inquiry message sent by the user. The received message is saved in a database. This database can be built using MySQL or PostgreSQL, for example. After saving, the server analyzes the message using a natural language processing engine (for example, Google Natural Language API). As a result of the analysis, the intent of the question and important keywords are extracted. For example, the keywords extracted are "before going to bed," "shortness of breath," and "measures."

[0890] The parsed query content is sent from the server to a generative AI (e.g., OpenAI's GPT-4), and the prompt text is also sent to the generative AI.

[0891] Example prompt sentence:

[0892] "Healthcare question: 'I'm having trouble breathing before bed. What should I do?' Generate an answer."

[0893] Generative AI generates appropriate answers based on the submitted data and prompts. The generated answers are based on medical training data and are designed to provide accurate and reliable information. For example, a generated answer might read, "Possible causes of shortness of breath include allergies, asthma, or heart problems. We recommend consulting a doctor first. Also, try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow."

[0894] The generated answer is sent to an expert for review. The expert reviews the generated answer and checks its accuracy and appropriateness. After the review process is complete, the server formats the final answer into an easy-to-understand format before sending it to the user. For example, it may use HTML formatting or highlight important keywords. The final formatted answer is then sent to the user's device as a direct message.

[0895] Finally, the user receives the final answer on their device and obtains appropriate medical advice for their question. This allows users to receive prompt and accurate medical consultations 24 hours a day. The system uses training data to generate appropriate answers to specific medical questions, reducing the burden on experts while maintaining the accuracy of the answers. This is expected to improve the quality and efficiency of medical care.

[0896] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0897] Step 1:

[0898] The user uses a dedicated application or website to input and submit a medical question. The input data here is the user's medical question. The device sends this input data to the server. Specifically, the user inputs a question such as, "I have shortness of breath before going to bed. What should I do?" and clicks the submit button.

[0899] Step 2:

[0900] The server receives the message sent by the user. The received data is the user's question, and to store it in the database, it executes an SQL query, for example, "INSERT INTO inquiries (user_id, question, timestamp) VALUES (?, ?, ?)". The input is the message data from the user, and the output is the record stored in the database.

[0901] Step 3:

[0902] The server sends the stored message to a natural language processing engine for analysis. The input data here is the user's question retrieved from the database. The natural language processing engine (for example, Google Natural Language API) analyzes the inquiry and extracts important keywords and intent. Specifically, it extracts keywords such as "before going to bed," "shortness of breath," and "measures." The output is the analysis results.

[0903] Step 4:

[0904] The server sends the analyzed query content to the generative AI. The input data consists of keywords from the analyzed query content and the corresponding prompt. For example, a prompt such as "This is a medical question. Please generate an answer to the question, 'I feel short of breath before going to bed. What should I do?'" is sent to the generative AI. The output is the generated answer.

[0905] Step 5:

[0906] The server receives the answer generated by the generative AI and then sends it to the expert. The input data is the answer received from the generative AI. The server sends this answer to the expert, who reviews it. The output is feedback from the expert and a revised answer.

[0907] Step 6:

[0908] The server receives the expert-edited answer and formats it. The input data is the expert-edited answer, which is then formatted into an easy-to-understand format, for example, using HTML to highlight important keywords. The output is the formatted answer.

[0909] Step 7:

[0910] The server sends a formatted final answer to the user. The input data is the formatted final answer, and the output is a message sent to the user's terminal. The user receives this final answer on their terminal and obtains the required medical advice.

[0911] In this way, each processing step works in tandem, allowing users to receive prompt and accurate medical advice 24 hours a day.

[0912] (Application example 1)

[0913] 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."

[0914] In modern society, users need to be able to quickly and accurately resolve their medical questions and concerns. However, current systems require users to directly contact medical experts, which often requires time and effort. In addition, general online information can be unreliable, and there is a lack of mechanisms for providing users with accurate medical advice. Furthermore, the lack of an efficient system that utilizes mobile devices such as smartphones makes it difficult to meet the needs of today's busy people.

[0915] 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.

[0916] In this invention, the server includes means for receiving medical inquiries from users, means for analyzing the received inquiry content using natural language processing, means for sending the analyzed inquiry content to a generative artificial intelligence, means for sending an answer generated by the generative artificial intelligence to an expert for supervision, means for sending the supervised answer to the user, and means for receiving inquiries from users using a smartphone application. This enables users to receive fast and reliable medical advice via their smartphones, and the accuracy of the answers is guaranteed through expert supervision, allowing users to alleviate their concerns and take appropriate measures quickly.

[0917] A "user" is an individual who has a medical question or concern.

[0918] An "inquiry" is a question sent by a user to resolve a medical question or doubt.

[0919] The "receiving means" is a system or device for electronically receiving an inquiry sent by a user.

[0920] "Natural language processing" is a technology that allows computers to understand and analyze human language.

[0921] "Generative artificial intelligence" is an AI system that automatically generates appropriate medical advice based on the content of the inquiry it receives.

[0922] An "expert" is an individual or team with a high level of expertise and experience in the medical field.

[0923] "Supervision" refers to the methods and processes by which experts verify the accuracy and appropriateness of the answers provided by the generative AI.

[0924] A "smartphone application" is a program that runs on a smartphone and allows users to send medical inquiries.

[0925] "Formatting methods" are techniques or processes used to prepare curated responses in a format that is easy for users to understand.

[0926] "Server" refers to a computer system that manages and executes a series of processes, such as receiving inquiries, analyzing them, generating responses, sending them to experts, and reviewing them.

[0927] "Learning data" refers to training data that is used in advance by generative artificial intelligence to generate appropriate medical advice.

[0928] MODE FOR CARRYING OUT THE INVENTION

[0929] The present invention is a system that allows users to quickly and accurately resolve medical questions and concerns. Each part of the system operates according to the following procedure.

[0930] 1. Program Generation

[0931] The program of the system for realizing the present invention mainly consists of the following components.

[0932] Receiving means: An interface (such as a smartphone application) for receiving medical inquiries from users.

[0933] Natural Language Processing: An NLP engine (e.g., SomeNLPModel) for analyzing query content and extracting important keywords.

[0934] Generative artificial intelligence: An AI system that generates medical advice based on control messages (e.g., GPT-4).

[0935] Editing tool: An interface for experts to review, correct, and approve answers generated by the AI.

[0936] Formatting: The process of sending expert-approved answers in a format that is easy for users to understand.

[0937] 2. Explain the program's processing

[0938] The program's processing begins when a query is sent from the user's smartphone application to the server. The server receives this query and analyzes its contents using a natural language processing (NLP) engine. During the analysis, important keywords are extracted. An NLP engine called SomeNLPModel is used in this step.

[0939] The extracted keywords are then sent to a generative artificial intelligence (GPT-4) to generate appropriate medical advice. The generated answers are then sent to experts, who review them for accuracy and appropriateness. This process is carried out using an expert review interface.

[0940] Once the expert-edited answers are ready for the user, they are formatted. This step ensures that the answers are easy for the user to understand. The final answers are then sent to the user via a smartphone application.

[0941] 3. Add specific examples

[0942] For example, suppose a user sends a question via a smartphone application: "Recently, I've been having trouble breathing before going to bed. Is there anything I can do about this?" In this case, the server receives the question and analyzes it using an NLP engine (SomeNLPModel). This analysis extracts keywords such as "night," "shortness of breath," and "measures." A generative artificial intelligence (GPT-4) generates appropriate medical advice based on these keywords. The generated answer is sent to an expert for review. After review, the answer is formatted before being provided to the user and sent to the user in the following format:

[0943] Example prompt sentence:

[0944] Q: Lately I've been having trouble breathing before going to bed. Is there anything I can do about it?

[0945] Keywords: night, shortness of breath, measures

[0946] Generate medical advice based on this keyword.

[0947] In this way, users can receive fast and accurate medical advice.

[0948] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0949] Step 1:

[0950] A user submits a medical inquiry through a smartphone application. The input is the user's question, and the output is the action of sending this question to the server. For example, if a user enters the question, "Recently, I've been having trouble breathing before going to bed. Is there anything I can do about it?", the question is sent to the server.

[0951] Step 2:

[0952] The server receives inquiries from users. The input is the user's question, and the output is the data of the received inquiry. The server receives and stores the data sent from the smartphone app. Specifically, the inquiry is saved in a database.

[0953] Step 3:

[0954] The query received by the server is analyzed using a natural language processing (NLP) engine. The input is the query data, and the output is the extracted keywords. For example, keywords such as "night," "shortness of breath," and "measures" are extracted. Specifically, SomeNLPModel, an NLP engine, analyzes the query and extracts important keywords.

[0955] Step 4:

[0956] The server sends the keywords from the analyzed query to a generative AI (GPT-4) system, which generates appropriate medical advice. The input is the extracted keywords, and the output is the generated medical advice. For example, the generated advice might read, "Possible causes of shortness of breath include allergies, asthma, and heart problems. We recommend that you first consult a doctor. Also, try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow." Specifically, the generative AI system outputs appropriate advice based on the prompt text.

[0957] Step 5:

[0958] The generated answer is sent to an expert for review. The input is the generated medical advice, and the output is an answer that has been reviewed and approved by an expert. Specifically, the answer output by the generative AI is checked by an expert review system, and any corrections that need to be made are made, and finally, it is approved.

[0959] Step 6:

[0960] The server formats the expert-edited answers for delivery to the user. The input is the edited answers, and the output is the final formatted answers. Specifically, the answers are formatted in a user-friendly format and delivered via a smartphone app.

[0961] Step 7:

[0962] The server sends the final answer to the user. The input is the formatted final answer, and the output is the answer that is displayed on the user's smartphone. Specifically, the server sends the formatted answer to a smartphone app, and the user views the answer.

[0963] This allows users to receive prompt and accurate medical advice.

[0964] 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.

[0965] This invention is a system that allows users to quickly resolve medical-related questions and concerns, and by combining it with an emotion engine that recognizes the user's emotions, it provides more appropriate responses. It is implemented in the following steps.

[0966] First, a user sends a medical question via a dedicated app or website in the form of a direct message. The user's device (smartphone or PC) then sends the question to the emotion engine, which analyzes the user's emotions. The emotion engine then uses the text data to analyze the user's emotional state (e.g., anxiety, anger, relief, etc.).

[0967] Next, after analyzing the user's emotions, the server receives the question message sent by the user. The received message is stored in a database, and then the inquiry is analyzed by a natural language processing (NLP) engine. During the analysis, the intent of the inquiry and important keywords are extracted. For example, if the question is "I get short of breath at night. What should I do?", the keywords "night," "short of breath," and "measures" are analyzed.

[0968] Along with the analyzed query content, the user's emotional information analyzed by the emotion engine is also sent to a generative AI (such as GPT-4). The generative AI generates an appropriate answer taking into account the user's emotions. For example, if the emotion is recognized as "anxiety," the answer will include a reassuring expression. In this way, the generated answer is based on medical training data and provides accurate and reliable information.

[0969] Instead of providing the answer directly to the user, the generated answer is first sent to an expert, who reviews the content of the generated answer and reviews it for accuracy and appropriateness. After this review process is complete, the server formats the answer in an easy-to-understand format before sending it to the user. The final answer is then sent to the user's device as a direct message.

[0970] Users can receive the final answer on their device and obtain accurate medical advice for their questions. This allows users to receive medical consultations 24 hours a day and quickly obtain reliable information supervised by experts.

[0971] For example, if a user sends a specific query such as, "I've been having trouble breathing before bed recently. I'm feeling very anxious. What should I do?", the emotion engine will interpret it as "anxiety." This information and the query are received and analyzed by the server and sent to the generative AI. The generative AI then generates a response such as, "Possible causes of your shortness of breath include allergies, asthma, and heart problems. We recommend that you first consult a doctor. We also recommend that you try improving your bedroom environment. For example, use an air purifier to remove dust and allergens and use a high pillow. We understand your anxiety, but please try these measures." A specialist will then review this response and confirm its appropriateness before sending it to the user. In this way, the user can receive an emotionally sensitive response, which can help alleviate their anxiety.

[0972] This system uses an emotion engine to recognize the user's emotions and provide that information to a generative AI system, which can then generate appropriate answers based on the user's emotions. Furthermore, after being supervised by experts, the system provides users with highly reliable information, greatly improving the quality and efficiency of medical consultations.

[0973] The processing flow will be explained below.

[0974] Step 1:

[0975] Users enter medical questions from their device (smartphone or PC) through a dedicated app or website and send them as direct messages.

[0976] Step 2:

[0977] The server receives the query message sent by the user and stores it in a database.

[0978] Step 3:

[0979] The server sends the message to the emotion engine, which analyzes the user's emotion. The emotion engine analyzes the text data to recognize the user's emotional state (e.g., anxiety, anger, relief, etc.) and stores this information for further processing.

[0980] Step 4:

[0981] The server sends the received message to a natural language processing (NLP) engine, which analyzes the message content, performing tokenization (dividing sentences into words), POS tagging (identifying parts of speech), and named entity recognition (recognizing specific names, places, diseases, etc.).

[0982] Step 5:

[0983] The server normalizes the query content analyzed by the NLP engine and the user's emotional information recognized by the emotion engine, and sends them to a generative AI (e.g., GPT-4). The normalization process converts the data into a format that is easy for the generative AI to understand.

[0984] Step 6:

[0985] Generative AI generates appropriate answers based on the normalized data. The system also takes into account the user's emotional information, so the generated answers include language tones and expressions that correspond to the emotion. For example, if the user's emotion is recognized as "anxiety," the system adds expressions that convey a sense of security to the answer.

[0986] Step 7:

[0987] The server sends the generated answer to the expert, who reviews the answer and supervises its accuracy and appropriateness, correcting or supplementing it as necessary.

[0988] Step 8:

[0989] The final answer, edited by an expert, is sent to the server, which formats it for presentation to the user. The formatted answer is formatted in a way that is easy for the user to understand.

[0990] Step 9:

[0991] The server sends the final formatted response to the user as a direct message.

[0992] Step 10:

[0993] The user receives the final answer on the device and obtains the medical advice relevant to their question. The user reviews the information displayed on the device and decides on the next step if necessary, such as making further inquiries or acting on the advice.

[0994] Example 2

[0995] 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."

[0996] In modern medical consultations, it is necessary to quickly and appropriately resolve users' anxieties and doubts. However, conventional systems generate mechanical answers without considering the user's feelings, which can lead to low user satisfaction. Furthermore, to ensure the reliability of the generated answers, expert supervision is essential, but there are only a limited number of systems that can do this efficiently.

[0997] 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.

[0998] In this invention, the server includes means for receiving a medical inquiry from a user, means for analyzing the received inquiry content using natural language processing, means for analyzing the user's emotions, means for transmitting the analyzed inquiry content and emotional information to the generative AI, means for transmitting the answer generated by the generative AI to an expert for review, and means for transmitting the answer after review to the user. This makes it possible to generate an appropriate medical answer that takes the user's emotions into consideration, and provide it to the user with a high level of reliability after it has been checked by the expert.

[0999] A "user" is someone who uses the system to resolve medical questions or concerns.

[1000] "Inquiry" refers to a medical question or consultation sent by a user.

[1001] "Natural language processing" is a technical method for analyzing the content of received inquiries and extracting intent and keywords.

[1002] "Sentiment analysis" is the process of identifying an emotional state from the text data contained in a user's query.

[1003] "Generative AI" refers to AI that generates appropriate medical answers based on the content of the inquiry and emotional information it receives.

[1004] An "expert" is someone who is qualified to oversee the content of medical answers created by generative AI and verify their accuracy and appropriateness.

[1005] "Supervision" is the act of an expert reviewing the content of answers generated by generative AI and ensuring their appropriateness and accuracy.

[1006] "Formatting" is the process of shaping the answers provided to the user into an understandable form.

[1007] "Training data" refers to the training data used by generative artificial intelligence to generate appropriate answers to medical-related questions.

[1008] The present invention is a system that enables users to receive prompt and appropriate advice regarding medical questions and concerns. By incorporating an emotion engine that analyzes the user's emotions, the system provides appropriate responses that take emotions into consideration.

[1009] First, a user sends a medical question in the form of a direct message through a dedicated application or website. For example, a user might use their smartphone to send a message along the lines of, "Lately, I've been having trouble breathing before going to bed. It's really worrying. What should I do?"

[1010] The user's device sends the message to an emotion engine. A commonly used emotion analysis tool can be used as the emotion engine. As a specific example, IBM Watson's Tone Analyzer can be used. This emotion engine analyzes the user's emotional state (anxiety, anger, relief, etc.) based on the text data and assigns an emotion label. In this specific example, the emotion label "anxiety" is obtained from the part "I am very anxious."

[1011] Next, the server receives the message sent by the user and stores it in a MySQL database. The stored data includes the user ID, message content, emotion label, etc. The stored message is then analyzed by a natural language processing (NLP) engine. NLP engines such as SpaCy and NLTK can be used as examples. During this analysis process, the intent of the inquiry and important keywords are extracted. For example, from the message "Recently, I've been having trouble breathing before going to bed," keywords such as "before going to bed," "trouble breathing," and "measures" are extracted.

[1012] The analyzed inquiry content and emotional information are sent to a generative AI. A large language model such as GPT-4 is used as the generative AI. At this time, the following message is input as a prompt: "The user has recently been experiencing difficulty breathing before going to bed, which has caused them great anxiety. Please provide appropriate medical advice."

[1013] The generative AI model generates an appropriate answer based on the prompt and emotional information. For example, it might say, "Possible causes of shortness of breath include allergies, asthma, or heart problems. We recommend that you first consult a doctor. Also, try improving your bedroom environment. For example, use an air purifier to remove dust and allergens, and use a high pillow. We understand that you may be feeling anxious, but please try these measures."

[1014] The generated answers are sent to experts, who review the answers for accuracy and appropriateness and make any necessary corrections. Once the review is complete, the server formats the answers in a way that is easy for users to understand, such as by organizing them into paragraphs or bullet points.

[1015] The final answer will be sent to the user's device as a direct message, allowing them to receive the final answer on their smartphone or PC and receive reliable medical advice for their question.

[1016] In this way, the present invention makes it possible to quickly provide appropriate medical answers that take into account the user's feelings.

[1017] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1018] Step 1:

[1019] A user submits a medical question through a dedicated application or website.

[1020] Input: User's question (e.g., "Recently, I've been having trouble breathing before going to bed. It's really worrying me. What should I do?")

[1021] Output: A query message is sent from the user's terminal to the server.

[1022] Step 2:

[1023] The user's terminal transmits the received question message to the emotion engine.

[1024] Input: User's question message

[1025] Data processing: The emotion engine analyzes the text data and assigns an emotion label (e.g., "anxiety").

[1026] Output: Question message with emotion label (e.g., "Anxious" from "I'm very anxious")

[1027] Step 3:

[1028] The server receives the query message sent from the user's terminal and stores it in a database.

[1029] Input: A question message with an emotion label.

[1030] Data processing: Received messages are stored in a MySQL database. The stored data includes user ID, message content, emotion label, etc.

[1031] Output: The question message stored in the database

[1032] Step 4:

[1033] The server sends the query message stored in the database to a natural language processing (NLP) engine, which analyzes the query.

[1034] Input: Question message stored in the database

[1035] Data calculation: Text analysis is performed by an NLP engine (e.g., SpaCy) to extract intent and important keywords (e.g., "before going to bed," "difficulty breathing," "measures").

[1036] Output: Parsed query content

[1037] Step 5:

[1038] The server sends the analyzed inquiry content and emotional information to the generative artificial intelligence.

[1039] Input: Parsed query content and sentiment label

[1040] Data processing: Generate prompt sentences (e.g., "The user has recently been experiencing difficulty breathing before going to bed, which has caused him great anxiety. Please provide appropriate medical advice.")

[1041] Output: The prompt sent to the generative AI model

[1042] Step 6:

[1043] Generative AI generates appropriate answers based on prompts and emotional information.

[1044] Input: A prompt sent to the generative AI model

[1045] Data computation: A generative AI model (e.g., GPT-4) generates an appropriate answer based on a prompt (e.g., "Possible causes of shortness of breath include allergies, asthma, heart problems, etc...").

[1046] Output: The generated answer

[1047] Step 7:

[1048] The server sends the generated answers to experts, who then review the content.

[1049] Input: Generated Answer

[1050] Data processing: Experts check the accuracy and appropriateness of the answers and make corrections if necessary (expert feedback)

[1051] Output: Edited answer

[1052] Step 8:

[1053] The server formats the edited answers and converts them into a user-friendly format.

[1054] Input: Edited answer

[1055] Data processing: Formatting responses, such as organizing them into paragraphs or bullet points

[1056] Output: Formatted final answer

[1057] Step 9:

[1058] The server sends the final response to the user's device as a direct message.

[1059] Input: Final formatted answer

[1060] Output: Final answer EX sent to the user's device (received by the user on their smartphone or PC)

[1061] (Application example 2)

[1062] 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."

[1063] Conventional systems have difficulty providing responses that take users' emotions into consideration, and often do not provide adequate support, especially to users who feel anxious. Furthermore, the generated answers require time-consuming supervision by experts, making it difficult to provide immediate responses. This has led to issues that make it difficult for users to obtain fast and reliable information.

[1064] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the received inquiry content using natural language processing, means for sending the analyzed inquiry content to the generative artificial intelligence, means for sending an answer generated by the generative artificial intelligence to an expert for supervision, means for sending the answer after supervision to the user, emotion analysis means for recognizing the user's emotional state, and means for providing the generative artificial intelligence with analysis results including emotion-analyzed information. This makes it possible to quickly provide a response that takes the user's emotions into consideration, and to provide highly reliable information that has been supervised by an expert.

[1065] "User inquiries" refer to questions or doubts that users send to the system.

[1066] "Natural language processing" is a technology that allows computers to understand and analyze human language.

[1067] "Generative AI" refers to AI technology that automatically generates appropriate answers based on input data.

[1068] An "expert" is someone who has advanced knowledge and experience in a particular field.

[1069] "Supervision" is the process in which experts check the accuracy and appropriateness of the generated answers.

[1070] "Emotion analysis means" is a technology that analyzes a user's text data and identifies the user's emotional state.

[1071] "Formatting means" refers to a technique that provides expert-edited answers in a format that is easy for users to understand.

[1072] "Learning data" is training data that generative artificial intelligence uses to generate appropriate answers.

[1073] A system for implementing the present invention has the following configuration.

[1074] 1. The server first receives an inquiry from the user. The user enters security questions or concerns through a dedicated application or website. For example, a user could use a device such as a smartphone or PC to send an inquiry such as, "I've been feeling uneasy about online banking lately. What can I do to make it safer?"

[1075] 2. When a query arrives at the server, the server analyzes it using a natural language processing (NLP) engine to extract important keywords and intent from the query. Libraries such as TextBlob and SentimentAnalysis are used here.

[1076] 3. Next, the server uses emotion analysis to recognize the user's emotions. Tools such as EmotionEngine provide technology that analyzes the user's text data and identifies their emotional state (e.g., anxiety, relief, etc.).

[1077] 4. The results of the sentiment analysis and the analysis results from the NLP engine are sent to a generative AI model (such as GPT-4). The generative AI model uses this data to generate an appropriate response that takes the user's emotions into consideration.

[1078] 5. The generated answers are sent to experts, who review them for accuracy and appropriateness. For example, a security expert might review an answer about how to use online banking safely.

[1079] 6. After the expert has verified the answer, the server formats the final answer for delivery to the user. This formatted answer is provided in a format that is easy for the user to understand.

[1080] 7. Finally, the server sends the final formatted response to the user's terminal.

[1081] As a specific example, if a user sends a query such as, "I've been feeling uneasy about online banking lately. What can I do to make it safer?", the emotion engine will interpret it as "uneasy." This information and the content of the query are received and analyzed by the server, and then sent to the generative AI. The generative AI will then generate a response such as, "To increase the security of online banking, it is important to use a strong password and set up two-step authentication. Also, be careful not to click on suspicious links." This is then reviewed by security experts, and the final response is sent to the user.

[1082] Examples of prompts used by generative AI models include:

[1083] "Users are worried and have asked about how to use online banking safely. Please reassure them and provide helpful advice."

[1084] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1085] Step 1:

[1086] A user uses a device such as a smartphone or PC to send a security inquiry to a server through a dedicated application or website. The input is the user's inquiry text, and the output is the inquiry data sent to the server. Specifically, the text sent might be something like, "I've been feeling uneasy about online banking lately. What can I do to make it safer?"

[1087] Step 2:

[1088] The server analyzes the received inquiry using a natural language processing (NLP) engine. The input is the inquiry text, and the output is the analyzed keywords and intent. Specifically, TextBlob and SentimentAnalysis are used to extract important keywords such as "online banking," "anxiety," and "safety."

[1089] Step 3:

[1090] The server recognizes the user's emotional state using an emotion analysis means. The input is the user's query text, and the output is the recognized emotional information. Specifically, the server uses the Emotion Engine to analyze the user's emotion (e.g., anxiety) and obtains it as data.

[1091] Step 4:

[1092] The server sends the analyzed keywords and emotional information to a generative artificial intelligence (generative AI model). The input is the analyzed keywords and emotional information, and the output is a generated answer. Specifically, it uses the OpenAI API (e.g., GPT-4) to generate an answer such as, "To increase the security of your online banking, it is important to use a strong password and set up two-factor authentication. Also, be careful not to click on suspicious links."

[1093] Step 5:

[1094] The generated answer is sent from the server to an expert for review. The input is the generated answer text, and the output is the answer that has been checked and corrected by the expert. Specifically, the security expert checks the accuracy and appropriateness of the answer content and makes corrections as necessary.

[1095] Step 6:

[1096] The server formats the expert-edited answers for delivery to the user. The input is the edited answer text, and the output is the formatted answer. Specifically, the server converts the answer content into a format that is easy for the user to understand.

[1097] Step 7:

[1098] Finally, the server sends the formatted answer to the user's device. The input is the formatted answer text, and the output is the answer received on the user's device. Specifically, the intuitive and easy-to-read answer is displayed on the user's smartphone or PC.

[1099] Through the above processing steps, users can receive prompt and reliable security advice that takes their emotions into consideration.

[1100] 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.

[1101] 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.

[1102] 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.

[1103] 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.

[1104] FIG. 9 is a diagram illustrating 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 actions 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.

[1105] 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.

[1106] 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).

[1107] 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.

[1108] 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."

[1109] 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.

[1110] 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).

[1111] 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.

[1112] 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.

[1113] 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.

[1114] 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.

[1115] 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.

[1116] 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.

[1117] 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.

[1118] 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.

[1119] 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.

[1120] 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.

[1121] The following is further disclosed regarding the above embodiment.

[1122] (Claim 1)

[1123] means for receiving a medical inquiry from a user;

[1124] A means for analyzing the content of the received inquiry by natural language processing;

[1125] means for transmitting the analyzed query content to a generative artificial intelligence;

[1126] A means to send the answers generated by the generative AI to experts for review, and

[1127] means for transmitting the edited answers to the user;

[1128] A system including:

[1129] (Claim 2)

[1130] 10. The system of claim 1, further comprising means for formatting the expert-curated answer content for presentation to the user.

[1131] (Claim 3)

[1132] 10. The system of claim 1, wherein the generative artificial intelligence includes training data for generating appropriate answers to specific medical-related questions.

[1133] "Example 1"

[1134] (Claim 1)

[1135] means for receiving a medical inquiry from a user;

[1136] A means for analyzing the content of the received inquiry by natural language processing;

[1137] means for transmitting the analyzed query content to a generative artificial intelligence;

[1138] A means to send the answers generated by the generative AI to experts for review, and

[1139] means for transmitting the edited answers to the user;

[1140] means for transmitting the final formatted answer to the user;

[1141] a means for the user to receive a final answer and obtain appropriate medical advice;

[1142] A system including:

[1143] (Claim 2)

[1144] 10. The system of claim 1, further comprising means for formatting the expert-curated answer content for presentation to the user.

[1145] (Claim 3)

[1146] 10. The system of claim 1, wherein the generative artificial intelligence includes training data for generating appropriate answers to specific medical-related questions.

[1147] "Application Example 1"

[1148] (Claim 1)

[1149] means for receiving a medical inquiry from a user;

[1150] A means for analyzing the content of the received inquiry by natural language processing;

[1151] means for transmitting the analyzed query content to a generative artificial intelligence;

[1152] A means to send the answers generated by the generative AI to experts for review, and

[1153] means for transmitting the edited answers to the user;

[1154] A means for receiving inquiries from users using a smartphone application;

[1155] A system including:

[1156] (Claim 2)

[1157] 10. The system of claim 1, further comprising means for formatting the expert-curated answer content for presentation to the user.

[1158] (Claim 3)

[1159] 10. The system of claim 1, wherein the generative artificial intelligence includes training data for generating appropriate answers to specific medical-related questions.

[1160] "Example 2: Combining Emotion Engines"

[1161] (Claim 1)

[1162] means for receiving a medical inquiry from a user;

[1163] A means for analyzing the content of the received inquiry by natural language processing;

[1164] means for analyzing user emotions;

[1165] means for transmitting the analyzed inquiry content and emotion information to the generative artificial intelligence;

[1166] A means to send the answers generated by the generative AI to experts for review, and

[1167] means for transmitting the edited answers to the user;

[1168] A system including:

[1169] (Claim 2)

[1170] 10. The system of claim 1, further comprising means for formatting the expert-curated answer content for presentation to the user.

[1171] (Claim 3)

[1172] 10. The system of claim 1, wherein the generative artificial intelligence includes training data for generating appropriate answers to specific medical-related questions.

[1173] "Application example 2 when combining emotion engines"

[1174] (Claim 1)

[1175] means for receiving a query from a user;

[1176] A means for analyzing the content of the received inquiry by natural language processing;

[1177] means for transmitting the analyzed query content to a generative artificial intelligence;

[1178] A means to send the answers generated by the generative AI to experts for review, and

[1179] means for transmitting the edited answers to the user;

[1180] emotion analysis means for recognizing the emotional state of a user;

[1181] A means for providing an analysis result including emotion-analyzed information to a generative artificial intelligence;

[1182] A system including:

[1183] (Claim 2)

[1184] 10. The system of claim 1, further comprising means for formatting the expert-curated answer content for presentation to the user.

[1185] (Claim 3)

[1186] 10. The system of claim 1, wherein the generative artificial intelligence includes training data for generating appropriate answers to specific questions. [Explanation of symbols]

[1187] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving a medical inquiry from a user; A means for analyzing the content of the received inquiry by natural language processing; means for transmitting the analyzed query content to a generative artificial intelligence; A means to send the answers generated by the generative AI to experts for review, and means for transmitting the edited answers to the user; A system including:

2. 10. The system of claim 1, further comprising means for formatting the expert-curated response content for presentation to the user.

3. The system of claim 1 , wherein the generative artificial intelligence includes training data for generating appropriate answers to specific medical-related questions.

Images