System

Abstract

To provide a system.SOLUTION: A system includes: first means which receives, from a user, input of a prompt sentence for generating disease condition data of a virtual patient with a specific genetic disease; second means for generating the disease condition data of the virtual patient by inputting the prompt sentence to a generative AI model; and third means for displaying the generated disease condition data of the virtual patient on a terminal of 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] In order for medical professionals to improve their medical history skills, it is important to gain experience interviewing many patients. However, interviewing patients takes time, and in rare cases, the number of patients is limited, making it difficult to gain experience. [Means for solving the problem]

[0005] We provide a system that utilizes generative AI to conduct medical interview training with virtual patients. This system generates virtual patients with various symptoms and medical histories, and evaluates the interview results. Based on the evaluation results, it provides feedback indicating areas for improvement in the interview results and ways to improve interview skills. [Brief explanation of the drawings]

[0006] [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. 2 is a sequence diagram showing a flow of processing in the data processing system according to the first embodiment of the first form example. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1. [Figure 13] FIG. 10 is a sequence diagram showing a processing flow of a data processing system in a second embodiment of the second form example. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Embodiment Example 2. [Figure 15] FIG. 10 is a sequence diagram showing the flow of processing in a data processing system according to a third embodiment of the third embodiment. [Figure 16]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Embodiment 3. [Figure 17] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when an emotion engine is combined. [Figure 19] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the second embodiment of the second form example when an emotion engine is combined. [Figure 20] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when an emotion engine is combined. [Figure 21] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the third embodiment of the third form example when an emotion engine is combined. [Figure 22] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0007] 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.

[0008] First, the terms used in the following description will be explained.

[0009] 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, the 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), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] 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.

[0011] 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.

[0012] 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.

[0013] 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."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 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.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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.

[0022] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] 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.

[0024] 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.

[0025] 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.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] One aspect of the present invention is a virtual patient interview training system that utilizes generative AI. Specifically, the generative AI generates virtual patients with various symptoms and medical histories. A medical professional interviews these virtual patients, and the system evaluates the results. Based on the evaluation results, the system provides feedback to the medical professional.

[0029] "Example 2"

[0030] As a concrete example, generative AI generates a virtual patient with diabetes symptoms. A healthcare professional interviews the virtual patient, and the system evaluates the results. Based on the evaluation results, the system provides feedback to the healthcare professional on how to improve their interview and how to improve their diabetes interview skills.

[0031] "Example 3"

[0032] Generative AI can also generate virtual patients with rare conditions, such as specific genetic diseases, allowing patients to gain experience in medical interviews with conditions that would be difficult to experience in real life.

[0033] The processing flow of each embodiment will be described below.

[0034] "Example 1"

[0035] Step 1: Generative AI generates virtual patients with various symptoms and medical histories.

[0036] Step 2: A medical professional interviews the generated virtual patient.

[0037] Step 3: The system evaluates the interview results and provides feedback to the healthcare professional based on the evaluation results.

[0038] "Example 2"

[0039] Step 1: The generative AI generates a virtual patient with diabetes symptoms.

[0040] Step 2: A medical professional interviews the generated virtual patient.

[0041] Step 3: The system evaluates the interview results and provides feedback to healthcare professionals based on the evaluation results, indicating areas for improvement in the interview and how to improve their diabetes interview skills.

[0042] "Example 3"

[0043] Step 1: Generative AI generates virtual patients with specific genetic diseases.

[0044] Step 2: A medical professional interviews the generated virtual patient.

[0045] Step 3: The system evaluates the interview results and provides feedback to the healthcare professional based on the evaluation results.

[0046] Example 1

[0047] Next, a description will be given of Example 1 of Form 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."

[0048] Conventional medical interview training for medical professionals often involves the use of actual patients, which is time-consuming and costly, and training opportunities are limited. Furthermore, evaluation and feedback of medical interview skills tend to be subjective, making objective evaluation difficult. Furthermore, there is a lack of systems for efficiently improving medical interview skills for patients with various symptoms and medical histories.

[0049] 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.

[0050] In this invention, the server includes a means for conducting medical interview training with a virtual patient using generative artificial intelligence, a means for evaluating the results of the medical interview, a means for providing feedback based on the evaluation results, a means for providing an interface for displaying information about the virtual patient, a means for a user to input questions to the virtual patient and for the virtual patient to respond, a means for analyzing the content of the medical interview and evaluating the user's medical interview skills using an evaluation algorithm, and a means for providing the evaluation results as feedback to the user. This allows medical professionals to efficiently improve their medical interview skills without being restricted by time or location. Furthermore, receiving objective evaluations and specific feedback promotes improvement in medical interview skills.

[0051] "Generative AI" is AI that has the ability to generate new information and responses based on various data.

[0052] A "virtual patient" is a fictitious patient generated by a computer system, rather than a real patient.

[0053] "History training" refers to the practice of questions and responses that medical professionals ask patients.

[0054] The "interview results" are a record of questions asked by a medical professional to a virtual patient and their responses.

[0055] An "evaluation algorithm" is a calculation method for analyzing the results of a medical interview and objectively evaluating the medical interview skills of medical professionals.

[0056] "Feedback" refers to information on areas for improvement and advice provided based on the evaluation results.

[0057] An "interface" is an operating environment containing screens and input fields through which a user interacts with a system.

[0058] "User" refers to a medical professional who uses this system to conduct interview training.

[0059] The "means for inputting a question" refers to a text box or input field for the user to input a question to the virtual patient.

[0060] "Means for returning a response" refers to the function of the virtual patient to generate and display an appropriate response to a user's question.

[0061] The "means for analyzing the contents of the medical interview" refers to a function for analyzing the user's questions and the virtual patient's responses, and making an evaluation based on an evaluation algorithm.

[0062] The "means for providing evaluation results" refers to a function for providing the evaluation results obtained by the evaluation algorithm to the user as feedback.

[0063] This invention relates to a virtual patient interview training system that utilizes generative artificial intelligence. This system allows medical professionals to interview virtual patients, evaluate the results, and provide feedback to improve their interview skills.

[0064] System configuration

[0065] 1. Virtual patient generation

[0066] The server uses generative artificial intelligence (e.g., natural language processing models) to generate virtual patients with various symptoms and medical histories. The generated virtual patient information includes age, gender, chief complaint, medical history, current symptoms, etc.

[0067] 2. Providing a medical interview interface

[0068] The terminal provides a user (healthcare professional) with an interface that displays information about the virtual patient, including basic information about the virtual patient and input fields for conducting a medical interview.

[0069] 3. Conducting a medical interview

[0070] The user inputs questions to the virtual patient through a terminal, and the virtual patient uses generative artificial intelligence to provide appropriate responses.

[0071] 4. Evaluation of interview results

[0072] The server analyzes the user's interview content and evaluates the user's interview skills using an evaluation algorithm, including the appropriateness of the questions, the comprehension of the responses, and the flow of the interview.

[0073] 5. Providing Feedback

[0074] The device provides the evaluation results to the user as feedback, which specifically indicates areas for improvement and merits.

[0075] Hardware and software used

[0076] Server: A computer system with high-performance computing power that runs generative AI models and analyzes data.

[0077] Device: A computer, tablet, or smartphone operated by a user, providing an interview interface and accepting user input.

[0078] Generative artificial intelligence model: A natural language processing model (e.g., GPT-4®) is used to generate responses for the virtual patient.

[0079] Evaluation algorithm: An algorithm for analyzing the interview results and evaluating the user's interview skills.

[0080] Specific examples

[0081] Example scenario:

[0082] The hypothetical patient is a 35-year-old man who presents with recent headaches and dizziness. He has been previously diagnosed with high blood pressure and is currently taking medication.

[0083] Example prompt sentence:

[0084] "A 35-year-old male patient presents with recent headaches and dizziness. He has been diagnosed with high blood pressure in the past and is currently taking medication. What questions would you ask the patient?"

[0085] Based on this prompt, the user can ask specific questions to the virtual patient, such as "When did your headache start?" or "In what situations does dizziness occur?"

[0086] The server records these questions and the virtual patient's responses, and uses an evaluation algorithm to evaluate the user's interview skills. The evaluation results are fed back as feedback, such as "appropriateness of questions: 90 points," "understandability of responses: 85 points," and "flow of interview: 80 points."

[0087] In this way, users can improve their skills by interviewing virtual patients.

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

[0089] Step 1: Creating a virtual patient

[0090] The server generates a virtual patient using a generative AI model. As input, it receives a prompt statement (e.g., "A 35-year-old male patient complains of recent headaches and dizziness. He has been diagnosed with high blood pressure in the past and is currently taking medication."). Based on this prompt statement, the generative AI model generates detailed information about the virtual patient (e.g., age, gender, chief complaint, medical history, current symptoms, etc.). As output, it obtains information about the generated virtual patient.

[0091] Specific behavior:

[0092] The server inputs prompts into the generative artificial intelligence model to generate detailed information about the virtual patient, which is used for subsequent interview training.

[0093] Step 2: Providing a medical interview interface

[0094] The terminal provides the user with an interface that displays information about the virtual patient. As input, the terminal uses the information about the virtual patient received from the server. As output, an interface is displayed that allows the user to check the information about the virtual patient.

[0095] Specific behavior:

[0096] The terminal displays information about the virtual patient (age, gender, chief complaint, etc.) on the screen and provides a text box where the user can enter questions.

[0097] Step 3: Conduct a medical interview

[0098] The user inputs a question to the virtual patient through the terminal. The question entered by the user in the text box is used as input. The server sends this question to a generative artificial intelligence model, which generates a response from the virtual patient. The output is the response of the virtual patient.

[0099] Specific behavior:

[0100] The user types "When did your headache start?" into a text box. The server sends this question to a generative artificial intelligence model, which generates the virtual patient's response: "My headache started a week ago." The device displays this response to the user.

[0101] Step 4: Evaluate the interview results

[0102] The server analyzes the user's interview content and evaluates the user's interview skills using an evaluation algorithm, using the user's questions and the virtual patient's responses as inputs, and obtaining the evaluation results as output.

[0103] Specific behavior:

[0104] The server analyzes the user's questions and the virtual patient's responses and applies an evaluation algorithm to generate results such as "appropriateness of question: 90 points," "understandability of response: 85 points," and "flow of medical interview: 80 points."

[0105] Step 5: Provide feedback

[0106] The terminal provides the evaluation results as feedback to the user. As input, it uses the evaluation results received from the server. As output, it displays feedback that allows the user to check the evaluation results.

[0107] Specific behavior:

[0108] The device displays the evaluation results on the screen, such as "Question appropriateness: 90 points," "Response comprehension: 85 points," and "Interview flow: 80 points," along with specific advice such as "You can get better results by changing the order of the questions."

[0109] (Application example 1)

[0110] Next, a description will be given of Application Example 1 of Embodiment 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."

[0111] In conventional medical interview training systems, medical professionals have limited opportunities to interact with actual patients, making it difficult to provide a realistic training environment. Furthermore, evaluation and feedback of medical interview results are not provided immediately, which slows down the improvement of medical professionals' skills.

[0112] 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.

[0113] In this invention, the server includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the interview, a means for providing feedback based on the evaluation results, a means for a factory robot to act as the virtual patient and for a medical professional to interview the robot, a means for simulating virtual patients with various symptoms and medical histories using generative AI, and a means for evaluating the results of the interview in real time and providing feedback, thereby enabling medical professionals to improve their medical interview skills in a realistic training environment.

[0114] "Generative AI" is an artificial intelligence technology that generates new information and content based on various data.

[0115] A "virtual patient" is a patient that is not a real patient, but is simulated by a computer system.

[0116] "Medical history training" is training to improve medical professionals' techniques and skills in conducting medical interviews with patients.

[0117] "Interview results" refers to the content of the interview conducted by a medical professional with a patient and the information obtained as a result.

[0118] "Evaluation" refers to analyzing the medical professional's interview techniques and skills based on the results of the interview and judging whether they are good or bad.

[0119] "Feedback" refers to information such as areas for improvement and advice provided to medical professionals based on the evaluation results.

[0120] A "factory robot" is an automated mechanical device used in production sites such as factories.

[0121] "Simulate" means to imitate and reproduce actual situations or conditions.

[0122] "Real-time" means that processing and response are immediate and without delay.

[0123] A system for implementing this invention includes means for conducting medical interview training with a virtual patient using generative AI, means for evaluating the results of the interview, means for providing feedback based on the evaluation results, means for a factory robot acting as a virtual patient and for a medical professional to interview the robot, means for simulating virtual patients with various symptoms and medical histories using generative AI, and means for evaluating the results of the interview in real time and providing feedback.

[0124] The server generates a virtual patient using generative AI. The generated virtual patient is simulated through an application installed on a factory robot. Medical personnel interview the factory robot. The interview results are sent to the server in real time, and the server evaluates the interview results. Based on the evaluation results, the server provides feedback to the medical personnel.

[0125] The hardware used includes servers, factory robots, and terminals used by medical personnel, while the software used includes generative AI models (e.g., OpenAI® API) and algorithms for evaluating medical interview results.

[0126] As a concrete example, a virtual patient can be generated by inputting the following prompt sentences into a generative AI model:

[0127] Virtual patient generation prompt:

[0128] "Generate virtual patients with a variety of symptoms and medical histories."

[0129] After the healthcare professional has completed the interview, they should evaluate the interview using the following prompts:

[0130] Interview assessment prompt:

[0131] Please rate the following questions:\nQuestion 1: What symptoms are you experiencing?\nAnswer 1: I have a headache.\nQuestion 2: How long have you had these symptoms?\nAnswer 2: About three days ago.

[0132] Based on the assessment, provide feedback using the following prompts:

[0133] Feedback prompt:

[0134] Please provide feedback based on the following assessment results:\n Assessment result: The healthcare professional asks appropriate questions and accurately understands the patient's symptoms.

[0135] This allows medical professionals to improve their medical history taking skills in a realistic training environment.

[0136] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0137] Step 1:

[0138] The server generates a virtual patient using a generative AI model. Specifically, the server inputs the following prompt sentences into the generative AI model:

[0139] "Generate virtual patients with a variety of symptoms and medical histories."

[0140] Based on this input, the generative AI model generates virtual patient data and outputs it to a server, including the virtual patient's symptoms and medical history.

[0141] Step 2:

[0142] The server sends the generated virtual patient data to the factory robot, which then acts as the virtual patient based on this data. Specifically, the factory robot simulates the virtual patient's symptoms and medical history and responds in real time to medical interviews with medical professionals.

[0143] Step 3:

[0144] The user (medical worker) conducts a medical interview with the factory robot. The contents of the interview are input by the user through a terminal. For example, the following questions are asked:

[0145] "What symptoms are you experiencing?"

[0146] The factory robot responds based on the virtual patient's data:

[0147] "I have a headache."

[0148] Step 4:

[0149] The terminal sends the details of the user's interview and the factory robot's response to the server, which receives this data and stores it as the interview results.

[0150] Step 5:

[0151] The server inputs the following prompt sentences into the generative AI model to evaluate the interview results:

[0152] Please rate the following questions:\nQuestion 1: What symptoms are you experiencing?\nAnswer 1: I have a headache.\nQuestion 2: How long have you had these symptoms?\nAnswer 2: About three days ago.

[0153] Based on this input, the generative AI model evaluates the interview results and outputs the evaluation results to the server.

[0154] Step 6:

[0155] The server inputs the following prompt sentences into the generative AI model to provide feedback based on the evaluation results:

[0156] Please provide feedback based on the following assessment results:\n Assessment result: The healthcare professional asks appropriate questions and accurately understands the patient's symptoms.

[0157] Based on this input, the generative AI model generates feedback and outputs it to the server.

[0158] Step 7:

[0159] The server transmits the generated feedback to the user's terminal, and the user can check the feedback through the terminal and use it to improve their interview skills.

[0160] Example 2

[0161] Next, a description will be given of Example 2 of Form 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."

[0162] Conventional medical interview training systems have had the problem of being unable to sufficiently improve medical personnel's medical interview skills due to the difficulty of simulating conversations with actual patients. In addition, evaluation of medical interview results and provision of feedback are often done manually, making efficient training difficult.

[0163] 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.

[0164] In this invention, the server includes means for conducting medical interview training with a virtual patient using generative artificial intelligence, means for evaluating the results of the interview, means for providing feedback based on the evaluation results, means for using prompt sentences to generate the virtual patient, means for transmitting the results of the interview to the server, and means for displaying the feedback on a terminal. This allows medical professionals to improve their skills through interviews with realistic virtual patients, thereby achieving efficient and effective training.

[0165] "Generative AI" is an AI technology that uses natural language processing and machine learning to generate data and information based on specific conditions.

[0166] A "virtual patient" is a fictional character simulated by generative artificial intelligence with symptoms and medical history similar to those of a real patient.

[0167] "Medical history training" involves medical professionals asking questions of virtual patients and determining diagnoses and treatment plans based on the answers.

[0168] The "interview results" are a record of questions asked by medical professionals to a virtual patient and the virtual patient's answers to those questions.

[0169] "Evaluation means" refers to algorithms or programs for analyzing the results of the interview and evaluating their content.

[0170] "Feedback" refers to advice provided to medical professionals on areas for improvement and skill development based on the evaluation results obtained through the evaluation tool.

[0171] A "prompt sentence" is a sentence used to input specific conditions or requests to generative artificial intelligence.

[0172] A "server" is a computer system for executing programs such as generative artificial intelligence, evaluation means, and feedback generation modules.

[0173] A "terminal" is a device such as a computer or tablet that medical professionals use to conduct interview training.

[0174] This invention is a system that utilizes generative artificial intelligence to conduct interview training with virtual patients, evaluate the interview results, and provide feedback based on the evaluation results. Specific embodiments of this system are described below.

[0175] Hardware and software used

[0176] Hardware: Servers, devices (PCs, tablets, etc.)

[0177] Software: Generative AI models (e.g., GPT-4), evaluation algorithms, feedback generation modules

[0178] Virtual patient generation

[0179] The server uses the generative AI model to generate a virtual patient with diabetes symptoms. Specifically, the server inputs the following prompt sentence into the generative AI model:

[0180] Example prompt: "Generate a hypothetical patient who is a 50-year-old male with a BMI of 30, HbA1c of 7.5%, and complains of frequent urination and dry mouth."

[0181] Based on this prompt, the generative AI model generates a detailed virtual patient profile, such as a "50-year-old male, BMI 30, HbA1c 7.5%, complaining of frequent urination and dry mouth" with specific symptoms and medical history.

[0182] Conducting medical interviews

[0183] The user (healthcare worker) uses a terminal to interview the generated virtual patient. The user inputs questions such as the following through the terminal interface.

[0184] "How's your diet been lately?"

[0185] "Do you have an exercise routine?"

[0186] The virtual patient provides answers to these questions through a generative AI model, which the user can use to ask more detailed questions.

[0187] Sending and evaluating medical interview results

[0188] The device sends the results of the medical interview entered by the user to the server, which then analyzes the results using an evaluation algorithm. For example, the server may make an evaluation such as, "There are not enough questions about diet" or "More detailed questions about exercise habits are needed."

[0189] Generating and Providing Feedback

[0190] The server generates feedback for the healthcare professional using the feedback generation module based on the evaluation results. Specifically, the following feedback is generated:

[0191] "It would be good to ask more questions about the diet."

[0192] "By asking specific questions about exercise habits, we can get a better understanding of the patient's lifestyle."

[0193] The device then displays the received feedback to the user, allowing the user to receive specific advice on how to improve their interview skills.

[0194] This system is an effective tool for medical professionals to improve their medical interview skills through virtual patients. The server uses a generative AI model to generate realistic virtual patients, evaluates the interview results, and provides feedback to help medical professionals improve their skills.

[0195] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0196] Step 1:

[0197] Virtual patient generation

[0198] The server uses the generative AI model to generate a virtual patient with diabetes symptoms.

[0199] Input: Prompt: "Create a virtual patient who is a 50-year-old male with a BMI of 30, HbA1c of 7.5%, and complains of frequent urination and dry mouth."

[0200] Data processing: The server inputs the prompt sentences into a generative AI model to generate a detailed profile of the virtual patient.

[0201] Output: Virtual patient profile (e.g., "50-year-old male, BMI 30, HbA1c 7.5%, frequent urination, dry mouth")

[0202] Step 2:

[0203] Conducting medical interviews

[0204] The user (medical worker) uses a terminal to interview the generated virtual patient.

[0205] Input: Questions entered by the user through the device interface (e.g., "How is your diet these days?", "Do you exercise regularly?")

[0206] Data processing: The virtual patient generates answers to these questions through a generative AI model.

[0207] Output: The hypothetical patient's response (e.g., "My diet is high in calories and I barely exercise.")

[0208] Step 3:

[0209] Sending medical interview results

[0210] The terminal transmits the results of the medical interview entered by the user to the server.

[0211] Input: User-entered questions and virtual patient answers

[0212] Data processing: The device converts all of this data into a format that can be sent to the server.

[0213] Output: Interview result data sent to the server

[0214] Step 4:

[0215] Evaluation of interview results

[0216] The server analyzes the received interview results using an evaluation algorithm.

[0217] Input: Submitted medical interview result data

[0218] Data processing: The server uses an evaluation algorithm to analyze the interview results and identify areas for improvement and shortcomings.

[0219] Output: Evaluation results (e.g., "There are not enough questions about diet," "More detailed questions about exercise habits are needed")

[0220] Step 5:

[0221] Generating and Providing Feedback

[0222] The server generates feedback for the healthcare professional based on the evaluation results using a feedback generation module.

[0223] Input: Evaluation result

[0224] Data processing: The server uses a feedback generation module to generate specific improvements and advice for skill development.

[0225] Output: Feedback (e.g., "It would be good to ask more questions about diet" or "By asking specific questions about exercise habits, we can get a better understanding of the patient's lifestyle")

[0226] Step 6:

[0227] View Feedback

[0228] The terminal displays the received feedback to the user.

[0229] Input: Feedback sent by the server

[0230] Data processing: The device displays the feedback in a user-friendly format.

[0231] Output: Feedback displayed to the user

[0232] In this way, healthcare professionals can receive specific advice on how to improve their interview skills through virtual patients.

[0233] (Application example 2)

[0234] Next, a description will be given of Application Example 2 of Form 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."

[0235] Conventional training systems for medical professionals make it difficult to practice medical interviews with actual patients, and often do not provide sufficient evaluation or feedback on the results of the interviews. This has led to problems such as delays in improving medical interview skills. Furthermore, there has been a lack of appropriate systems for providing health management training to operators and safety managers working on-site at factories and other facilities. To solve these issues, an effective training system using virtual patients is needed.

[0236] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0237] In this invention, the server includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the medical interview, a means for providing feedback based on the evaluation results, a means for interacting with the virtual patient using a smart device, and a means for generating the virtual patient and evaluating the results of the medical interview via a cloud service. This enables medical professionals and factory operators to effectively improve their medical interview skills and health management skills without having to meet face-to-face with actual patients or workers.

[0238] "Generative AI" refers to artificial intelligence techniques used to generate virtual patients and scenarios.

[0239] "Medical history training" refers to practice in which medical professionals or operators conduct medical history interviews with virtual patients.

[0240] "Evaluation of interview results" is the process of analyzing the results of an interview with a virtual patient and evaluating their quality.

[0241] "Feedback" is information that provides areas for improvement and ways to improve skills based on the evaluation of the interview results.

[0242] "Smart devices" refers to interactive electronic devices such as smartphones, smart glasses, and head-mounted displays.

[0243] "Interaction" refers to the act of a user interacting with a virtual patient.

[0244] "Cloud services" refers to computing resources and services provided over the Internet.

[0245] A system for implementing this invention includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the interview, a means for providing feedback based on the evaluation results, a means for interacting with the virtual patient using a smart device, and a means for generating the virtual patient and evaluating the results of the interview via a cloud service.

[0246] System configuration

[0247] 1. Generative AI

[0248] The server generates virtual patients using a generative AI. The generative AI uses pre-trained models to generate virtual patients with various symptoms and medical histories. For example, to generate a virtual patient with diabetes, the following prompt sentence is used:

[0249] Generate a virtual patient with diabetes symptoms.

[0250] 2. Medical interview training

[0251] The user interacts with the virtual patient using a smart device (e.g., smart glasses), conducts a medical interview with the virtual patient, and inputs the results into the system.

[0252] 3. Evaluation of interview results

[0253] The server uses generative AI to evaluate the questionnaire results entered by the user. An example of an evaluation prompt is as follows:

[0254] Please rate the results of the following interview.

[0255] Interview results: [Virtual patient information]

[0256] evaluation:

[0257] 4. Providing Feedback

[0258] The server generates feedback based on the evaluation results. An example of a feedback prompt is:

[0259] Based on the evaluation below, please indicate areas for improvement in your interview and how you can improve your skills.

[0260] Evaluation: [Evaluation result]

[0261] feedback:

[0262] The generated feedback is displayed on the user's smart device.

[0263] Hardware and software used

[0264] Hardware: Smart devices (smart glasses, smartphones, head-mounted displays, etc.)

[0265] Software: Generative AI models (e.g., OpenAI API), cloud services (e.g., AWS (registered trademark))

[0266] Specific examples

[0267] For example, consider a scenario in which a factory operator wears smart glasses and interviews a virtual patient. In this case, the operator checks the virtual patient's information through the smart glasses and proceeds with the interview. The interview results are sent to a server via a cloud service, where they are evaluated by generative AI. Based on the evaluation results, the operator can receive feedback on areas where and how to improve their interview skills.

[0268] In this way, medical professionals and factory operators can effectively improve their medical history and health management skills without having to meet actual patients or workers face-to-face.

[0269] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0270] Step 1:

[0271] The server generates a virtual patient using a generative AI. The prompt sentence "Please generate a virtual patient with symptoms of diabetes" is used as input. The generative AI model generates data for the virtual patient based on this prompt sentence and provides detailed information about the virtual patient as output.

[0272] Step 2:

[0273] The user interacts with the virtual patient using a smart device (e.g., smart glasses). The user interviews the virtual patient and inputs the results into the smart device. The input includes the interview content the user conducted with the virtual patient. The output is the interview results stored in the smart device.

[0274] Step 3:

[0275] The terminal sends the medical interview results to the server via the cloud service. The input includes the medical interview results entered by the user into the smart device. The output is the medical interview results sent to the server.

[0276] Step 4:

[0277] The server uses a generative AI to evaluate the received medical interview results. As input, it uses the medical interview results and the evaluation prompt "Please rate the following medical interview results.\n\nMedical interview results: [Virtual patient information]\n\nEvaluation:". The generative AI model evaluates the medical interview results based on this prompt and provides the evaluation result as output.

[0278] Step 5:

[0279] The server generates feedback based on the evaluation results. As input, it uses the evaluation results and a feedback prompt: "Based on the evaluation below, please indicate areas for improvement in the interview and ways to improve your skills.\n\nEvaluation: [Evaluation result]\n\nFeedback:". The generative AI model generates feedback based on this prompt and provides the feedback information as output.

[0280] Step 6:

[0281] The server sends the generated feedback to the user's smart device through a cloud service. As an input, the feedback information is included. As an output, the feedback information is displayed on the user's smart device.

[0282] Step 7:

[0283] The user can check the feedback through their smart device and use it to improve their medical interview and health management skills. The input includes feedback information sent from the server. The output is the user receiving the feedback and reflecting it in their actual training.

[0284] Example 3

[0285] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0286] With conventional medical interview training systems, it was difficult to gain experience interviewing real patients, especially patients with rare conditions or specific genetic diseases. Furthermore, there was a problem that the improvement of medical interview skills was delayed because the evaluation and feedback of the interview results was not sufficiently carried out.

[0287] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0288] In this invention, the server includes means for a user to input a prompt sentence, means for a terminal to send the prompt sentence to the server, means for the server to receive the prompt sentence and send it to the generative AI model, means for the generative AI model to generate medical condition data of the virtual patient, means for the server to receive the generated medical condition data and return it to the terminal, and means for the terminal to display the medical condition data to the user. This allows the user to gain experience in interviewing virtual patients with rare conditions or specific genetic diseases, and improves their interview skills through evaluation of the interview results and feedback.

[0289] "Generative AI" is a system that uses artificial intelligence technology to generate data and information based on specific conditions.

[0290] A "virtual patient" is not a real patient, but a fictitious patient generated on a computer system.

[0291] "Medical history training" is training for medical professionals to understand the patient's condition and make a diagnosis through dialogue with the patient.

[0292] A "prompt statement" is an instruction statement entered into a generative AI to cause it to generate specific data or information.

[0293] A "terminal" is a device such as a computer or smartphone operated by a user.

[0294] A "server" is a computer system that processes and stores data over a network.

[0295] A "generative AI model" is an algorithm or program that forms the core of generative AI and generates data based on input prompts.

[0296] "Medical condition data" is data that includes information about a patient's symptoms and medical history.

[0297] "Feedback" refers to information such as areas for improvement and advice provided based on the results of the interview.

[0298] MODE FOR CARRYING OUT THE INVENTION

[0299] This invention is a system for conducting interview training with virtual patients using generative AI. The system involves a series of processes: a user inputs a prompt, a generative AI model generates medical condition data for the virtual patient, the interview results are evaluated, and feedback is provided.

[0300] Hardware and software used

[0301] The system uses the following hardware and software:

[0302] Device: A device such as a computer or smartphone operated by a user.

[0303] Server: A computer system that processes and stores data over a network.

[0304] Generative AI model: The core algorithm or program of generative AI. For example, a model that uses natural language processing technology (e.g., GPT-4).

[0305] Specific operation of the system

[0306] The user accesses the terminal interface and inputs a prompt to generate a medical condition for a virtual patient with a specific genetic disease, for example, "Please generate a medical condition for a virtual patient with Fabry disease. The patient is a 30-year-old male whose main symptoms are pain and rash in the hands and feet."

[0307] The device sends the prompt text entered by the user to the server. The server receives the prompt text sent from the device and sends it to the generative AI model. The generative AI model generates medical condition data for the virtual patient based on the prompt text.

[0308] The generated medical condition data is sent back to the server, which then sends the data back to the terminal. The terminal displays the medical condition data received from the server to the user. The user can check the displayed medical condition data and gain experience in interviewing the virtual patient.

[0309] Specific examples

[0310] For example, if a user wants to generate the symptoms of a virtual patient with a specific genetic disease, Fabry disease, he or she would enter the following prompt into the terminal:

[0311] "Generate the symptoms of a hypothetical patient with Fabry disease. The patient is a 30-year-old man whose main symptoms are pain in the hands and feet and a rash."

[0312] The server receives this prompt and sends it to the generative AI model, which generates the following medical condition data:

[0313] "A 30-year-old male patient suffers from Fabry disease, with the main symptoms being severe pain in the hands and feet and a red rash. In addition, the patient complains of frequent fatigue and gastrointestinal upset. Blood tests have confirmed that he has reduced α-galactosidase A activity."

[0314] In this way, the user can gain experience in interviewing patients with conditions that would be difficult to experience in real life. Furthermore, through evaluation of the interview results and feedback, the user's interview skills can be improved. The flow of the identification process in the third embodiment will be described with reference to FIG. 15.

[0315] Program processing flow

[0316] Step 1: The user enters a prompt statement

[0317] The user accesses the terminal interface and inputs a prompt to generate a medical condition for a virtual patient with a specific genetic disease, for example, "Please generate a medical condition for a virtual patient with Fabry disease. The patient is a 30-year-old male whose main symptoms are pain and rash in the hands and feet."

[0318] Input: The prompt text that the user enters

[0319] Output: The prompt text displayed on the terminal

[0320] Step 2: The terminal sends the prompt to the server

[0321] The terminal sends the prompt text entered by the user to the server, converting it into a suitable format so that the server can understand it.

[0322] Input: The prompt text entered by the user

[0323] Output: The prompt sent to the server

[0324] Step 3: The server receives the prompt and sends it to the generative AI model

[0325] The server receives the prompt sentence sent from the terminal. Then, the server sends this prompt sentence to the generative AI model. The server performs the necessary preprocessing so that the generative AI model can process the prompt sentence.

[0326] Input: Prompt sent from the terminal

[0327] Output: The prompt sent to the generative AI model

[0328] Step 4: The generative AI model generates the virtual patient's medical condition data

[0329] The generative AI model generates medical condition data for virtual patients based on prompt sentences sent from the server. The model uses natural language processing technology to create data for virtual patients with detailed medical conditions related to the specified genetic disease.

[0330] Input: Prompt text sent from the server

[0331] Output: Medical condition data of the generated virtual patient

[0332] Step 5: The server receives the generated medical condition data and sends it back to the device.

[0333] The server receives the medical condition data generated by the generative AI model, then sends this data back to the device, where it converts the data into a format that the device can display.

[0334] Input: Medical condition data sent from the generative AI model

[0335] Output: Medical condition data returned to the terminal

[0336] Step 6: The device displays the medical condition data to the user

[0337] The terminal displays the medical condition data received from the server to the user, who can check the displayed medical condition data and gain experience in interviewing the virtual patient.

[0338] Input: Medical condition data sent from the server

[0339] Output: Medical condition data displayed to the user

[0340] (Application example 3)

[0341] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0342] Conventional medical interview training for healthcare professionals is based on common symptoms and medical histories, making it difficult to gain experience interviewing patients with rare conditions. Training using actual patients also poses ethical issues and resource constraints, making it difficult to provide sufficient training opportunities. Furthermore, evaluation and feedback of interview results is insufficient, preventing healthcare professionals from efficiently improving their skills.

[0343] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0344] In this invention, the server includes: means for conducting medical interview training with a virtual patient using generative AI; means for evaluating the results of the medical interview; means for providing feedback based on the evaluation results; means for including a simulator that is installed on a factory robot and enables medical professionals to interact with virtual patients with rare medical conditions and gain medical interview experience; and means for generating the virtual patient's medical condition in real time using a generative AI model and providing appropriate responses to questions from the medical professionals. This allows medical professionals to gain experience in medical interviews with virtual patients with rare medical conditions, efficiently improving their medical interview skills.

[0345] "Generative AI" is a system that uses artificial intelligence technology to generate new data and information.

[0346] A "virtual patient" is a patient that does not actually exist but is recreated through computer simulation.

[0347] "Medical history training" is training to improve medical professionals' skills in listening to patients' symptoms and medical history.

[0348] "Interview results" is a general term for information obtained from patients by medical professionals.

[0349] "Evaluation" means analyzing the skills and responses of medical professionals based on the results of the medical interview and judging whether they are good or bad.

[0350] "Feedback" means providing suggestions for improvement or advice based on the evaluation results.

[0351] A "factory robot" is an automated mechanical device designed to perform tasks in a factory.

[0352] A "simulator" is a device or software that simulates a real-world situation.

[0353] A "generative AI model" is an algorithm or program that forms the core of generative AI and is used to generate new data and information.

[0354] "Real-time" refers to immediate processing or response without delay.

[0355] "Disease condition" refers to the symptoms or disease state exhibited by a patient.

[0356] A "question" is a question that a medical professional asks a patient to obtain information.

[0357] A "response" is an answer or reaction to a question.

[0358] A system for implementing this invention includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the medical interview, a means for providing feedback based on the evaluation results, a means for including a simulator that is installed on a factory robot and allows medical professionals to interact with virtual patients with rare medical conditions and gain medical interview experience, and a means for generating the virtual patient's medical condition in real time using a generative AI model and providing appropriate responses to questions from medical professionals.

[0359] Hardware and software used

[0360] Hardware: Factory robots

[0361] Software: OpenAI API, Python (registered trademark) program

[0362] System processing overview

[0363] 1. Data entry: Medical professionals input questions into the robot, which then uses voice recognition technology to convert the questions into text data.

[0364] 2. Prompt generation: Generate a prompt based on the input question. The prompt is input data that allows the generative AI model to generate an appropriate response.

[0365] 3. Invoke the generative AI model: Using the OpenAI API, send prompts to the generative AI model (e.g., GPT-3®) to generate a response for the virtual patient.

[0366] 4. Response display: The generated response is displayed to the medical staff. The robot outputs the response aloud using speech synthesis technology.

[0367] Specific examples

[0368] When a healthcare professional asks, "What are your recent symptoms?", the robot converts the question into text data and generates the following prompt:

[0369] Prompt Sentence Examples

[0370] Healthcare professional questions: What are your recent symptoms?

[0371] Virtual patient response:

[0372] Based on this prompt, the generative AI model generates the virtual patient's response, "I've been having frequent headaches lately, especially when I wake up in the morning." The robot then relays this response to the healthcare professional via voice.

[0373] Data processing and calculation

[0374] Speech recognition: Converting healthcare professional questions from speech to text.

[0375] Prompt generation: Generate a prompt based on the question text.

[0376] Invoke the generative AI model: Send the prompt to the generative AI model and generate a response.

[0377] Speech synthesis: Converts the generated response into speech and delivers it to the healthcare professional.

[0378] This system allows medical professionals to gain experience in interviewing virtual patients with rare medical conditions, efficiently improving their interview skills.

[0379] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0380] Step 1:

[0381] The user (medical worker) inputs a question to the robot by voice, and the robot converts the question into text data using voice recognition technology.

[0382] Input: Voice question from healthcare professional

[0383] Output: Question data in text format

[0384] Step 2:

[0385] The server receives textual question data and generates prompts, which are input data for the generative AI model to generate appropriate responses.

[0386] Input: Question data in text format

[0387] Output: prompt statement

[0388] Step 3:

[0389] The server sends the generated prompt sentences to the generative AI model using the OpenAI API to generate responses for the virtual patient. The generative AI model generates appropriate responses based on the prompt sentences.

[0390] Input: prompt statement

[0391] Output: Virtual patient response data

[0392] Step 4:

[0393] The server receives the generated virtual patient response data and sends it to the robot, which uses speech synthesis technology to convert the response into voice and convey it to medical professionals.

[0394] Input: Virtual patient response data

[0395] Output: Audio response

[0396] Step 5:

[0397] The server evaluates the results of the interview and provides feedback based on the evaluation results, including information indicating how to improve the interview results and how to improve the interview skills.

[0398] Input: Medical interview result data

[0399] Output: Feedback information

[0400] In this way, medical professionals can gain experience in interviewing virtual patients with rare conditions, and their interview skills can be improved efficiently.

[0401] 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.

[0402] "Example 1"

[0403] One embodiment of the present invention is a system in which generative AI includes an emotion engine that recognizes a user's emotions. In this system, as the user interviews a virtual patient, the emotion engine recognizes the user's emotional state in real time. Specifically, the emotion engine infers the user's emotional state from the user's tone of voice, facial expression, and choice of words. The system then adjusts the virtual patient's response based on the user's emotional state. For example, if the user feels irritated, the virtual patient will respond more considerately.

[0404] "Example 2"

[0405] In another embodiment of the present invention, the system includes an emotion engine that adjusts the virtual patient's response according to the user's emotional state. In this system, the emotion engine recognizes the user's emotional state and adjusts the virtual patient's response according to the state. For example, if the user feels depressed, the virtual patient will choose more encouraging words.

[0406] "Example 3"

[0407] Furthermore, in another embodiment of the present invention, there is a system in which the feedback includes information indicating the correlation between the user's emotional state and the interview result. This system provides feedback on the correlation between the user's emotional state and the interview result, along with an evaluation of the interview result. For example, if the user felt anger or anxiety, information indicating how the emotion affected the interview result is provided. This allows the user to understand the importance of managing their own emotions and identify areas for improvement to conduct better interviews.

[0408] The processing flow of each embodiment will be described below.

[0409] "Example 1"

[0410] Step 1: The user starts the system and begins an interview with the virtual patient.

[0411] Step 2: The emotion engine recognizes the user's emotional state in real time from their tone of voice, facial expressions, and choice of words.

[0412] Step 3: The emotion engine adjusts the virtual patient's response depending on the perceived emotional state. For example, if the user feels irritated, the virtual patient will react more considerately.

[0413] "Example 2"

[0414] Step 1: The user starts the system and begins an interview with the virtual patient.

[0415] Step 2: The emotion engine recognizes the user's emotional state in real time.

[0416] Step 3: The emotion engine adjusts the virtual patient's response depending on the perceived emotional state. For example, if the user feels depressed, the virtual patient will choose more encouraging words.

[0417] "Example 3"

[0418] Step 1: The user starts the system and begins an interview with the virtual patient.

[0419] Step 2: Once the interview is complete, the system evaluates the results.

[0420] Step 3: The system provides feedback on the user's emotional state and its relevance to the results of the interview. For example, if the user felt angry or anxious, the system provides information on how those emotions influenced the results of the interview.

[0421] Example 1

[0422] Next, a description will be given of Example 1 of Form 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."

[0423] Conventional medical interview training systems do not provide feedback that takes into account the user's emotional state when medical professionals interview virtual patients, making it difficult to fully improve their ability to respond in real-life clinical situations. In addition, the virtual patient's reactions are fixed, making it difficult to provide a realistic medical interview experience because the system is unable to respond flexibly to the user's emotional state.

[0424] 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.

[0425] In this invention, the server includes means for conducting medical interview training with a virtual patient using generative artificial intelligence, means for evaluating the results of the medical interview, means for providing feedback based on the evaluation results, means for recognizing the emotional state of the user in real time, and means for adjusting the response of the virtual patient according to the recognized emotional state. This enables a realistic medical interview experience that takes the emotional state of the user into consideration, thereby improving the response capabilities of medical professionals.

[0426] "Generative AI" is AI that has the ability to generate new data and information based on user input.

[0427] A "virtual patient" is a patient that is not a real patient, but is simulated by a computer system.

[0428] "History-taking training" involves medical professionals asking questions of virtual patients to gather information for diagnosis and treatment.

[0429] The "interview results" are the sum of the questions asked by the medical professional to the virtual patient and the answers given by the virtual patient to those questions.

[0430] "Evaluation" is the process of analyzing the results of the interview and determining their appropriateness and validity.

[0431] "Feedback" refers to information on areas for improvement and advice provided to healthcare professionals based on the evaluation results.

[0432] "Emotional state" indicates the user's psychological state and is estimated from the tone of voice, facial expression, choice of words, etc.

[0433] "Real-time" means processing or reacting immediately, without delay.

[0434] "Adjusting responses" means changing the virtual patient's responses and behavior depending on the user's emotional state.

[0435] This invention is a virtual patient interview training system that utilizes generative artificial intelligence. This system allows medical professionals to interview virtual patients, evaluate the results, and provide feedback. Furthermore, it has the ability to recognize the user's emotional state in real time and adjust the virtual patient's responses accordingly.

[0436] Hardware and software used

[0437] Hardware: High-performance servers, devices used by users (PCs, tablets, smartphones)

[0438] Software: Generative AI models (e.g., GPT-4), emotion engines, speech recognition software

[0439] Specific operation of the system

[0440] Virtual patient generation

[0441] The server uses a generative AI model to generate virtual patients with various symptoms and medical histories. For example, by entering a prompt such as "Please generate a virtual patient who is a 40-year-old male with a history of smoking and whose recent symptoms are cough and chest pain," the server generates data for the virtual patient.

[0442] Beginning of medical interview

[0443] The user uses the device to initiate a consultation with the virtual patient. The device uses voice recognition software to convert the user's question into text data. For example, if the user asks, "What are your recent symptoms?", the device converts this into text data and sends it to the server.

[0444] Emotion recognition

[0445] The device uses an emotion engine to recognize the user's emotional state in real time from their tone of voice, facial expressions, and choice of words. For example, if the user is irritated, the emotion engine will recognize this as "irritation" and send that data to the server.

[0446] Virtual patient response adjustment

[0447] The server adjusts the virtual patient's response based on the output of the emotion engine. For example, if the server detects that the user is irritated, it instructs the virtual patient to respond with a considerate response, such as, "Can you tell me more?"

[0448] Evaluation of interview results

[0449] The server evaluates the content of the medical interview and analyzes the appropriateness of the user's questions and the quality of the response. For example, it may evaluate the user's questions as appropriate and improve their understanding of the patient's symptoms.

[0450] Providing Feedback

[0451] The server provides feedback to the user based on the evaluation results, such as "Next time, please ask questions with more consideration for the patient's feelings."

[0452] This completes the embodiment of the present invention. This system enables a realistic medical interview experience that takes into account the user's emotional state, thereby improving the response capabilities of medical professionals.

[0453] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0454] Step 1:

[0455] Virtual patient generation

[0456] The server uses a generative AI model to generate a virtual patient by inputting a prompt sentence.

[0457] Input: Prompt: "Create a virtual patient: a 40-year-old male with a history of smoking, whose current symptoms are cough and chest pain."

[0458] Data processing: A generative AI model analyzes the prompts and generates a detailed virtual patient profile and symptoms.

[0459] Output: Virtual patient data (e.g., age, gender, medical history, current symptoms).

[0460] Specific operation: The server stores the generated virtual patient data in a database.

[0461] Step 2:

[0462] Beginning of medical interview

[0463] The user uses the terminal to begin an interview with the virtual patient.

[0464] Input: A user spoken question (e.g., "What are your recent symptoms?").

[0465] Data processing: The device uses voice recognition software to convert the user's speech into text data.

[0466] Output: User's question converted to text data.

[0467] Specific operation: The device sends text data to the server.

[0468] Step 3:

[0469] Emotion recognition

[0470] The device uses an emotion engine to recognize the user's emotional state in real time from their tone of voice, facial expressions, and choice of words.

[0471] Input: User audio and video data.

[0472] Data processing: The emotion engine analyzes audio and video data to estimate the user's emotional state.

[0473] Output: User's emotional state (e.g., annoyed, calm).

[0474] Specific operation: The device sends the recognized emotion data to the server.

[0475] Step 4:

[0476] Virtual patient response adjustment

[0477] The server adjusts the virtual patient's response based on the output of the emotion engine.

[0478] Input: User emotional state data.

[0479] Data processing: The server adjusts the virtual patient's responses according to their emotional state using a generative AI model.

[0480] Output: Tailored virtual patient response (e.g., "Can you tell me more?").

[0481] Specific operation: The server sends the adjusted response to the terminal, and the virtual patient reacts appropriately to the user.

[0482] Step 5:

[0483] Evaluation of interview results

[0484] The server evaluates the contents of the medical interview and analyzes the appropriateness of the user's questions and the quality of the response.

[0485] Input: User questions and virtual patient response data.

[0486] Data processing: The server analyzes the interview content and uses an evaluation algorithm to determine its appropriateness and validity.

[0487] Output: Evaluation result (e.g., "The user's question was appropriate and provided a better understanding of the patient's condition").

[0488] Specific operation: The server stores the evaluation results in a database.

[0489] Step 6:

[0490] Providing feedback

[0491] The server provides feedback to the user based on the evaluation results.

[0492] Input: Evaluation result data.

[0493] Data processing: The server generates improvements and advice based on the evaluation results.

[0494] Output: Feedback (e.g., "Next time, please be more considerate of the patient's feelings when asking questions").

[0495] Specific operation: The server sends the feedback content to the user's device.

[0496] (Application example 1)

[0497] Next, a description will be given of Application Example 1 of Embodiment 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."

[0498] Conventional training systems provide uniform feedback without considering the user's emotional state, which means they are unable to respond to the stress and emotional fluctuations that occur in real work. In addition, the responses of virtual objects are fixed, which means users are unable to fully develop the ability to respond to diverse scenarios.

[0499] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means. In this invention, the server includes means for training with a virtual object using generative AI, means for evaluating the training results, means for providing feedback based on the evaluation results, emotion recognition means for recognizing the user's emotional state in real time, and means for adjusting the response of the virtual object according to the recognized emotional state. This makes it possible to provide appropriate feedback according to the user's emotional state and develop the ability to respond to a variety of scenarios.

[0500] "Generative AI" is an artificial intelligence technology that generates new information and scenarios from data.

[0501] A "virtual object" is an object that does not exist in reality but is simulated on a computer system.

[0502] "Training means" refers to a function that allows users to improve their skills through interaction with and manipulation of virtual objects.

[0503] The "means for evaluating training results" is a function for analyzing the content of the training performed by the user and evaluating the results.

[0504] The "means for providing feedback" is a function that provides the user with suggestions for improvement and advice based on the evaluation results.

[0505] The "emotion recognition means" is a function that recognizes the user's emotional state in real time from the user's tone of voice, facial expression, choice of words, etc.

[0506] The "means for adjusting the reaction of the virtual object" is a function that dynamically changes the reaction of the virtual object depending on the recognized emotional state of the user.

[0507] A "scenario" is a set of circumstances or events that a virtual object has.

[0508] "Skill improvement methods" are specific techniques and advice for users to improve specific abilities or techniques.

[0509] As an embodiment of the present invention, the following system is constructed.

[0510] The server includes a means for training with a virtual object using generative AI, a means for evaluating the training results, a means for providing feedback based on the evaluation results, an emotion recognition means for recognizing the user's emotional state in real time, and a means for adjusting the response of the virtual object according to the recognized emotional state.

[0511] Specifically, the server uses a generative AI model to generate virtual objects with various scenarios. Users interact with and train these virtual objects through smartphones, smart glasses, head-mounted displays, or robots. The server evaluates the training results and provides feedback based on the evaluation results.

[0512] The system uses a camera and microphone to recognize the user's emotional state in real time based on their tone of voice, facial expressions, and choice of words. The system uses the EmotionRecognizer library for emotion recognition. The virtual object's response is dynamically adjusted according to the recognized emotional state.

[0513] For example, if a user responds to a virtual robot by saying, "I'll check the robot's malfunction," the server evaluates the response using a generative AI model. If the server recognizes that the user is irritated, the virtual robot adjusts its response to respond calmly.

[0514] Examples of prompts include:

[0515] Scenario generation prompt: "Generate a scenario in which a factory robot breaks down."

[0516] Response evaluation prompt: "Evaluate the operator's response: Identify the faulty part of the robot."

[0517] In this way, it is possible to provide appropriate feedback according to the user's emotional state and develop the ability to respond to a variety of scenarios.

[0518] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0519] Step 1:

[0520] The server uses the generative AI model to generate a scenario for the virtual object. Specifically, the server inputs a prompt statement, "Please generate a scenario in the event that a factory robot breaks down," into the generative AI model and generates a scenario. The input is the prompt statement, and the output is the generated scenario.

[0521] Step 2:

[0522] The server transmits the generated scenario to a user terminal, which can be a smartphone, smart glasses, a head-mounted display, or a robot. The input is the generated scenario, and the output is the display of the scenario on the user terminal.

[0523] Step 3:

[0524] The user interacts with the virtual object and performs training. The user's response is input as voice or text. The input is the user's response, and the output is the transmission of the response data to the server.

[0525] Step 4:

[0526] The server uses a generative AI model to evaluate the user's response. Specifically, it inputs the prompt statement "Please evaluate the operator's response: Check the robot's fault location" into the generative AI model and obtains the evaluation result. The input is the user's response and the prompt statement, and the output is the evaluation result.

[0527] Step 5:

[0528] The server generates feedback based on the evaluation results and sends it to the user's device. The feedback includes points to improve the training results and ways to improve skills. The input is the evaluation results, and the output is the display of the feedback on the user's device.

[0529] Step 6:

[0530] The user device uses a camera and microphone to recognize the user's emotional state in real time. The EmotionRecognizer library is used for emotion recognition. The input is the user's tone of voice, facial expression, and word choice, and the output is the recognized emotional state.

[0531] Step 7:

[0532] The server adjusts the virtual object's response based on the recognized emotional state. For example, if the user is recognized as irritated, the virtual robot adjusts its response to respond calmly. The input is the recognized emotional state, and the output is the adjusted virtual object's response.

[0533] Step 8:

[0534] The server transmits the adjusted reaction of the virtual object to the user terminal, and the user terminal displays the adjusted reaction and provides feedback to the user. The input is the adjusted reaction of the virtual object, and the output is the display of the reaction to the user terminal.

[0535] Example 2

[0536] Next, a description will be given of Example 2 of Form 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."

[0537] Conventional medical interview training systems generate virtual patients and evaluate the results of the interviews, but they do not provide feedback based on the user's emotional state or adjust the virtual patient's reactions, which means they have limited applicability in actual medical settings. Furthermore, there is a lack of specific feedback on improving medical interview skills, which means that medical professionals are unable to fully improve their skills.

[0538] 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.

[0539] In this invention, the server includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the medical interview, a means for providing feedback based on the evaluation results, and a means for recognizing the emotional state of the user and adjusting the virtual patient's response according to that state. This makes it possible to provide feedback according to the user's emotional state and adjust the virtual patient's response, improving applicability in actual medical settings and improving the medical interview skills of medical professionals.

[0540] "Generative AI" is a system that uses artificial intelligence technology to generate new data or information based on specific conditions or prompts.

[0541] A "virtual patient" is a fictitious patient with specific symptoms and medical history, generated by generative AI.

[0542] "Medical history training" involves medical professionals asking questions of virtual patients and determining diagnoses and treatment plans based on the answers.

[0543] "Interview results" is a general term for questions asked by medical professionals to virtual patients and the answers given by the virtual patients to those questions.

[0544] "Evaluation" is the process of analyzing the results of the interview and determining their quality and appropriateness.

[0545] "Feedback" is information provided based on the evaluation results regarding areas for improvement in interviews and ways to improve skills.

[0546] "Emotional state" refers to the user's psychological state or mood, as recognized by the emotion engine.

[0547] The "emotion engine" is a system that analyzes the user's emotional state from facial expressions, voice, etc. and provides that information.

[0548] "Response modulation" is the process of modifying the responses and behavior of a virtual patient depending on the user's emotional state.

[0549] This invention is a virtual patient interview training system that utilizes generative AI, evaluates the interview results, provides feedback based on the evaluation results, and includes a function to recognize the user's emotional state and adjust the virtual patient's response accordingly.

[0550] Hardware and software used

[0551] Hardware: Servers, devices (PCs, tablets, smartphones)

[0552] Software: Generative AI models (e.g., GPT-4), emotion engines

[0553] Specific operation of the system

[0554] Virtual patient generation

[0555] The server uses the generative AI model to generate a virtual patient with diabetes symptoms. Specifically, the server inputs a prompt statement such as "50-year-old male, with a 5-year history of diabetes and recent elevated blood sugar levels" into the generative AI model, and generates a detailed profile of the virtual patient.

[0556] Conducting medical interviews

[0557] The user (healthcare professional) uses a terminal to interview the virtual patient. The user inputs questions to the virtual patient through the terminal interface, such as "How has your diet been recently?" The terminal then sends these questions to the server.

[0558] Evaluation of interview results

[0559] The server analyzes and evaluates the received medical interview content. Specifically, the server uses a generative AI model to analyze the virtual patient's answers and evaluate the quality of the interview. For example, if a virtual patient answers, "I've been trying to eat more vegetables lately," the server evaluates the answer and generates appropriate feedback.

[0560] Providing Feedback

[0561] Based on the evaluation results, the server generates feedback indicating areas for improvement in the interview and ways to improve diabetes interview skills. Specifically, the server generates feedback such as "There were not enough questions about dietary details, so next time you should ask about specific dietary details," and sends it to the device. The device then displays this feedback to the user.

[0562] Response adjustment by emotion engine

[0563] The server uses an emotion engine to recognize the user's emotional state. Specifically, if the server analyzes the data obtained from the emotion engine and recognizes that the user is depressed, it adjusts the virtual patient's response. For example, the virtual patient will choose encouraging words such as, "Your efforts are great. Keep it up."

[0564] Examples of specific examples and prompts

[0565] Specific examples

[0566] Generating a virtual patient: The server inputs prompt statements such as "50-year-old male, with a 5-year history of diabetes and recent elevated blood sugar levels" into a generative AI model to generate a detailed profile.

[0567] Conducting the medical interview: The user asks the question "How has your diet been lately?" through the terminal, and the terminal sends this question to the server.

[0568] Evaluation of the interview results: The server analyzes the virtual patient's answer, "I've been trying to eat more vegetables recently," and makes an evaluation.

[0569] Providing feedback: The server generates feedback such as "There were not enough questions about the meal contents, so next time you should ask about the specific meal contents," and sends it to the device.

[0570] Response adjustment by emotion engine: The server analyzes the data obtained from the emotion engine, and if it recognizes that the user is depressed, the virtual patient adjusts its response to choose encouraging words such as "Your efforts are great. Keep going."

[0571] Prompt Sentence Examples

[0572] "Generate a hypothetical patient who is a 50-year-old male with a 5-year history of diabetes and recent elevated blood glucose levels."

[0573] "Ask your virtual patient about their recent diet."

[0574] Please evaluate the results of the interview and provide feedback on areas for improvement.

[0575] "If the user is feeling down, tailor your response so that the virtual patient chooses encouraging words."

[0576] The above is an embodiment of the present invention.

[0577] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0578] Step 1:

[0579] Virtual patient generation

[0580] The server uses the generative AI model to generate a virtual patient with diabetes symptoms.

[0581] Input: A prompt such as "50 year old male with 5 years of diabetes, recent high blood sugar levels."

[0582] Data processing: The server inputs the prompt sentences into a generative AI model to generate a detailed profile of the virtual patient.

[0583] Output: A detailed profile of the virtual patient.

[0584] Specific operation: The server inputs prompt sentences into the generative AI model and stores the generated virtual patient profile in the database.

[0585] Step 2:

[0586] Conducting medical interviews

[0587] The user (medical worker) uses a terminal to interview the virtual patient.

[0588] Input: A question typed by the user through the device interface (e.g., "How has your diet been lately?").

[0589] Data processing: The device sends the user's question to the server.

[0590] Output: The question sent to the server.

[0591] Specific operation: The user inputs a question through the terminal, and the terminal sends the question to the server.

[0592] Step 3:

[0593] Evaluation of interview results

[0594] The server analyzes the received interview contents and performs an evaluation.

[0595] Input: The hypothetical patient's response (e.g., "I've been trying to eat more vegetables lately").

[0596] Data processing: The server uses a generative AI model to analyze the virtual patient's responses and evaluate the quality of the interview.

[0597] Output: Evaluation result.

[0598] Specific operation: The server analyzes the answers of the virtual patient and generates an evaluation result.

[0599] Step 4:

[0600] Providing Feedback

[0601] Based on the evaluation results, the server generates feedback indicating areas for improvement in the interview and ways to improve diabetes interview skills.

[0602] Input: Evaluation result.

[0603] Data processing: The server generates feedback based on the evaluation results and sends it to the device.

[0604] Output: Feedback (e.g., "You didn't ask enough questions about dietary habits. Next time, please ask more specifically about your dietary habits.").

[0605] Specific operation: The server generates feedback based on the evaluation results and sends it to the device, which then displays this feedback to the user.

[0606] Step 5:

[0607] Response adjustment by emotion engine

[0608] The server uses an emotion engine to recognize the user's emotional state.

[0609] Input: User emotional state data (e.g. facial expressions and voice data).

[0610] Data processing: The server analyzes the data obtained from the emotion engine and recognizes the user's emotional state.

[0611] Output: The perceived emotional state of the user.

[0612] Specific behavior: The server analyzes the data obtained from the emotion engine, and if it recognizes that the user is depressed, it adjusts the virtual patient's response. For example, the virtual patient will choose encouraging words such as "Your efforts are great. Keep going."

[0613] (Application example 2)

[0614] Next, a description will be given of Application Example 2 of Form 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."

[0615] Conventional medical interview training systems have the problem of being unable to adequately simulate the emotions and reactions of medical professionals when they meet with actual patients. Furthermore, the inability to provide real-time feedback or respond to emotional states limits the improvement of actual medical interview skills. Furthermore, there is a lack of support for conducting appropriate medical interviews with customers in brick-and-mortar stores.

[0616] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0617] In this invention, the server includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the medical interview, a means for providing feedback based on the evaluation results, a means for analyzing the emotional state of a user using an emotion recognition engine, a means for adjusting the response of the virtual patient according to the emotional state of the user, a means for capturing video in real time using smart glasses to support the medical interview, and a means for providing feedback. This enables medical professionals and store clerks to conduct appropriate medical interviews according to the emotional state of customers and patients and receive feedback in real time.

[0618] "Generative AI" is an artificial intelligence technology that generates virtual patients and simulates their symptoms and reactions.

[0619] "Medical history training" is training for medical professionals and store clerks to conduct medical history interviews with virtual patients and improve their skills.

[0620] The "means for evaluating the results of medical interviews" is a system for analyzing the content of medical interviews conducted by medical professionals or store clerks and evaluating their quality and appropriateness.

[0621] The "means for providing feedback" is a system that provides medical professionals and store clerks with advice on areas for improvement and skill development based on the evaluation of the interview results.

[0622] An "emotion recognition engine" is a technology that analyzes a user's facial expressions and tone of voice to recognize their emotional state.

[0623] The "means for analyzing the user's emotional state" is a system that uses an emotion recognition engine to analyze the user's emotions in real time.

[0624] A "means for adjusting the virtual patient's reactions" is a system for changing the virtual patient's reactions depending on the user's emotional state.

[0625] "Smart glasses" are wearable devices that have built-in cameras and displays and are worn by users.

[0626] The "means of capturing video in real time and supporting medical interviews" is a system that uses smart glasses to capture video of customers or patients in real time and supports medical interviews based on that information.

[0627] The system for implementing this invention integrates a generative AI, an emotion recognition engine, smart glasses, and a feedback system. The following describes the details of each component and its operation.

[0628] 1. Generative AI

[0629] The generative AI generates virtual patients and simulates their symptoms and reactions. The server uses the generative AI to generate virtual patients with various symptoms and medical histories, allowing medical professionals and store clerks to train in how to handle a variety of cases.

[0630] 2. Emotion Recognition Engine

[0631] The emotion recognition engine analyzes the user's facial expressions and tone of voice to recognize their emotional state. The server captures the user's video using the camera built into the smart glasses and analyzes it with the emotion recognition engine. This allows the server to grasp the user's emotional state in real time.

[0632] 3. Smart Glasses

[0633] Smart glasses are wearable devices equipped with a built-in camera and display that are worn by the user. The smart glasses capture video in real time and send the information to a server. The server then analyzes the received video using an emotion recognition engine to understand the user's emotional state.

[0634] 4. Feedback System

[0635] The feedback system provides advice to medical professionals and store clerks on areas for improvement and skill development based on the evaluation of the interview results. The server evaluates the interview results and generates feedback based on the evaluation results. The feedback is displayed on the display of the smart glasses.

[0636] Specific examples

[0637] For example, consider a situation where a store clerk is wearing smart glasses and a customer asks, "I've been feeling thirsty a lot lately. What should I do?" If the emotion recognition engine senses anxiety from the customer's facial expression, the virtual patient will select encouraging words such as, "That may be an early symptom of diabetes. Let's start by measuring your blood sugar level." The feedback system will provide feedback to the store clerk, such as, "It would be good to suggest specific actions to ease the customer's anxiety."

[0638] Prompt Sentence Examples

[0639] "If a client is consulting about their diabetes symptoms, what questions should they ask?"

[0640] In this way, medical professionals and store clerks can tailor their interviews to the emotional state of their customers or patients and receive real-time feedback.

[0641] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0642] Step 1:

[0643] The server generates a virtual patient using generative AI. As input, it receives data about the virtual patient's symptoms and medical history, and uses a generative AI model to generate a profile of the virtual patient. As output, it obtains a detailed profile of the virtual patient.

[0644] Step 2:

[0645] The user puts on the smart glasses and starts the medical interview. The camera in the smart glasses captures the video in real time and sends the video data to the server. The user's video data is used as input, and the video data is sent to the server as output.

[0646] Step 3:

[0647] The server analyzes the received video data using an emotion recognition engine. As input, it receives the user's video data and analyzes the user's emotional state using the emotion recognition engine. As output, it obtains data related to the user's emotional state.

[0648] Step 4:

[0649] The server adjusts the virtual patient's reactions according to the user's emotional state. As input, the server uses the user's emotional state data and the virtual patient profile to generate the virtual patient's reactions. As output, the server obtains the adjusted virtual patient's reactions.

[0650] Step 5:

[0651] The user interviews the virtual patient. The user inputs a question, and the question data is sent to the server. The user's question data is used as input, and the question data is sent to the server as output.

[0652] Step 6:

[0653] The server generates and provides the virtual patient's responses to the user. As input, the server uses the user's question data and the virtual patient's profile to generate the virtual patient's responses. As output, the virtual patient's response data is obtained and displayed on the smart glasses display.

[0654] Step 7:

[0655] The server evaluates the interview results and generates feedback. As input, it uses the user's question data and the virtual patient's response data, and performs the evaluation using a feedback system. As output, it obtains feedback data.

[0656] Step 8:

[0657] The server provides the generated feedback to the user. As an input, it uses the feedback data and displays the feedback on the display of the smart glasses. As an output, the user can receive the feedback.

[0658] In this way, medical professionals and store clerks can tailor their interviews to the emotional state of their customers or patients and receive real-time feedback.

[0659] Example 3

[0660] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0661] Conventional medical interview training systems make it difficult for users to gain experience interviewing patients with rare conditions, and lack a means to evaluate the impact of the user's emotional state on the interview results. This makes it difficult to improve users' medical interview skills.

[0662] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0663] In this invention, the server includes means for conducting medical interview training with a virtual patient using generative artificial intelligence, means for evaluating the results of the medical interview, means for providing feedback based on the evaluation results, means for recording the user's emotional state, means for analyzing the user's emotional state, and means for evaluating the correlation between the results of the medical interview and the user's emotional state. This allows the user to gain experience in medical interviews with virtual patients with rare conditions, further understanding the importance of managing their own emotions and improving their medical interview skills.

[0664] "Generative AI" is an AI technology that generates virtual patients and other information based on prompts entered by the user.

[0665] "Medical history training" is training in which a user asks questions to a virtual patient to understand the condition and symptoms.

[0666] The "results of medical interview" is a general term for the contents of the medical interview conducted by the user with the virtual patient and the answers thereto.

[0667] The "evaluation result" is an evaluation of the user's interview skills and other related information generated based on the interview result.

[0668] "Feedback" is information about areas and methods for improving interview skills that is provided to the user based on the evaluation results.

[0669] "Emotional state" refers to the emotional state the user feels during the interview, and includes, for example, anxiety, anger, joy, etc.

[0670] The "means for recording emotional state" refers to a device or software for recording the user's facial expressions, tone of voice, etc.

[0671] The "means for analyzing emotional state" refers to a device or software for analyzing the recorded emotional data of a user and identifying that emotional state.

[0672] The "means for evaluating the correlation between the questionnaire result and the emotional state" is a device or software for evaluating the correlation between the questionnaire result and the emotional state of the user.

[0673] This invention is a system that utilizes generative artificial intelligence to conduct medical interview training with virtual patients, evaluate the interview results, and provide feedback based on the evaluation results. Furthermore, it records and analyzes the user's emotional state, and evaluates the correlation between the interview results and the emotional state, thereby improving the user's medical interview skills.

[0674] Hardware and software used

[0675] Server: A server equipped with a high-performance GPU (e.g., a server equipped with a high-performance graphics processing unit) is used to run the generative artificial intelligence model.

[0676] Generative AI model: Uses a natural language processing model (e.g., an advanced natural language processing model) to generate a virtual patient's medical condition based on a prompt sentence.

[0677] Emotion Analysis Software: Uses an emotion analysis API (e.g., an application programming interface for emotion analysis) to analyze the user's emotional state.

[0678] Program processing

[0679] Virtual patient generation

[0680] The user inputs a prompt for the generative AI model through the terminal interface. For example, the user might input, "Please generate a virtual patient, a 30-year-old male with a specific genetic disease." The terminal then sends this prompt to the server. The server uses the generative AI model to generate the virtual patient's medical condition based on the prompt. Information about the generated virtual patient is sent to the terminal and displayed to the user.

[0681] Conducting medical interviews

[0682] The user interviews the virtual patient through the device. For example, the user might ask, "How are you feeling lately?", to which the virtual patient might respond, "Recently, my hands and feet have been shaking a lot, and I've been dropping things more easily." During the interview process, the device uses a camera and microphone to record the user's facial expressions and tone of voice in real time.

[0683] Sentiment analysis and feedback provision

[0684] Once the interview is complete, the server uses emotion analysis software to analyze the user's emotional state. The server evaluates the correlation between the interview results and the user's emotional state and generates feedback. For example, the server generates feedback such as, "Because you were feeling anxious, the questions were vague and we were unable to elicit accurate information," and sends it to the device. The user can review the feedback displayed on the device, understand the importance of managing their own emotions, and identify areas for improvement in their interview skills.

[0685] Specific examples

[0686] Example prompt: Create a hypothetical patient, a 30-year-old male with a specific genetic disease.

[0687] The virtual patient generated was a 30-year-old male with a specific genetic disorder, characterized by motor and cognitive impairment and mental fluctuations.

[0688] Conducting a medical interview: The user asks, "How are you feeling these days?" and the virtual patient answers, "Recently, my hands and feet have been shaking a lot, and I've been dropping things more easily."

[0689] Feedback: The user is informed that their anxiety led them to ask vague questions and fail to elicit specific symptoms from the virtual patient.

[0690] This system allows users to gain experience in interviewing virtual patients with rare conditions, and furthermore, to understand the importance of managing their own emotions and improve their interviewing skills.The flow of the identification process in Example 3 will be described with reference to Figure 21.

[0691] Step 1:

[0692] The user enters a prompt statement.

[0693] The user inputs a prompt for the generative AI model through the terminal interface, for example, "Please generate a virtual patient, a 30-year-old male with a specific genetic disease." The input prompt is stored in the terminal's memory.

[0694] Step 2:

[0695] The terminal sends the prompt to the server.

[0696] The terminal sends the prompt text entered by the user to the server using the HTTPS protocol, and the sent prompt text is received by the server's receiving module.

[0697] Step 3:

[0698] The server generates a virtual patient using the generative AI model.

[0699] The server inputs the received prompt sentences into a generative AI model to generate a virtual patient's medical condition. The generated virtual patient information includes the patient's age, gender, medical history, current symptoms, etc. The generated data is stored in the server's database.

[0700] Step 4:

[0701] The server transmits information about the generated virtual patient to the terminal.

[0702] The server sends the generated virtual patient information in JSON format to the terminal, which then analyzes the received information and displays it to the user.

[0703] Step 5:

[0704] The user interviews the virtual patient.

[0705] The user asks the virtual patient questions through the terminal. For example, "How are you feeling lately?", and the virtual patient responds, "Recently, my hands and feet have been shaking a lot, and I've been dropping things more easily." The user's questions and the virtual patient's answers are recorded on the terminal.

[0706] Step 6:

[0707] The device records the user's emotional state.

[0708] The device uses a camera and microphone to record the user's facial expressions and tone of voice in real time, and the recorded data is stored in the device's memory.

[0709] Step 7:

[0710] The server performs emotion analysis and evaluates the correlation between the interview results and the emotional state.

[0711] The device sends the recorded emotional data to a server, which then uses emotion analysis software to analyze the user's emotional state. The analysis results are stored in the server's database. The server then evaluates the correlation between the interview results and the user's emotional state and generates an evaluation result.

[0712] Step 8:

[0713] The server generates feedback and sends it to the device.

[0714] The server generates feedback based on the evaluation results. For example, it might say, "Because you were feeling anxious, the questions were vague and we were unable to elicit accurate information," and sends it to the device. The device then displays the received feedback to the user.

[0715] Step 9:

[0716] The user checks the feedback.

[0717] Users can view feedback displayed on the device screen, which helps them understand the importance of managing their own emotions and identify areas for improvement in their interview skills.

[0718] (Application example 3)

[0719] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0720] In modern factories, it is important to monitor the health status of workers in real time and detect abnormalities early. However, it is difficult to take appropriate action quickly when an abnormality in a worker's health is detected. Another challenge is understanding the impact of a worker's emotional state on their health and work efficiency, and providing appropriate feedback. To solve these challenges, a system is needed that monitors the health status of workers, generates a virtual patient to conduct an interview when an abnormality is detected, and provides feedback showing the relationship between the emotional state and the interview results.

[0721] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes a means for conducting interview training with a virtual patient using generative AI, a means for evaluating the interview results, a means for providing feedback based on the evaluation results, a means for monitoring the worker's health condition, a means for generating a virtual patient and conducting an interview when an abnormality is detected, and a means for providing feedback indicating the correlation between the interview results and the worker's emotional state. This makes it possible to monitor the worker's health condition in real time, detect abnormalities early, and take appropriate measures promptly. It is also possible to understand the impact of the worker's emotional state on their health condition and work efficiency and provide appropriate feedback.

[0722] "Generative AI" is a system that uses artificial intelligence technology to generate new data and information.

[0723] A "virtual patient" is a computer-generated model of a patient that does not actually exist.

[0724] "Medical history training" is training to improve medical professionals' skills in interviewing patients.

[0725] "Interview results" refers to information and data obtained through interviews.

[0726] "Evaluation results" are the results of evaluations conducted based on the interview results.

[0727] "Feedback" refers to suggestions for improvement or advice provided based on the evaluation results.

[0728] "Worker" means a person who works in a factory or on a work site.

[0729] "Health status" refers to the physical and mental condition of a worker.

[0730] "Monitoring" means to continuously watch over a particular condition or situation.

[0731] An "abnormality" is a state or situation that is different from the normal.

[0732] "Emotional state" refers to the worker's current feelings and mood.

[0733] "Relationship" indicates that two or more events or data are related to each other.

[0734] The system for implementing this invention is configured as follows: First, the server uses generative AI to generate virtual patients and conduct medical interview training. The generated virtual patients can have a variety of symptoms and medical histories, allowing users to improve their medical interview skills in dealing with a variety of cases.

[0735] The server then evaluates the results of the interview and provides feedback based on the evaluation. The feedback includes suggestions for improving the interview results and ways to improve interview skills. Sensors are also installed to monitor the worker's health, and if an abnormality is detected, the server automatically generates a virtual patient and conducts an interview.

[0736] Furthermore, the server provides feedback indicating the correlation between the interview results and the worker's emotional state, allowing the worker to understand the impact of their own emotional state on their health and work efficiency and take appropriate measures.

[0737] Specifically, the server performs the following process. First, sensors (e.g., heart rate monitor, thermometer, blood pressure monitor, etc.) are used to monitor the worker's health condition in real time. If an abnormality is detected, the generative AI generates a virtual patient and begins a medical interview. The medical interview is conducted based on prompts entered by the user.

[0738] For example, the following prompt sentences are used:

[0739] Please describe your current emotional state (e.g., anxious, angry, happy): Anxiety

[0740] A virtual patient was created: {'disease': 'Genetic disease A', 'symptoms': ['Symptom 1', 'Symptom 2']}

[0741] How are you feeling these days? I'm not feeling well.

[0742] What symptoms are you experiencing? I have headaches and dizziness.

[0743] When did the symptoms start? Three days ago.

[0744] Emotional state: Anxiety

[0745] Interview results:

[0746] I'm not feeling well.

[0747] I have a headache and dizziness.

[0748] It started three days ago.

[0749] Consider how emotional state can affect interview results.

[0750] Based on the prompt, the server evaluates the interview results and provides feedback, including suggestions for improving the interview results and ways to improve interview skills. It also provides information on the correlation between the worker's emotional state and the interview results.

[0751] This system makes it possible to monitor workers' health conditions in real time, detect abnormalities early, and take appropriate measures promptly.It also understands the impact of workers' emotional states on their health and work efficiency, and provides appropriate feedback.

[0752] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0753] Step 1:

[0754] The server monitors the worker's health condition in real time. It collects data from sensors (e.g., heart rate monitors, thermometers, blood pressure monitors, etc.) and checks for any abnormalities. The input is health data from the sensors, and the output is the presence or absence of abnormalities. Specifically, the server periodically obtains data from the sensors and evaluates whether it is within the normal range.

[0755] Step 2:

[0756] When an abnormality is detected, the server uses generative AI to generate a virtual patient. The input is the result of the abnormality detection, and the output is the virtual patient data. Specifically, the server generates an appropriate virtual patient according to the type of abnormality and sets the patient's symptoms and medical history.

[0757] Step 3:

[0758] The server starts an interview with the generated virtual patient. The user inputs a prompt sentence, and the interview is conducted based on that. The input is the prompt sentence from the user, and the output is the interview result. In concrete terms, the server generates a response for the virtual patient to the question from the user, and progresses the interview.

[0759] Step 4:

[0760] The server evaluates the results of the medical interview. The input is the medical interview result, and the output is the evaluation result. Specifically, the server analyzes the medical interview result and makes an appropriate evaluation. The evaluation includes the accuracy and appropriateness of the medical interview.

[0761] Step 5:

[0762] The server provides feedback based on the evaluation results. The input is the evaluation results, and the output is feedback information. Specifically, the server generates advice for improving the interview and skills based on the evaluation results.

[0763] Step 6:

[0764] The server provides feedback indicating the correlation between the interview results and the worker's emotional state. The input is the interview results and the worker's emotional state, and the output is feedback information indicating the correlation. Specifically, the server analyzes the effect of the worker's emotional state on the interview results and provides the results as feedback.

[0765] Step 7:

[0766] The user receives the provided feedback and uses it to improve their interview skills and emotional management. The input is feedback information, and the output is improvements to the user's behavior. Specifically, the user reviews and improves their own interview methods and emotional management methods based on the feedback.

[0767] 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.

[0768] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ) 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.

[0769] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0770] 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.

[0771] [Second embodiment]

[0772] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0773] 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.

[0774] 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).

[0775] 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.

[0776] 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.

[0777] 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).

[0778] 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.

[0779] 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.

[0780] 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.

[0781] 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.

[0782] 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.

[0783] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0784] "Example 1"

[0785] One aspect of the present invention is a virtual patient interview training system that utilizes generative AI. Specifically, the generative AI generates virtual patients with various symptoms and medical histories. A medical professional interviews these virtual patients, and the system evaluates the results. Based on the evaluation results, the system provides feedback to the medical professional.

[0786] "Example 2"

[0787] As a concrete example, generative AI generates a virtual patient with diabetes symptoms. A healthcare professional interviews the virtual patient, and the system evaluates the results. Based on the evaluation results, the system provides feedback to the healthcare professional on how to improve their interview and how to improve their diabetes interview skills.

[0788] "Example 3"

[0789] Generative AI can also generate virtual patients with rare conditions, such as specific genetic diseases, allowing patients to gain experience in medical interviews with conditions that would be difficult to experience in real life.

[0790] The processing flow of each embodiment will be described below.

[0791] "Example 1"

[0792] Step 1: Generative AI generates virtual patients with various symptoms and medical histories.

[0793] Step 2: A medical professional interviews the generated virtual patient.

[0794] Step 3: The system evaluates the interview results and provides feedback to the healthcare professional based on the evaluation results.

[0795] "Example 2"

[0796] Step 1: The generative AI generates a virtual patient with diabetes symptoms.

[0797] Step 2: A medical professional interviews the generated virtual patient.

[0798] Step 3: The system evaluates the interview results and provides feedback to healthcare professionals based on the evaluation results, indicating areas for improvement in the interview and how to improve their diabetes interview skills.

[0799] "Example 3"

[0800] Step 1: Generative AI generates virtual patients with specific genetic diseases.

[0801] Step 2: A medical professional interviews the generated virtual patient.

[0802] Step 3: The system evaluates the interview results and provides feedback to the healthcare professional based on the evaluation results.

[0803] Example 1

[0804] Next, a description will be given of Example 1 of Form 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."

[0805] Conventional medical interview training for medical professionals often involves the use of actual patients, which is time-consuming and costly, and training opportunities are limited. Furthermore, evaluation and feedback of medical interview skills tend to be subjective, making objective evaluation difficult. Furthermore, there is a lack of systems for efficiently improving medical interview skills for patients with various symptoms and medical histories.

[0806] 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.

[0807] In this invention, the server includes a means for conducting medical interview training with a virtual patient using generative artificial intelligence, a means for evaluating the results of the medical interview, a means for providing feedback based on the evaluation results, a means for providing an interface for displaying information about the virtual patient, a means for a user to input questions to the virtual patient and for the virtual patient to respond, a means for analyzing the content of the medical interview and evaluating the user's medical interview skills using an evaluation algorithm, and a means for providing the evaluation results as feedback to the user. This allows medical professionals to efficiently improve their medical interview skills without being restricted by time or location. Furthermore, receiving objective evaluations and specific feedback promotes improvement in medical interview skills.

[0808] "Generative AI" is AI that has the ability to generate new information and responses based on various data.

[0809] A "virtual patient" is a fictitious patient generated by a computer system, rather than a real patient.

[0810] "History training" refers to the practice of questions and responses that medical professionals ask patients.

[0811] The "interview results" are a record of questions asked by a medical professional to a virtual patient and their responses.

[0812] An "evaluation algorithm" is a calculation method for analyzing the results of a medical interview and objectively evaluating the medical interview skills of medical professionals.

[0813] "Feedback" refers to information on areas for improvement and advice provided based on the evaluation results.

[0814] An "interface" is an operating environment containing screens and input fields through which a user interacts with a system.

[0815] "User" refers to a medical professional who uses this system to conduct interview training.

[0816] The "means for inputting a question" refers to a text box or input field for the user to input a question to the virtual patient.

[0817] "Means for returning a response" refers to the function of the virtual patient to generate and display an appropriate response to a user's question.

[0818] The "means for analyzing the contents of the medical interview" refers to a function for analyzing the user's questions and the virtual patient's responses, and making an evaluation based on an evaluation algorithm.

[0819] The "means for providing evaluation results" refers to a function for providing the evaluation results obtained by the evaluation algorithm to the user as feedback.

[0820] This invention relates to a virtual patient interview training system that utilizes generative artificial intelligence. This system allows medical professionals to interview virtual patients, evaluate the results, and provide feedback to improve their interview skills.

[0821] System configuration

[0822] 1. Virtual patient generation

[0823] The server uses generative artificial intelligence (e.g., natural language processing models) to generate virtual patients with various symptoms and medical histories. The generated virtual patient information includes age, gender, chief complaint, medical history, current symptoms, etc.

[0824] 2. Providing a medical interview interface

[0825] The terminal provides a user (healthcare worker) with an interface that displays information about the virtual patient, including basic information about the virtual patient and input fields for conducting a medical interview.

[0826] 3. Conducting a medical interview

[0827] The user inputs questions to the virtual patient through a terminal, and the virtual patient uses generative artificial intelligence to provide appropriate responses.

[0828] 4. Evaluation of interview results

[0829] The server analyzes the user's interview content and evaluates the user's interview skills using an evaluation algorithm, including the appropriateness of the questions, the comprehension of the responses, and the flow of the interview.

[0830] 5. Providing Feedback

[0831] The device provides the evaluation results to the user as feedback, which specifically indicates areas for improvement and merits.

[0832] Hardware and software used

[0833] Server: A computer system with high-performance computing power that runs generative AI models and analyzes data.

[0834] Device: A computer, tablet, or smartphone operated by a user, providing an interview interface and accepting user input.

[0835] Generative artificial intelligence models: use natural language processing models (e.g., GPT-4) to generate responses for virtual patients.

[0836] Evaluation algorithm: An algorithm for analyzing the interview results and evaluating the user's interview skills.

[0837] Specific examples

[0838] Example scenario:

[0839] The hypothetical patient is a 35-year-old man who presents with recent headaches and dizziness. He has been previously diagnosed with high blood pressure and is currently taking medication.

[0840] Example prompt sentence:

[0841] "A 35-year-old male patient presents with recent headaches and dizziness. He has been diagnosed with high blood pressure in the past and is currently taking medication. What questions would you ask the patient?"

[0842] Based on this prompt, the user can ask specific questions to the virtual patient, such as "When did your headache start?" or "In what situations does dizziness occur?"

[0843] The server records these questions and the virtual patient's responses, and uses an evaluation algorithm to evaluate the user's interview skills. The evaluation results are fed back as feedback, such as "appropriateness of questions: 90 points," "understandability of responses: 85 points," and "flow of interview: 80 points."

[0844] In this way, users can improve their skills by interviewing virtual patients.

[0845] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0846] Step 1: Creating a virtual patient

[0847] The server generates a virtual patient using a generative AI model. As input, it receives a prompt statement (e.g., "A 35-year-old male patient complains of recent headaches and dizziness. He has been diagnosed with high blood pressure in the past and is currently taking medication."). Based on this prompt statement, the generative AI model generates detailed information about the virtual patient (e.g., age, gender, chief complaint, medical history, current symptoms, etc.). As output, it obtains information about the generated virtual patient.

[0848] Specific behavior:

[0849] The server inputs prompts into the generative artificial intelligence model to generate detailed information about the virtual patient, which is used for subsequent interview training.

[0850] Step 2: Providing a medical interview interface

[0851] The terminal provides the user with an interface that displays information about the virtual patient. As input, the terminal uses the information about the virtual patient received from the server. As output, an interface is displayed that allows the user to check the information about the virtual patient.

[0852] Specific behavior:

[0853] The terminal displays information about the virtual patient (age, gender, chief complaint, etc.) on the screen and provides a text box where the user can enter questions.

[0854] Step 3: Conduct a medical interview

[0855] The user inputs a question to the virtual patient through the terminal. The question entered by the user in the text box is used as input. The server sends this question to a generative artificial intelligence model, which generates a response from the virtual patient. The output is the response of the virtual patient.

[0856] Specific behavior:

[0857] The user enters "When did your headache start?" into a text box. The server sends this question to a generative artificial intelligence model, which generates the virtual patient's response: "My headache started a week ago." The device displays this response to the user.

[0858] Step 4: Evaluate the interview results

[0859] The server analyzes the user's interview content and evaluates the user's interview skills using an evaluation algorithm. The input is the user's questions and the virtual patient's responses. The output is the evaluation result.

[0860] Specific behavior:

[0861] The server analyzes the user's questions and the virtual patient's responses and applies an evaluation algorithm to generate results such as "appropriateness of question: 90 points," "understandability of response: 85 points," and "flow of medical interview: 80 points."

[0862] Step 5: Provide feedback

[0863] The terminal provides the evaluation results as feedback to the user. As input, it uses the evaluation results received from the server. As output, it displays feedback that allows the user to check the evaluation results.

[0864] Specific behavior:

[0865] The device displays the evaluation results on the screen, such as "Question appropriateness: 90 points," "Response comprehension: 85 points," and "Interview flow: 80 points," along with specific advice such as "You can get better results by changing the order of the questions."

[0866] (Application example 1)

[0867] Next, a description will be given of Application Example 1 of Form 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."

[0868] In conventional medical interview training systems, medical professionals have limited opportunities to interact with actual patients, making it difficult to provide a realistic training environment. Furthermore, evaluation and feedback of medical interview results are not provided immediately, which slows down the improvement of medical professionals' skills.

[0869] 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.

[0870] In this invention, the server includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the interview, a means for providing feedback based on the evaluation results, a means for a factory robot to act as the virtual patient and for a medical professional to interview the robot, a means for simulating virtual patients with various symptoms and medical histories using generative AI, and a means for evaluating the results of the interview in real time and providing feedback, thereby enabling medical professionals to improve their medical interview skills in a realistic training environment.

[0871] "Generative AI" is an artificial intelligence technology that generates new information and content based on various data.

[0872] A "virtual patient" is a patient that is not a real patient, but is simulated by a computer system.

[0873] "Medical history training" is training to improve medical professionals' techniques and skills in conducting medical interviews with patients.

[0874] "Interview results" refers to the content of the interview conducted by a medical professional with a patient and the information obtained as a result.

[0875] "Evaluation" refers to analyzing the medical professional's interview techniques and skills based on the results of the interview and judging whether they are good or bad.

[0876] "Feedback" refers to information such as areas for improvement and advice provided to medical professionals based on the evaluation results.

[0877] A "factory robot" is an automated mechanical device used in production sites such as factories.

[0878] "Simulate" means to imitate and reproduce actual situations or conditions.

[0879] "Real-time" means that processing and response are immediate and without delay.

[0880] A system for implementing this invention includes means for conducting medical interview training with a virtual patient using generative AI, means for evaluating the results of the interview, means for providing feedback based on the evaluation results, means for a factory robot acting as a virtual patient and for a medical professional to interview the robot, means for simulating virtual patients with various symptoms and medical histories using generative AI, and means for evaluating the results of the interview in real time and providing feedback.

[0881] The server generates a virtual patient using generative AI. The generated virtual patient is simulated through an application installed on a factory robot. Medical personnel interview the factory robot. The interview results are sent to the server in real time, and the server evaluates the interview results. Based on the evaluation results, the server provides feedback to the medical personnel.

[0882] The hardware used includes servers, factory robots, and devices used by medical professionals, while the software used includes generative AI models (e.g., OpenAI API) and algorithms for evaluating the results of medical interviews.

[0883] As a concrete example, a virtual patient can be generated by inputting the following prompt sentences into a generative AI model:

[0884] Virtual patient generation prompt:

[0885] "Generate virtual patients with a variety of symptoms and medical histories."

[0886] After the healthcare professional has completed the interview, they should evaluate the interview using the following prompts:

[0887] Interview assessment prompt:

[0888] Please rate the following questions:\nQuestion 1: What symptoms are you experiencing?\nAnswer 1: I have a headache.\nQuestion 2: How long have you had these symptoms?\nAnswer 2: About three days ago.

[0889] Based on the assessment, provide feedback using the following prompts:

[0890] Feedback prompt:

[0891] Please provide feedback based on the following assessment results:\n Assessment result: The healthcare professional asks appropriate questions and accurately understands the patient's symptoms.

[0892] This allows medical professionals to improve their medical history taking skills in a realistic training environment.

[0893] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0894] Step 1:

[0895] The server generates a virtual patient using a generative AI model. Specifically, the server inputs the following prompt sentences into the generative AI model:

[0896] "Generate virtual patients with a variety of symptoms and medical histories."

[0897] Based on this input, the generative AI model generates virtual patient data and outputs it to a server, including the virtual patient's symptoms and medical history.

[0898] Step 2:

[0899] The server sends the generated virtual patient data to the factory robot, which then acts as the virtual patient based on this data. Specifically, the factory robot simulates the virtual patient's symptoms and medical history and responds in real time to medical interviews with medical professionals.

[0900] Step 3:

[0901] The user (medical worker) conducts a medical interview with the factory robot. The contents of the interview are input by the user through a terminal. For example, the following questions are asked:

[0902] "What symptoms are you experiencing?"

[0903] The factory robot responds based on the virtual patient's data:

[0904] "I have a headache."

[0905] Step 4:

[0906] The terminal sends the details of the user's interview and the factory robot's response to the server, which receives this data and stores it as the interview results.

[0907] Step 5:

[0908] The server inputs the following prompt sentences into the generative AI model to evaluate the interview results:

[0909] Please rate the following questions:\nQuestion 1: What symptoms are you experiencing?\nAnswer 1: I have a headache.\nQuestion 2: How long have you had these symptoms?\nAnswer 2: About three days ago.

[0910] Based on this input, the generative AI model evaluates the interview results and outputs the evaluation results to the server.

[0911] Step 6:

[0912] The server inputs the following prompt sentences into the generative AI model to provide feedback based on the evaluation results:

[0913] Please provide feedback based on the following assessment results:\n Assessment result: The healthcare professional asks appropriate questions and accurately understands the patient's symptoms.

[0914] Based on this input, the generative AI model generates feedback and outputs it to the server.

[0915] Step 7:

[0916] The server transmits the generated feedback to the user's terminal, and the user can check the feedback through the terminal and use it to improve their interview skills.

[0917] Example 2

[0918] Next, a description will be given of Example 2 of Form 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."

[0919] Conventional medical interview training systems have had the problem of being unable to sufficiently improve medical personnel's medical interview skills due to the difficulty of simulating conversations with actual patients. In addition, evaluation of medical interview results and provision of feedback are often done manually, making efficient training difficult.

[0920] 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.

[0921] In this invention, the server includes means for conducting medical interview training with a virtual patient using generative artificial intelligence, means for evaluating the results of the interview, means for providing feedback based on the evaluation results, means for using prompt sentences to generate the virtual patient, means for transmitting the results of the interview to the server, and means for displaying the feedback on a terminal. This allows medical professionals to improve their skills through interviews with realistic virtual patients, thereby achieving efficient and effective training.

[0922] "Generative AI" is an AI technology that uses natural language processing and machine learning to generate data and information based on specific conditions.

[0923] A "virtual patient" is a fictional character simulated by generative artificial intelligence with symptoms and medical history similar to those of a real patient.

[0924] "Medical history training" involves medical professionals asking questions of virtual patients and determining diagnoses and treatment plans based on the answers.

[0925] The "interview results" are a record of questions asked by medical professionals to a virtual patient and the virtual patient's answers to those questions.

[0926] "Evaluation means" refers to algorithms or programs for analyzing the results of the interview and evaluating their content.

[0927] "Feedback" refers to advice provided to medical professionals on areas for improvement and skill development based on the evaluation results obtained through the evaluation tool.

[0928] A "prompt sentence" is a sentence used to input specific conditions or requests to generative artificial intelligence.

[0929] A "server" is a computer system for executing programs such as generative artificial intelligence, evaluation means, and feedback generation modules.

[0930] A "terminal" is a device such as a computer or tablet that medical professionals use to conduct interview training.

[0931] This invention is a system that utilizes generative artificial intelligence to conduct interview training with virtual patients, evaluate the interview results, and provide feedback based on the evaluation results. Specific embodiments of this system are described below.

[0932] Hardware and software used

[0933] Hardware: Servers, devices (PCs, tablets, etc.)

[0934] Software: Generative AI models (e.g., GPT-4), evaluation algorithms, feedback generation modules

[0935] Virtual patient generation

[0936] The server uses the generative AI model to generate a virtual patient with diabetes symptoms. Specifically, the server inputs the following prompt sentence into the generative AI model:

[0937] Example prompt: "Generate a hypothetical patient who is a 50-year-old male with a BMI of 30, HbA1c of 7.5%, and complains of frequent urination and dry mouth."

[0938] Based on this prompt, the generative AI model generates a detailed virtual patient profile, such as a "50-year-old male, BMI 30, HbA1c 7.5%, complaining of frequent urination and dry mouth" with specific symptoms and medical history.

[0939] Conducting medical interviews

[0940] The user (healthcare worker) uses a terminal to interview the generated virtual patient. The user inputs questions such as the following through the terminal interface.

[0941] "How's your diet been lately?"

[0942] "Do you have an exercise routine?"

[0943] The virtual patient provides answers to these questions through a generative AI model, which the user can use to ask more detailed questions.

[0944] Sending and evaluating medical interview results

[0945] The device sends the results of the medical interview entered by the user to the server, which then analyzes the results using an evaluation algorithm. For example, the server may make an evaluation such as, "There are not enough questions about diet" or "More detailed questions about exercise habits are needed."

[0946] Generating and Providing Feedback

[0947] The server generates feedback for the healthcare professional using the feedback generation module based on the evaluation results. Specifically, the following feedback is generated:

[0948] "It would be good to ask more questions about the diet."

[0949] "By asking specific questions about exercise habits, we can get a better understanding of the patient's lifestyle."

[0950] The device then displays the received feedback to the user, allowing the user to receive specific advice on how to improve their interview skills.

[0951] This system is an effective tool for medical professionals to improve their medical interview skills through virtual patients. The server uses a generative AI model to generate realistic virtual patients, evaluates the interview results, and provides feedback to help medical professionals improve their skills.

[0952] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0953] Step 1:

[0954] Virtual patient generation

[0955] The server uses the generative AI model to generate a virtual patient with diabetes symptoms.

[0956] Input: Prompt: "Create a virtual patient who is a 50-year-old male with a BMI of 30, HbA1c of 7.5%, and complains of frequent urination and dry mouth."

[0957] Data processing: The server inputs the prompt sentences into a generative AI model to generate a detailed profile of the virtual patient.

[0958] Output: Virtual patient profile (e.g., "50-year-old male, BMI 30, HbA1c 7.5%, frequent urination, dry mouth")

[0959] Step 2:

[0960] Conducting medical interviews

[0961] The user (medical worker) uses a terminal to interview the generated virtual patient.

[0962] Input: Questions entered by the user through the device interface (e.g., "How is your diet these days?", "Do you exercise regularly?")

[0963] Data processing: The virtual patient generates answers to these questions through a generative AI model.

[0964] Output: The hypothetical patient's response (e.g., "My diet is high in calories and I barely exercise.")

[0965] Step 3:

[0966] Sending medical interview results

[0967] The terminal transmits the results of the medical interview entered by the user to the server.

[0968] Input: User-entered questions and virtual patient answers

[0969] Data processing: The device converts all of this data into a format that can be sent to the server.

[0970] Output: Interview result data sent to the server

[0971] Step 4:

[0972] Evaluation of interview results

[0973] The server analyzes the received interview results using an evaluation algorithm.

[0974] Input: Submitted medical interview result data

[0975] Data processing: The server uses an evaluation algorithm to analyze the interview results and identify areas for improvement and shortcomings.

[0976] Output: Evaluation results (e.g., "There are not enough questions about diet," "More detailed questions about exercise habits are needed")

[0977] Step 5:

[0978] Generating and Providing Feedback

[0979] The server generates feedback for the healthcare professional based on the evaluation results using a feedback generation module.

[0980] Input: Evaluation result

[0981] Data processing: The server uses a feedback generation module to generate specific improvements and advice for skill development.

[0982] Output: Feedback (e.g., "It would be good to ask more questions about diet," "Ask specific questions about exercise habits to get a better understanding of the patient's lifestyle")

[0983] Step 6:

[0984] View Feedback

[0985] The terminal displays the received feedback to the user.

[0986] Input: Feedback sent by the server

[0987] Data processing: The device displays the feedback in a user-friendly format.

[0988] Output: Feedback displayed to the user

[0989] In this way, healthcare professionals can receive specific advice on how to improve their interview skills through virtual patients.

[0990] (Application example 2)

[0991] Next, a description will be given of Application Example 2 of Form 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."

[0992] Conventional training systems for medical professionals make it difficult to practice medical interviews with actual patients, and often do not provide sufficient evaluation or feedback on the results of the interviews. This has led to problems such as delays in improving medical interview skills. Furthermore, there has been a lack of appropriate systems for providing health management training to operators and safety managers working on-site at factories and other facilities. To solve these issues, an effective training system using virtual patients is needed.

[0993] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0994] In this invention, the server includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the medical interview, a means for providing feedback based on the evaluation results, a means for interacting with the virtual patient using a smart device, and a means for generating the virtual patient and evaluating the results of the medical interview via a cloud service. This enables medical professionals and factory operators to effectively improve their medical interview skills and health management skills without having to meet face-to-face with actual patients or workers.

[0995] "Generative AI" refers to artificial intelligence techniques used to generate virtual patients and scenarios.

[0996] "Medical history training" refers to practice in which medical professionals or operators conduct medical history interviews with virtual patients.

[0997] "Evaluation of interview results" is the process of analyzing the results of an interview with a virtual patient and evaluating their quality.

[0998] "Feedback" is information that provides areas for improvement and ways to improve skills based on the evaluation of the interview results.

[0999] "Smart devices" refers to interactive electronic devices such as smartphones, smart glasses, and head-mounted displays.

[1000] "Interaction" refers to the act of a user interacting with a virtual patient.

[1001] "Cloud services" refers to computing resources and services provided over the Internet.

[1002] A system for implementing this invention includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the interview, a means for providing feedback based on the evaluation results, a means for interacting with the virtual patient using a smart device, and a means for generating the virtual patient and evaluating the results of the interview via a cloud service.

[1003] System configuration

[1004] 1. Generative AI

[1005] The server generates virtual patients using a generative AI. The generative AI uses pre-trained models to generate virtual patients with various symptoms and medical histories. For example, to generate a virtual patient with diabetes, the following prompt sentence is used:

[1006] Generate a virtual patient with diabetes symptoms.

[1007] 2. Medical interview training

[1008] The user interacts with the virtual patient using a smart device (e.g., smart glasses), conducts a medical interview with the virtual patient, and inputs the results into the system.

[1009] 3. Evaluation of interview results

[1010] The server uses generative AI to evaluate the questionnaire results entered by the user. An example of an evaluation prompt is as follows:

[1011] Please rate the results of the following interview.

[1012] Interview results: [Virtual patient information]

[1013] evaluation:

[1014] 4. Providing Feedback

[1015] The server generates feedback based on the evaluation results. An example of a feedback prompt is:

[1016] Based on the evaluation below, please indicate areas for improvement in your interview and how you can improve your skills.

[1017] Evaluation: [Evaluation result]

[1018] feedback:

[1019] The generated feedback is displayed on the user's smart device.

[1020] Hardware and software used

[1021] Hardware: Smart devices (smart glasses, smartphones, head-mounted displays, etc.)

[1022] Software: Generative AI models (e.g., OpenAI API), cloud services (e.g., AWS, Google Cloud)

[1023] Specific examples

[1024] For example, consider a scenario in which a factory operator wears smart glasses and interviews a virtual patient. In this case, the operator checks the virtual patient's information through the smart glasses and proceeds with the interview. The interview results are sent to a server via a cloud service, where they are evaluated by generative AI. Based on the evaluation results, the operator can receive feedback on areas where and how to improve their interview skills.

[1025] In this way, medical professionals and factory operators can effectively improve their medical history and health management skills without having to meet actual patients or workers face-to-face.

[1026] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1027] Step 1:

[1028] The server generates a virtual patient using generative AI. The prompt sentence "Please generate a virtual patient with diabetes symptoms" is used as input. The generative AI model generates data for the virtual patient based on this prompt sentence and provides detailed information about the virtual patient as output.

[1029] Step 2:

[1030] The user interacts with the virtual patient using a smart device (e.g., smart glasses). The user interviews the virtual patient and inputs the results into the smart device. The input includes the interview content the user conducted with the virtual patient. The output is the interview results stored in the smart device.

[1031] Step 3:

[1032] The terminal sends the medical interview results to the server via the cloud service. The input includes the medical interview results entered by the user into the smart device. The output is the medical interview results sent to the server.

[1033] Step 4:

[1034] The server uses a generative AI to evaluate the received medical interview results. As input, it uses the medical interview results and the evaluation prompt "Please rate the following medical interview results.\n\nMedical interview results: [Virtual patient information]\n\nEvaluation:". The generative AI model evaluates the medical interview results based on this prompt and provides the evaluation result as output.

[1035] Step 5:

[1036] The server generates feedback based on the evaluation results. As input, it uses the evaluation results and a feedback prompt: "Based on the evaluation below, please indicate areas for improvement in the interview and ways to improve your skills.\n\nEvaluation: [Evaluation result]\n\nFeedback:". The generative AI model generates feedback based on this prompt and provides the feedback information as output.

[1037] Step 6:

[1038] The server sends the generated feedback to the user's smart device through a cloud service. As an input, the feedback information is included. As an output, the feedback information is displayed on the user's smart device.

[1039] Step 7:

[1040] The user can check the feedback through their smart device and use it to improve their medical interview and health management skills. The input includes feedback information sent from the server. The output is the user receiving the feedback and reflecting it in their actual training.

[1041] Example 3

[1042] Next, a description will be given of Example 3 of Form Example 3. 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."

[1043] With conventional medical interview training systems, it was difficult to gain experience interviewing real patients, especially patients with rare conditions or specific genetic diseases. Furthermore, there was a problem that the improvement of medical interview skills was delayed because the evaluation and feedback of the interview results was not sufficiently carried out.

[1044] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1045] In this invention, the server includes means for a user to input a prompt sentence, means for a terminal to send the prompt sentence to the server, means for the server to receive the prompt sentence and send it to the generative AI model, means for the generative AI model to generate medical condition data of the virtual patient, means for the server to receive the generated medical condition data and return it to the terminal, and means for the terminal to display the medical condition data to the user. This allows the user to gain experience in interviewing virtual patients with rare conditions or specific genetic diseases, and improves their interview skills through evaluation of the interview results and feedback.

[1046] "Generative AI" is a system that uses artificial intelligence technology to generate data and information based on specific conditions.

[1047] A "virtual patient" is not a real patient, but a fictitious patient generated on a computer system.

[1048] "Medical history training" is training for medical professionals to understand the patient's condition and make a diagnosis through dialogue with the patient.

[1049] A "prompt statement" is an instruction statement entered into a generative AI to cause it to generate specific data or information.

[1050] A "terminal" is a device such as a computer or smartphone operated by a user.

[1051] A "server" is a computer system that processes and stores data over a network.

[1052] A "generative AI model" is an algorithm or program that forms the core of generative AI and generates data based on input prompts.

[1053] "Medical condition data" is data that includes information about a patient's symptoms and medical history.

[1054] "Feedback" refers to information such as areas for improvement and advice provided based on the results of the interview.

[1055] MODE FOR CARRYING OUT THE INVENTION

[1056] This invention is a system for conducting interview training with virtual patients using generative AI. The system involves a series of processes: a user inputs a prompt sentence, a generative AI model generates medical condition data for the virtual patient, the interview results are evaluated, and feedback is provided.

[1057] Hardware and software used

[1058] The system uses the following hardware and software:

[1059] Device: A device such as a computer or smartphone operated by a user.

[1060] Server: A computer system that processes and stores data over a network.

[1061] Generative AI model: The core algorithm or program of generative AI. For example, a model that uses natural language processing technology (e.g., GPT-4).

[1062] Specific operation of the system

[1063] The user accesses the terminal interface and inputs a prompt to generate a medical condition for a virtual patient with a specific genetic disease, for example, "Please generate a medical condition for a virtual patient with Fabry disease. The patient is a 30-year-old male whose main symptoms are pain and rash in the hands and feet."

[1064] The device sends the prompt text entered by the user to the server. The server receives the prompt text sent from the device and sends it to the generative AI model. The generative AI model generates medical condition data for the virtual patient based on the prompt text.

[1065] The generated medical condition data is sent back to the server, which then sends the data back to the terminal. The terminal displays the medical condition data received from the server to the user. The user can check the displayed medical condition data and gain experience in interviewing the virtual patient.

[1066] Specific examples

[1067] For example, if a user wants to generate the symptoms of a virtual patient with a specific genetic disease, Fabry disease, he or she would enter the following prompt into the terminal:

[1068] "Generate the symptoms of a hypothetical patient with Fabry disease. The patient is a 30-year-old man whose main symptoms are pain in the hands and feet and a rash."

[1069] The server receives this prompt and sends it to the generative AI model, which generates the following medical condition data:

[1070] "A 30-year-old male patient suffers from Fabry disease, with the main symptoms being severe pain in the hands and feet and a red rash. In addition, the patient complains of frequent fatigue and gastrointestinal upset. Blood tests have confirmed that he has reduced α-galactosidase A activity."

[1071] In this way, the user can gain experience in interviewing patients with conditions that would be difficult to experience in real life. Furthermore, through evaluation of the interview results and feedback, the user's interview skills can be improved. The flow of the identification process in the third embodiment will be described with reference to FIG. 15.

[1072] Program processing flow

[1073] Step 1: The user enters a prompt statement

[1074] The user accesses the terminal interface and inputs a prompt to generate a medical condition for a virtual patient with a specific genetic disease, for example, "Please generate a medical condition for a virtual patient with Fabry disease. The patient is a 30-year-old male whose main symptoms are pain and rash in the hands and feet."

[1075] Input: The prompt text that the user enters

[1076] Output: The prompt text displayed on the terminal

[1077] Step 2: The terminal sends the prompt to the server

[1078] The terminal sends the prompt text entered by the user to the server, converting it into a suitable format so that the server can understand it.

[1079] Input: The prompt text entered by the user

[1080] Output: The prompt sent to the server

[1081] Step 3: The server receives the prompt and sends it to the generative AI model

[1082] The server receives the prompt sentence sent from the terminal. Then, the server sends this prompt sentence to the generative AI model. The server performs the necessary preprocessing so that the generative AI model can process the prompt sentence.

[1083] Input: Prompt sent from the terminal

[1084] Output: The prompt sent to the generative AI model

[1085] Step 4: The generative AI model generates the virtual patient's medical condition data

[1086] The generative AI model generates medical condition data for virtual patients based on prompt sentences sent from the server. The model uses natural language processing technology to create data for virtual patients with detailed medical conditions related to the specified genetic disease.

[1087] Input: Prompt text sent from the server

[1088] Output: Medical condition data of the generated virtual patient

[1089] Step 5: The server receives the generated medical condition data and sends it back to the device.

[1090] The server receives the medical condition data generated by the generative AI model, then sends this data back to the device, where it converts the data into a format that the device can display.

[1091] Input: Medical condition data sent from the generative AI model

[1092] Output: Medical condition data returned to the terminal

[1093] Step 6: The device displays the medical condition data to the user

[1094] The terminal displays the medical condition data received from the server to the user, who can check the displayed medical condition data and gain experience in interviewing the virtual patient.

[1095] Input: Medical condition data sent from the server

[1096] Output: Medical condition data displayed to the user

[1097] (Application example 3)

[1098] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1099] Conventional medical interview training for healthcare professionals is based on common symptoms and medical histories, making it difficult to gain experience interviewing patients with rare conditions. Training using actual patients also poses ethical issues and resource constraints, making it difficult to provide sufficient training opportunities. Furthermore, evaluation and feedback of interview results is insufficient, preventing healthcare professionals from efficiently improving their skills.

[1100] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1101] In this invention, the server includes: means for conducting medical interview training with a virtual patient using generative AI; means for evaluating the results of the medical interview; means for providing feedback based on the evaluation results; means for including a simulator that is installed on a factory robot and enables medical professionals to interact with virtual patients with rare medical conditions and gain medical interview experience; and means for generating the virtual patient's medical condition in real time using a generative AI model and providing appropriate responses to questions from the medical professionals. This allows medical professionals to gain experience in medical interviews with virtual patients with rare medical conditions, efficiently improving their medical interview skills.

[1102] "Generative AI" is a system that uses artificial intelligence technology to generate new data and information.

[1103] A "virtual patient" is a patient that does not actually exist but is recreated through computer simulation.

[1104] "Medical history training" is training to improve medical professionals' skills in listening to patients' symptoms and medical history.

[1105] "Interview results" is a general term for information obtained from patients by medical professionals.

[1106] "Evaluation" means analyzing the skills and responses of medical professionals based on the results of the medical interview and judging whether they are good or bad.

[1107] "Feedback" means providing suggestions for improvement or advice based on the evaluation results.

[1108] A "factory robot" is an automated mechanical device designed to perform tasks in a factory.

[1109] A "simulator" is a device or software that simulates a real-world situation.

[1110] A "generative AI model" is an algorithm or program that forms the core of generative AI and is used to generate new data and information.

[1111] "Real-time" refers to immediate processing or response without delay.

[1112] "Disease condition" refers to the symptoms or disease state exhibited by a patient.

[1113] A "question" is a question that a medical professional asks a patient to obtain information.

[1114] A "response" is an answer or reaction to a question.

[1115] A system for implementing this invention includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the medical interview, a means for providing feedback based on the evaluation results, a means for including a simulator that is installed on a factory robot and allows medical professionals to interact with virtual patients with rare medical conditions and gain medical interview experience, and a means for generating the virtual patient's medical condition in real time using a generative AI model and providing appropriate responses to questions from medical professionals.

[1116] Hardware and software used

[1117] Hardware: Factory robots

[1118] Software: OpenAI API, Python program

[1119] System processing overview

[1120] 1. Data entry: Medical professionals input questions into the robot, which then uses voice recognition technology to convert the questions into text data.

[1121] 2. Prompt generation: Generates a prompt based on the input question. The prompt is input data that allows the generative AI model to generate an appropriate response.

[1122] 3. Invoke the generative AI model: Using the OpenAI API, send prompts to the generative AI model (e.g., GPT-3) to generate a response for the virtual patient.

[1123] 4. Response display: The generated response is displayed to the medical staff. The robot outputs the response aloud using speech synthesis technology.

[1124] Specific examples

[1125] When a healthcare professional asks, "What are your recent symptoms?", the robot converts the question into text data and generates the following prompt:

[1126] Prompt Sentence Examples

[1127] Healthcare professional questions: What are your recent symptoms?

[1128] Virtual patient response:

[1129] Based on this prompt, the generative AI model generates the virtual patient's response, "I've been having frequent headaches lately, especially when I wake up in the morning." The robot then relays this response to the healthcare professional via voice.

[1130] Data processing and calculation

[1131] Speech recognition: Converting healthcare professional questions from speech to text.

[1132] Prompt generation: Generate a prompt based on the question text.

[1133] Invoke the generative AI model: Send the prompt to the generative AI model and generate a response.

[1134] Speech synthesis: Converts the generated response into speech and delivers it to the healthcare professional.

[1135] This system allows medical professionals to gain experience in interviewing virtual patients with rare medical conditions, efficiently improving their interview skills.

[1136] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1137] Step 1:

[1138] The user (healthcare worker) inputs a question to the robot by voice, and the robot converts the question into text data using voice recognition technology.

[1139] Input: Voice question from healthcare professional

[1140] Output: Question data in text format

[1141] Step 2:

[1142] The server receives textual question data and generates prompts, which are input data for the generative AI model to generate appropriate responses.

[1143] Input: Question data in text format

[1144] Output: prompt statement

[1145] Step 3:

[1146] The server sends the generated prompt sentences to the generative AI model using the OpenAI API to generate responses for the virtual patient. The generative AI model generates appropriate responses based on the prompt sentences.

[1147] Input: prompt statement

[1148] Output: Virtual patient response data

[1149] Step 4:

[1150] The server receives the generated virtual patient response data and sends it to the robot, which uses speech synthesis technology to convert the response into voice and convey it to medical professionals.

[1151] Input: Virtual patient response data

[1152] Output: Audio response

[1153] Step 5:

[1154] The server evaluates the results of the interview and provides feedback based on the evaluation results, including information indicating how to improve the interview results and how to improve the interview skills.

[1155] Input: Medical interview result data

[1156] Output: Feedback information

[1157] In this way, medical professionals can gain experience in interviewing virtual patients with rare conditions, and their interview skills can be improved efficiently.

[1158] 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.

[1159] "Example 1"

[1160] One embodiment of the present invention is a system in which generative AI includes an emotion engine that recognizes a user's emotions. In this system, as the user interviews a virtual patient, the emotion engine recognizes the user's emotional state in real time. Specifically, the emotion engine infers the user's emotional state from the user's tone of voice, facial expression, and choice of words. The system then adjusts the virtual patient's response based on the user's emotional state. For example, if the user feels irritated, the virtual patient will respond more considerately.

[1161] "Example 2"

[1162] In another embodiment of the present invention, the system includes an emotion engine that adjusts the virtual patient's response according to the user's emotional state. In this system, the emotion engine recognizes the user's emotional state and adjusts the virtual patient's response according to the state. For example, if the user feels depressed, the virtual patient will choose more encouraging words.

[1163] "Example 3"

[1164] Furthermore, in another embodiment of the present invention, there is a system in which the feedback includes information indicating the correlation between the user's emotional state and the interview result. This system provides feedback on the correlation between the user's emotional state and the interview result, along with an evaluation of the interview result. For example, if the user felt anger or anxiety, information indicating how the emotion affected the interview result is provided. This allows the user to understand the importance of managing their own emotions and identify areas for improvement to conduct better interviews.

[1165] The processing flow of each embodiment will be described below.

[1166] "Example 1"

[1167] Step 1: The user starts the system and begins an interview with the virtual patient.

[1168] Step 2: The emotion engine recognizes the user's emotional state in real time from their tone of voice, facial expressions, and choice of words.

[1169] Step 3: The emotion engine adjusts the virtual patient's response depending on the perceived emotional state. For example, if the user feels irritated, the virtual patient will react more considerately.

[1170] "Example 2"

[1171] Step 1: The user starts the system and begins an interview with the virtual patient.

[1172] Step 2: The emotion engine recognizes the user's emotional state in real time.

[1173] Step 3: The emotion engine adjusts the virtual patient's response depending on the perceived emotional state. For example, if the user feels depressed, the virtual patient will choose more encouraging words.

[1174] "Example 3"

[1175] Step 1: The user starts the system and begins an interview with the virtual patient.

[1176] Step 2: Once the interview is complete, the system evaluates the results.

[1177] Step 3: The system provides feedback on the user's emotional state and its relevance to the results of the interview. For example, if the user felt angry or anxious, the system provides information on how those emotions influenced the results of the interview.

[1178] Example 1

[1179] Next, a description will be given of Example 1 of Form 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."

[1180] Conventional medical interview training systems do not provide feedback that takes into account the user's emotional state when medical professionals interview virtual patients, making it difficult to fully improve their ability to respond in real clinical situations. In addition, the virtual patient's reactions are fixed, making it difficult to respond flexibly according to the user's emotional state, making it difficult to provide a realistic medical interview experience.

[1181] 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.

[1182] In this invention, the server includes means for conducting medical interview training with a virtual patient using generative artificial intelligence, means for evaluating the results of the medical interview, means for providing feedback based on the evaluation results, means for recognizing the emotional state of the user in real time, and means for adjusting the response of the virtual patient according to the recognized emotional state. This enables a realistic medical interview experience that takes the emotional state of the user into consideration, thereby improving the response capabilities of medical professionals.

[1183] "Generative AI" is AI that has the ability to generate new data and information based on user input.

[1184] A "virtual patient" is a patient that is not a real patient, but is simulated by a computer system.

[1185] "History-taking training" involves medical professionals asking questions of virtual patients to gather information for diagnosis and treatment.

[1186] The "interview results" are the total of questions asked by the medical professional to the virtual patient and the answers given by the virtual patient to those questions.

[1187] "Evaluation" is the process of analyzing the results of the interview and determining their appropriateness and validity.

[1188] "Feedback" refers to information on areas for improvement and advice provided to healthcare professionals based on the evaluation results.

[1189] "Emotional state" indicates the user's psychological state and is estimated from the tone of voice, facial expression, choice of words, etc.

[1190] "Real-time" means processing or reacting immediately, without delay.

[1191] "Adjusting responses" means changing the virtual patient's responses and behavior depending on the user's emotional state.

[1192] This invention is a virtual patient interview training system that utilizes generative artificial intelligence. This system allows medical professionals to interview virtual patients, evaluate the results, and provide feedback. Furthermore, it has the ability to recognize the user's emotional state in real time and adjust the virtual patient's responses accordingly.

[1193] Hardware and software used

[1194] Hardware: High-performance servers, devices used by users (PCs, tablets, smartphones)

[1195] Software: Generative AI models (e.g., GPT-4), emotion engines, speech recognition software

[1196] Specific operation of the system

[1197] Virtual patient generation

[1198] The server uses a generative AI model to generate virtual patients with various symptoms and medical histories. For example, by entering a prompt such as "Please generate a virtual patient who is a 40-year-old male with a history of smoking and whose recent symptoms are cough and chest pain," the server generates data for the virtual patient.

[1199] Beginning of medical interview

[1200] The user uses the device to initiate a consultation with the virtual patient. The device uses voice recognition software to convert the user's question into text data. For example, if the user asks, "What are your recent symptoms?", the device converts this into text data and sends it to the server.

[1201] Emotion recognition

[1202] The device uses an emotion engine to recognize the user's emotional state in real time from their tone of voice, facial expressions, and choice of words. For example, if the user is irritated, the emotion engine will recognize this as "irritation" and send that data to the server.

[1203] Virtual patient response adjustment

[1204] The server adjusts the virtual patient's response based on the output of the emotion engine. For example, if the server detects that the user is irritated, it instructs the virtual patient to respond with a considerate response, such as, "Can you tell me more?"

[1205] Evaluation of interview results

[1206] The server evaluates the content of the medical interview and analyzes the appropriateness of the user's questions and the quality of the response. For example, it may evaluate the user's questions as appropriate and improve their understanding of the patient's symptoms.

[1207] Providing feedback

[1208] The server provides feedback to the user based on the evaluation results, such as "Next time, please ask questions with more consideration for the patient's feelings."

[1209] This completes the embodiment of the present invention. This system enables a realistic medical interview experience that takes into account the user's emotional state, thereby improving the response capabilities of medical professionals.

[1210] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1211] Step 1:

[1212] Virtual patient generation

[1213] The server uses a generative AI model to generate a virtual patient by inputting a prompt sentence.

[1214] Input: Prompt: "Create a virtual patient: a 40-year-old male, with a history of smoking, whose current symptoms are cough and chest pain."

[1215] Data processing: A generative AI model analyzes the prompts and generates a detailed virtual patient profile and symptoms.

[1216] Output: Virtual patient data (e.g., age, gender, medical history, current symptoms).

[1217] Specific operation: The server stores the generated virtual patient data in a database.

[1218] Step 2:

[1219] Beginning of medical interview

[1220] The user uses the terminal to begin an interview with the virtual patient.

[1221] Input: A user spoken question (e.g., "What are your recent symptoms?").

[1222] Data processing: The device uses voice recognition software to convert the user's speech into text data.

[1223] Output: User's question converted to text data.

[1224] Specific operation: The device sends text data to the server.

[1225] Step 3:

[1226] Emotion recognition

[1227] The device uses an emotion engine to recognize the user's emotional state in real time from their tone of voice, facial expressions, and choice of words.

[1228] Input: User audio and video data.

[1229] Data processing: The emotion engine analyzes audio and video data to estimate the user's emotional state.

[1230] Output: User's emotional state (e.g., annoyed, calm).

[1231] Specific operation: The device sends the recognized emotion data to the server.

[1232] Step 4:

[1233] Virtual patient response adjustment

[1234] The server adjusts the virtual patient's response based on the output of the emotion engine.

[1235] Input: User emotional state data.

[1236] Data processing: The server adjusts the virtual patient's responses according to their emotional state using a generative AI model.

[1237] Output: Tailored virtual patient response (e.g., "Can you tell me more?").

[1238] Specific operation: The server sends the adjusted response to the terminal, and the virtual patient reacts appropriately to the user.

[1239] Step 5:

[1240] Evaluation of interview results

[1241] The server evaluates the contents of the medical interview and analyzes the appropriateness of the user's questions and the quality of the response.

[1242] Input: User questions and virtual patient response data.

[1243] Data processing: The server analyzes the interview content and uses an evaluation algorithm to determine its appropriateness and validity.

[1244] Output: Evaluation result (e.g., "The user's question was appropriate and provided a better understanding of the patient's condition").

[1245] Specific operation: The server stores the evaluation results in a database.

[1246] Step 6:

[1247] Providing Feedback

[1248] The server provides feedback to the user based on the evaluation results.

[1249] Input: Evaluation result data.

[1250] Data processing: The server generates improvements and advice based on the evaluation results.

[1251] Output: Feedback (e.g., "Next time, please be more considerate of the patient's feelings when asking questions").

[1252] Specific operation: The server sends the feedback content to the user's device.

[1253] (Application example 1)

[1254] Next, a description will be given of Application Example 1 of Form 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."

[1255] Conventional training systems provide uniform feedback without considering the user's emotional state, which means they are unable to respond to the stress and emotional fluctuations that occur in real work. In addition, the responses of virtual objects are fixed, which means users are unable to fully develop the ability to respond to diverse scenarios.

[1256] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means. In this invention, the server includes means for training with a virtual object using generative AI, means for evaluating the training results, means for providing feedback based on the evaluation results, emotion recognition means for recognizing the user's emotional state in real time, and means for adjusting the response of the virtual object according to the recognized emotional state. This makes it possible to provide appropriate feedback according to the user's emotional state and develop the ability to respond to a variety of scenarios.

[1257] "Generative AI" is an artificial intelligence technology that generates new information and scenarios from data.

[1258] A "virtual object" is an object that does not exist in reality but is simulated on a computer system.

[1259] "Training means" refers to a function that allows users to improve their skills through interaction with and manipulation of virtual objects.

[1260] The "means for evaluating training results" is a function for analyzing the content of the training performed by the user and evaluating the results.

[1261] The "means for providing feedback" is a function that provides the user with suggestions for improvement and advice based on the evaluation results.

[1262] The "emotion recognition means" is a function that recognizes the user's emotional state in real time from the user's tone of voice, facial expression, choice of words, etc.

[1263] The "means for adjusting the reaction of the virtual object" is a function that dynamically changes the reaction of the virtual object depending on the recognized emotional state of the user.

[1264] A "scenario" is a set of circumstances or events that a virtual object has.

[1265] "Skill improvement methods" are specific techniques and advice for users to improve specific abilities or techniques.

[1266] As an embodiment of the present invention, the following system is constructed.

[1267] The server includes a means for training with a virtual object using generative AI, a means for evaluating the training results, a means for providing feedback based on the evaluation results, an emotion recognition means for recognizing the user's emotional state in real time, and a means for adjusting the response of the virtual object according to the recognized emotional state.

[1268] Specifically, the server uses a generative AI model to generate virtual objects with various scenarios. Users interact with and train these virtual objects through smartphones, smart glasses, head-mounted displays, or robots. The server evaluates the training results and provides feedback based on the evaluation results.

[1269] The system uses a camera and microphone to recognize the user's emotional state in real time based on their tone of voice, facial expressions, and choice of words. The system uses the EmotionRecognizer library for emotion recognition. The virtual object's response is dynamically adjusted according to the recognized emotional state.

[1270] For example, if a user responds to a virtual robot by saying, "I'll check the robot's malfunction," the server evaluates the response using a generative AI model. If the server recognizes that the user is irritated, the virtual robot adjusts its response to respond calmly.

[1271] Examples of prompts include:

[1272] Scenario generation prompt: "Generate a scenario in which a factory robot breaks down."

[1273] Response evaluation prompt: "Evaluate the operator's response: Identify the faulty part of the robot."

[1274] In this way, it is possible to provide appropriate feedback according to the user's emotional state and develop the ability to respond to a variety of scenarios.

[1275] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1276] Step 1:

[1277] The server uses the generative AI model to generate a scenario for the virtual object. Specifically, the server inputs a prompt statement, "Please generate a scenario in the event that a factory robot breaks down," into the generative AI model and generates a scenario. The input is the prompt statement, and the output is the generated scenario.

[1278] Step 2:

[1279] The server transmits the generated scenario to a user terminal, which can be a smartphone, smart glasses, a head-mounted display, or a robot. The input is the generated scenario, and the output is the display of the scenario on the user terminal.

[1280] Step 3:

[1281] The user interacts with the virtual object and performs training. The user's response is input as voice or text. The input is the user's response, and the output is the transmission of the response data to the server.

[1282] Step 4:

[1283] The server uses a generative AI model to evaluate the user's response. Specifically, it inputs the prompt statement "Please evaluate the operator's response: Check the robot's fault location" into the generative AI model and obtains the evaluation result. The input is the user's response and the prompt statement, and the output is the evaluation result.

[1284] Step 5:

[1285] The server generates feedback based on the evaluation results and sends it to the user's device. The feedback includes points to improve the training results and ways to improve skills. The input is the evaluation results, and the output is the display of the feedback on the user's device.

[1286] Step 6:

[1287] The user device uses a camera and microphone to recognize the user's emotional state in real time. The EmotionRecognizer library is used for emotion recognition. The input is the user's tone of voice, facial expression, and word choice, and the output is the recognized emotional state.

[1288] Step 7:

[1289] The server adjusts the virtual object's response based on the recognized emotional state. For example, if the user is recognized as irritated, the virtual robot adjusts its response to respond calmly. The input is the recognized emotional state, and the output is the adjusted virtual object's response.

[1290] Step 8:

[1291] The server transmits the adjusted reaction of the virtual object to the user terminal, and the user terminal displays the adjusted reaction and provides feedback to the user. The input is the adjusted reaction of the virtual object, and the output is the display of the reaction to the user terminal.

[1292] Example 2

[1293] Next, a description will be given of Example 2 of Form 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."

[1294] Conventional medical interview training systems generate virtual patients and evaluate the results of the interviews, but they do not provide feedback based on the user's emotional state or adjust the virtual patient's reactions, which means they have limited applicability in actual medical settings. Furthermore, there is a lack of specific feedback on improving medical interview skills, which means that medical professionals are unable to fully improve their skills.

[1295] 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.

[1296] In this invention, the server includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the medical interview, a means for providing feedback based on the evaluation results, and a means for recognizing the emotional state of the user and adjusting the virtual patient's response according to that state. This makes it possible to provide feedback according to the user's emotional state and adjust the virtual patient's response, improving applicability in actual medical settings and improving the medical interview skills of medical professionals.

[1297] "Generative AI" is a system that uses artificial intelligence technology to generate new data or information based on specific conditions or prompts.

[1298] A "virtual patient" is a fictitious patient with specific symptoms and medical history, generated by generative AI.

[1299] "Medical history training" involves medical professionals asking questions of virtual patients and determining diagnoses and treatment plans based on the answers.

[1300] "Interview results" is a general term for questions asked by medical professionals to virtual patients and the answers given by the virtual patients to those questions.

[1301] "Evaluation" is the process of analyzing the results of the interview and determining their quality and appropriateness.

[1302] "Feedback" is information provided based on the evaluation results regarding areas for improvement in interviews and ways to improve skills.

[1303] "Emotional state" refers to the user's psychological state or mood, as recognized by the emotion engine.

[1304] The "emotion engine" is a system that analyzes the user's emotional state from facial expressions, voice, etc. and provides that information.

[1305] "Response modulation" is the process of modifying the responses and behavior of a virtual patient depending on the user's emotional state.

[1306] This invention is a virtual patient interview training system that utilizes generative AI, evaluates the interview results, provides feedback based on the evaluation results, and includes a function to recognize the user's emotional state and adjust the virtual patient's response accordingly.

[1307] Hardware and software used

[1308] Hardware: Servers, devices (PCs, tablets, smartphones)

[1309] Software: Generative AI models (e.g., GPT-4), emotion engines

[1310] Specific operation of the system

[1311] Virtual patient generation

[1312] The server uses the generative AI model to generate a virtual patient with diabetes symptoms. Specifically, the server inputs a prompt statement such as "50-year-old male, with a 5-year history of diabetes and recent elevated blood sugar levels" into the generative AI model, and generates a detailed profile of the virtual patient.

[1313] Conducting medical interviews

[1314] The user (healthcare professional) uses a terminal to interview the virtual patient. The user inputs questions to the virtual patient through the terminal interface, such as "How has your diet been recently?" The terminal then sends these questions to the server.

[1315] Evaluation of interview results

[1316] The server analyzes and evaluates the received medical interview content. Specifically, the server uses a generative AI model to analyze the virtual patient's answers and evaluate the quality of the interview. For example, if a virtual patient answers, "I've been trying to eat more vegetables lately," the server evaluates the answer and generates appropriate feedback.

[1317] Providing Feedback

[1318] Based on the evaluation results, the server generates feedback indicating areas for improvement in the interview and ways to improve diabetes interview skills. Specifically, the server generates feedback such as "There were not enough questions about dietary details, so next time you should ask about specific dietary details," and sends it to the device. The device then displays this feedback to the user.

[1319] Response adjustment by emotion engine

[1320] The server uses an emotion engine to recognize the user's emotional state. Specifically, if the server analyzes the data obtained from the emotion engine and recognizes that the user is depressed, it adjusts the virtual patient's response. For example, the virtual patient will choose encouraging words such as, "Your efforts are great. Keep it up."

[1321] Examples of specific examples and prompts

[1322] Specific examples

[1323] Generating a virtual patient: The server inputs prompt statements such as "50-year-old male, with a 5-year history of diabetes and recent elevated blood sugar levels" into a generative AI model to generate a detailed profile.

[1324] Conducting the medical interview: The user asks the question "How has your diet been lately?" through the terminal, and the terminal sends this question to the server.

[1325] Evaluation of the interview results: The server analyzes the virtual patient's answer, "I've been trying to eat more vegetables recently," and makes an evaluation.

[1326] Providing feedback: The server generates feedback such as "There were not enough questions about the meal contents, so next time you should ask about the specific meal contents," and sends it to the device.

[1327] Response adjustment by emotion engine: The server analyzes the data obtained from the emotion engine, and if it recognizes that the user is depressed, the virtual patient adjusts its response to choose encouraging words such as "Your efforts are great. Keep going."

[1328] Prompt Sentence Examples

[1329] "Generate a hypothetical patient who is a 50-year-old male with a 5-year history of diabetes and recent elevated blood glucose levels."

[1330] "Ask your virtual patient about their recent diet."

[1331] Please evaluate the results of the interview and provide feedback on areas for improvement.

[1332] "If the user is feeling down, tailor your response so that the virtual patient chooses encouraging words."

[1333] The above is an embodiment of the present invention.

[1334] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1335] Step 1:

[1336] Virtual patient generation

[1337] The server uses the generative AI model to generate a virtual patient with diabetes symptoms.

[1338] Input: A prompt such as "50 year old male with 5 years of diabetes, recent high blood sugar levels."

[1339] Data processing: The server inputs the prompt sentences into a generative AI model to generate a detailed profile of the virtual patient.

[1340] Output: A detailed profile of the virtual patient.

[1341] Specific operation: The server inputs prompt sentences into the generative AI model and stores the generated virtual patient profile in the database.

[1342] Step 2:

[1343] Conducting medical interviews

[1344] The user (medical worker) uses a terminal to interview the virtual patient.

[1345] Input: A question typed by the user through the device interface (e.g., "How has your diet been lately?").

[1346] Data processing: The device sends the user's question to the server.

[1347] Output: The question sent to the server.

[1348] Specific operation: The user inputs a question through the terminal, and the terminal sends the question to the server.

[1349] Step 3:

[1350] Evaluation of interview results

[1351] The server analyzes the received interview contents and performs an evaluation.

[1352] Input: The hypothetical patient's response (e.g., "I've been trying to eat more vegetables lately").

[1353] Data processing: The server uses a generative AI model to analyze the virtual patient's responses and evaluate the quality of the interview.

[1354] Output: Evaluation result.

[1355] Specific operation: The server analyzes the answers of the virtual patient and generates an evaluation result.

[1356] Step 4:

[1357] Providing Feedback

[1358] Based on the evaluation results, the server generates feedback indicating areas for improvement in the interview and ways to improve diabetes interview skills.

[1359] Input: Evaluation result.

[1360] Data processing: The server generates feedback based on the evaluation results and sends it to the device.

[1361] Output: Feedback (e.g., "You didn't ask enough questions about dietary habits. Next time, please ask more specifically about your dietary habits.").

[1362] Specific operation: The server generates feedback based on the evaluation results and sends it to the device, which then displays this feedback to the user.

[1363] Step 5:

[1364] Response adjustment by emotion engine

[1365] The server uses an emotion engine to recognize the user's emotional state.

[1366] Input: User emotional state data (e.g. facial expressions and voice data).

[1367] Data processing: The server analyzes the data obtained from the emotion engine and recognizes the user's emotional state.

[1368] Output: The perceived emotional state of the user.

[1369] Specific behavior: The server analyzes the data obtained from the emotion engine, and if it recognizes that the user is depressed, it adjusts the virtual patient's response. For example, the virtual patient will choose encouraging words such as "Your efforts are great. Keep going."

[1370] (Application example 2)

[1371] Next, a description will be given of Application Example 2 of Form 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."

[1372] Conventional medical interview training systems have the problem of being unable to adequately simulate the emotions and reactions of medical professionals when they meet with actual patients. Furthermore, the inability to provide real-time feedback or respond to emotional states limits the improvement of actual medical interview skills. Furthermore, there is a lack of support for conducting appropriate medical interviews with customers in brick-and-mortar stores.

[1373] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1374] In this invention, the server includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the medical interview, a means for providing feedback based on the evaluation results, a means for analyzing the emotional state of a user using an emotion recognition engine, a means for adjusting the response of the virtual patient according to the emotional state of the user, a means for capturing video in real time using smart glasses to support the medical interview, and a means for providing feedback. This enables medical professionals and store clerks to conduct appropriate medical interviews according to the emotional state of customers and patients and receive feedback in real time.

[1375] "Generative AI" is an artificial intelligence technology that generates virtual patients and simulates their symptoms and reactions.

[1376] "Medical history training" is training for medical professionals and store clerks to conduct medical history interviews with virtual patients and improve their skills.

[1377] The "means for evaluating the results of medical interviews" is a system for analyzing the content of medical interviews conducted by medical professionals or store clerks and evaluating their quality and appropriateness.

[1378] The "means for providing feedback" is a system that provides medical professionals and store clerks with advice on areas for improvement and skill development based on the evaluation of the interview results.

[1379] An "emotion recognition engine" is a technology that analyzes a user's facial expressions and tone of voice to recognize their emotional state.

[1380] The "means for analyzing the user's emotional state" is a system that uses an emotion recognition engine to analyze the user's emotions in real time.

[1381] A "means for adjusting the virtual patient's reactions" is a system for changing the virtual patient's reactions depending on the user's emotional state.

[1382] "Smart glasses" are wearable devices that have built-in cameras and displays and are worn by users.

[1383] The "means of capturing video in real time and supporting medical interviews" is a system that uses smart glasses to capture video of customers or patients in real time and supports medical interviews based on that information.

[1384] The system for implementing this invention integrates a generative AI, an emotion recognition engine, smart glasses, and a feedback system. The following describes the details of each component and its operation.

[1385] 1. Generative AI

[1386] The generative AI generates virtual patients and simulates their symptoms and reactions. The server uses the generative AI to generate virtual patients with various symptoms and medical histories, allowing medical professionals and store clerks to train in how to handle a variety of cases.

[1387] 2. Emotion Recognition Engine

[1388] The emotion recognition engine analyzes the user's facial expressions and tone of voice to recognize their emotional state. The server captures the user's video using the camera built into the smart glasses and analyzes it with the emotion recognition engine. This allows the server to grasp the user's emotional state in real time.

[1389] 3. Smart Glasses

[1390] Smart glasses are wearable devices equipped with a built-in camera and display that are worn by the user. The smart glasses capture video in real time and send the information to a server. The server then analyzes the received video using an emotion recognition engine to understand the user's emotional state.

[1391] 4. Feedback System

[1392] The feedback system provides advice to medical professionals and store clerks on areas for improvement and skill development based on the evaluation of the interview results. The server evaluates the interview results and generates feedback based on the evaluation results. The feedback is displayed on the display of the smart glasses.

[1393] Specific examples

[1394] For example, consider a situation where a store clerk is wearing smart glasses and a customer asks, "I've been feeling thirsty a lot lately. What should I do?" If the emotion recognition engine senses anxiety from the customer's facial expression, the virtual patient will select encouraging words such as, "That may be an early symptom of diabetes. Let's start by measuring your blood sugar level." The feedback system will provide feedback to the store clerk, such as, "It would be good to suggest specific actions to ease the customer's anxiety."

[1395] Prompt Sentence Examples

[1396] "If a client is consulting about their diabetes symptoms, what questions should they ask?"

[1397] In this way, medical professionals and store clerks can tailor their interviews to the emotional state of their customers or patients and receive real-time feedback.

[1398] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1399] Step 1:

[1400] The server generates a virtual patient using generative AI. As input, it receives data about the virtual patient's symptoms and medical history, and uses a generative AI model to generate a profile of the virtual patient. As output, it obtains a detailed profile of the virtual patient.

[1401] Step 2:

[1402] The user puts on the smart glasses and starts the medical interview. The camera in the smart glasses captures the video in real time and sends the video data to the server. The user's video data is used as input, and the video data is sent to the server as output.

[1403] Step 3:

[1404] The server analyzes the received video data using an emotion recognition engine. As input, it receives the user's video data and analyzes the user's emotional state using the emotion recognition engine. As output, it obtains data related to the user's emotional state.

[1405] Step 4:

[1406] The server adjusts the virtual patient's reactions according to the user's emotional state. As input, the server uses the user's emotional state data and the virtual patient profile to generate the virtual patient's reactions. As output, the server obtains the adjusted virtual patient's reactions.

[1407] Step 5:

[1408] The user interviews the virtual patient. The user inputs a question, and the question data is sent to the server. The user's question data is used as input, and the question data is sent to the server as output.

[1409] Step 6:

[1410] The server generates and provides the virtual patient's responses to the user. As input, the server uses the user's question data and the virtual patient's profile to generate the virtual patient's responses. As output, the virtual patient's response data is obtained and displayed on the smart glasses display.

[1411] Step 7:

[1412] The server evaluates the interview results and generates feedback. As input, it uses the user's question data and the virtual patient's response data, and performs the evaluation using a feedback system. As output, it obtains feedback data.

[1413] Step 8:

[1414] The server provides the generated feedback to the user. As an input, it uses the feedback data and displays the feedback on the display of the smart glasses. As an output, the user can receive the feedback.

[1415] In this way, medical professionals and store clerks can tailor their interviews to the emotional state of their customers or patients and receive real-time feedback.

[1416] Example 3

[1417] Next, a description will be given of Example 3 of Form Example 3. 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."

[1418] Conventional medical interview training systems make it difficult for users to gain experience interviewing patients with rare conditions, and lack a means to evaluate the impact of the user's emotional state on the interview results. This makes it difficult to improve users' medical interview skills.

[1419] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1420] In this invention, the server includes means for conducting medical interview training with a virtual patient using generative artificial intelligence, means for evaluating the results of the medical interview, means for providing feedback based on the evaluation results, means for recording the user's emotional state, means for analyzing the user's emotional state, and means for evaluating the correlation between the results of the medical interview and the user's emotional state. This allows the user to gain experience in medical interviews with virtual patients with rare conditions, further understanding the importance of managing their own emotions and improving their medical interview skills.

[1421] "Generative AI" is an AI technology that generates virtual patients and other information based on prompts entered by the user.

[1422] "Medical history training" is training in which a user asks questions to a virtual patient to understand the condition and symptoms.

[1423] The "results of medical interview" is a general term for the contents of the medical interview conducted by the user with the virtual patient and the answers thereto.

[1424] The "evaluation result" is an evaluation of the user's interview skills and other related information generated based on the interview result.

[1425] "Feedback" is information about areas and methods for improving interview skills that is provided to the user based on the evaluation results.

[1426] "Emotional state" refers to the emotional state the user feels during the interview, and includes, for example, anxiety, anger, joy, etc.

[1427] The "means for recording emotional state" refers to a device or software for recording the user's facial expressions, tone of voice, etc.

[1428] The "means for analyzing emotional state" refers to a device or software for analyzing the recorded emotional data of a user and identifying that emotional state.

[1429] The "means for evaluating the correlation between the questionnaire result and the emotional state" is a device or software for evaluating the correlation between the questionnaire result and the emotional state of the user.

[1430] This invention is a system that utilizes generative artificial intelligence to conduct medical interview training with virtual patients, evaluate the interview results, and provide feedback based on the evaluation results. Furthermore, it records and analyzes the user's emotional state, and evaluates the correlation between the interview results and the emotional state, thereby improving the user's medical interview skills.

[1431] Hardware and software used

[1432] Server: A server equipped with a high-performance GPU (e.g., a server equipped with a high-performance graphics processing unit) is used to run the generative artificial intelligence model.

[1433] Generative AI model: Uses a natural language processing model (e.g., an advanced natural language processing model) to generate a virtual patient's medical condition based on a prompt sentence.

[1434] Emotion Analysis Software: Uses an emotion analysis API (e.g., an application programming interface for emotion analysis) to analyze the user's emotional state.

[1435] Program processing

[1436] Virtual patient generation

[1437] The user inputs a prompt for the generative AI model through the terminal interface. For example, the user might input, "Please generate a virtual patient, a 30-year-old male with a specific genetic disease." The terminal then sends this prompt to the server. The server uses the generative AI model to generate the virtual patient's medical condition based on the prompt. Information about the generated virtual patient is sent to the terminal and displayed to the user.

[1438] Conducting medical interviews

[1439] The user interviews the virtual patient through the device. For example, the user might ask, "How are you feeling lately?", to which the virtual patient might respond, "Recently, my hands and feet have been shaking a lot, and I've been dropping things more easily." During the interview process, the device uses a camera and microphone to record the user's facial expressions and tone of voice in real time.

[1440] Sentiment analysis and feedback provision

[1441] Once the interview is complete, the server uses emotion analysis software to analyze the user's emotional state. The server evaluates the correlation between the interview results and the user's emotional state and generates feedback. For example, the server generates feedback such as, "Because you were feeling anxious, the questions were vague and we were unable to elicit accurate information," and sends it to the device. The user can review the feedback displayed on the device, understand the importance of managing their own emotions, and identify areas for improvement in their interview skills.

[1442] Specific examples

[1443] Example prompt: Create a hypothetical patient, a 30-year-old male with a specific genetic disease.

[1444] The virtual patient generated was a 30-year-old male with a specific genetic disorder, characterized by motor and cognitive impairment and mental fluctuations.

[1445] Conducting a medical interview: The user asks, "How are you feeling these days?" and the virtual patient answers, "Recently, my hands and feet have been shaking a lot, and I've been dropping things more easily."

[1446] Feedback: The user is informed that their anxiety led them to ask vague questions and fail to elicit specific symptoms from the virtual patient.

[1447] This system allows users to gain experience in interviewing virtual patients with rare medical conditions, and further enables them to understand the importance of managing their own emotions and improve their interview skills.The flow of the identification process in Example 3 will be described with reference to Figure 21.

[1448] Step 1:

[1449] The user enters a prompt statement.

[1450] The user inputs a prompt for the generative AI model through the terminal interface, for example, "Please generate a virtual patient, a 30-year-old male with a specific genetic disease." The input prompt is stored in the terminal's memory.

[1451] Step 2:

[1452] The terminal sends the prompt to the server.

[1453] The terminal sends the prompt text entered by the user to the server using the HTTPS protocol, and the sent prompt text is received by the server's receiving module.

[1454] Step 3:

[1455] The server generates a virtual patient using the generative AI model.

[1456] The server inputs the received prompt sentences into a generative AI model to generate a virtual patient's medical condition. The generated virtual patient information includes the patient's age, gender, medical history, current symptoms, etc. The generated data is stored in the server's database.

[1457] Step 4:

[1458] The server transmits information about the generated virtual patient to the terminal.

[1459] The server sends the generated virtual patient information in JSON format to the terminal, which then analyzes the received information and displays it to the user.

[1460] Step 5:

[1461] The user interviews the virtual patient.

[1462] The user asks the virtual patient questions through the terminal. For example, "How are you feeling lately?", and the virtual patient responds, "Recently, my hands and feet have been shaking a lot, and I've been dropping things more easily." The user's questions and the virtual patient's answers are recorded on the terminal.

[1463] Step 6:

[1464] The device records the user's emotional state.

[1465] The device uses a camera and microphone to record the user's facial expressions and tone of voice in real time, and the recorded data is stored in the device's memory.

[1466] Step 7:

[1467] The server performs emotion analysis and evaluates the correlation between the interview results and the emotional state.

[1468] The device sends the recorded emotional data to a server, which then uses emotion analysis software to analyze the user's emotional state. The analysis results are stored in the server's database. The server then evaluates the correlation between the interview results and the user's emotional state and generates an evaluation result.

[1469] Step 8:

[1470] The server generates feedback and sends it to the device.

[1471] The server generates feedback based on the evaluation results. For example, it might say, "Because you were feeling anxious, the questions were vague and we were unable to elicit accurate information," and sends it to the device. The device then displays the received feedback to the user.

[1472] Step 9:

[1473] The user checks the feedback.

[1474] Users can view feedback displayed on the device screen, which helps them understand the importance of managing their own emotions and identify areas for improvement in their interview skills.

[1475] (Application example 3)

[1476] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1477] In modern factories, it is important to monitor the health status of workers in real time and detect abnormalities early. However, it is difficult to take appropriate action quickly when an abnormality in a worker's health is detected. Another challenge is understanding the impact of a worker's emotional state on their health and work efficiency, and providing appropriate feedback. To solve these challenges, a system is needed that monitors the health status of workers, generates a virtual patient to conduct an interview when an abnormality is detected, and provides feedback showing the relationship between the emotional state and the interview results.

[1478] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes a means for conducting interview training with a virtual patient using generative AI, a means for evaluating the interview results, a means for providing feedback based on the evaluation results, a means for monitoring the worker's health condition, a means for generating a virtual patient and conducting an interview when an abnormality is detected, and a means for providing feedback indicating the correlation between the interview results and the worker's emotional state. This makes it possible to monitor the worker's health condition in real time, detect abnormalities early, and take appropriate measures promptly. It is also possible to understand the impact of the worker's emotional state on their health condition and work efficiency and provide appropriate feedback.

[1479] "Generative AI" is a system that uses artificial intelligence technology to generate new data and information.

[1480] A "virtual patient" is a computer-generated model of a patient that does not actually exist.

[1481] "Medical history training" is training to improve medical professionals' skills in interviewing patients.

[1482] "Interview results" refers to information and data obtained through interviews.

[1483] "Evaluation results" are the results of evaluations conducted based on the interview results.

[1484] "Feedback" refers to suggestions for improvement or advice provided based on the evaluation results.

[1485] "Worker" means a person who works in a factory or on a work site.

[1486] "Health status" refers to the physical and mental condition of a worker.

[1487] "Monitoring" means to continuously watch over a particular condition or situation.

[1488] An "abnormality" is a state or situation that is different from the normal.

[1489] "Emotional state" refers to the worker's current feelings and mood.

[1490] "Relationship" indicates that two or more events or data are related to each other.

[1491] The system for implementing this invention is configured as follows: First, the server uses generative AI to generate virtual patients and conduct medical interview training. The generated virtual patients can have a variety of symptoms and medical histories, allowing users to improve their medical interview skills in dealing with a variety of cases.

[1492] The server then evaluates the results of the interview and provides feedback based on the evaluation. The feedback includes suggestions for improving the interview results and ways to improve interview skills. Sensors are also installed to monitor the worker's health, and if an abnormality is detected, the server automatically generates a virtual patient and conducts an interview.

[1493] Furthermore, the server provides feedback indicating the correlation between the interview results and the worker's emotional state, allowing the worker to understand the impact of their own emotional state on their health and work efficiency and take appropriate measures.

[1494] Specifically, the server performs the following process. First, sensors (e.g., heart rate monitor, thermometer, blood pressure monitor, etc.) are used to monitor the worker's health condition in real time. If an abnormality is detected, the generative AI generates a virtual patient and begins a medical interview. The medical interview is conducted based on prompts entered by the user.

[1495] For example, the following prompt sentences are used:

[1496] Please describe your current emotional state (e.g., anxious, angry, happy): Anxiety

[1497] A virtual patient was created: {'disease': 'Genetic disease A', 'symptoms': ['Symptom 1', 'Symptom 2']}

[1498] How are you feeling these days? I'm not feeling well.

[1499] What symptoms are you experiencing? I have headaches and dizziness.

[1500] When did the symptoms start? Three days ago.

[1501] Emotional state: Anxiety

[1502] Interview results:

[1503] I'm not feeling well.

[1504] I have a headache and dizziness.

[1505] It started three days ago.

[1506] Consider how emotional state can affect interview results.

[1507] Based on the prompt, the server evaluates the interview results and provides feedback, including suggestions for improving the interview results and ways to improve interview skills. It also provides information on the correlation between the worker's emotional state and the interview results.

[1508] This system makes it possible to monitor workers' health conditions in real time, detect abnormalities early, and take appropriate measures promptly.It also understands the impact of workers' emotional states on their health and work efficiency, and provides appropriate feedback.

[1509] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1510] Step 1:

[1511] The server monitors the worker's health condition in real time. It collects data from sensors (e.g., heart rate monitors, thermometers, blood pressure monitors, etc.) and checks for any abnormalities. The input is health data from the sensors, and the output is the presence or absence of abnormalities. Specifically, the server periodically obtains data from the sensors and evaluates whether it is within the normal range.

[1512] Step 2:

[1513] When an abnormality is detected, the server uses generative AI to generate a virtual patient. The input is the result of the abnormality detection, and the output is the virtual patient data. Specifically, the server generates an appropriate virtual patient according to the type of abnormality and sets the patient's symptoms and medical history.

[1514] Step 3:

[1515] The server starts an interview with the generated virtual patient. The user inputs a prompt sentence, and the interview is conducted based on that. The input is the prompt sentence from the user, and the output is the interview result. In concrete terms, the server generates a response for the virtual patient to the question from the user, and progresses the interview.

[1516] Step 4:

[1517] The server evaluates the results of the medical interview. The input is the medical interview result, and the output is the evaluation result. Specifically, the server analyzes the medical interview result and makes an appropriate evaluation. The evaluation includes the accuracy and appropriateness of the medical interview.

[1518] Step 5:

[1519] The server provides feedback based on the evaluation results. The input is the evaluation results, and the output is feedback information. Specifically, the server generates advice for improving the interview and skills based on the evaluation results.

[1520] Step 6:

[1521] The server provides feedback indicating the correlation between the interview results and the worker's emotional state. The input is the interview results and the worker's emotional state, and the output is feedback information indicating the correlation. Specifically, the server analyzes the effect of the worker's emotional state on the interview results and provides the results as feedback.

[1522] Step 7:

[1523] The user receives the provided feedback and uses it to improve their interview skills and emotional management. The input is feedback information, and the output is improvements to the user's behavior. Specifically, the user reviews and improves their own interview methods and emotional management methods based on the feedback.

[1524] 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.

[1525] 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> ) 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.

[1526] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1527] 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.

[1528] [Third embodiment]

[1529] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1530] 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.

[1531] 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).

[1532] 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.

[1533] 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.

[1534] 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).

[1535] 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.

[1536] 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.

[1537] 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.

[1538] 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.

[1539] 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.

[1540] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1541] "Example 1"

[1542] One aspect of the present invention is a virtual patient interview training system that utilizes generative AI. Specifically, the generative AI generates virtual patients with various symptoms and medical histories. A medical professional interviews these virtual patients, and the system evaluates the results. Based on the evaluation results, the system provides feedback to the medical professional.

[1543] "Example 2"

[1544] As a concrete example, generative AI generates a virtual patient with diabetes symptoms. A healthcare professional interviews the virtual patient, and the system evaluates the results. Based on the evaluation results, the system provides feedback to the healthcare professional on how to improve their interview and how to improve their diabetes interview skills.

[1545] "Example 3"

[1546] Generative AI can also generate virtual patients with rare conditions, such as specific genetic diseases, allowing patients to gain experience in medical interviews with conditions that would be difficult to experience in real life.

[1547] The processing flow of each embodiment will be described below.

[1548] "Example 1"

[1549] Step 1: Generative AI generates virtual patients with various symptoms and medical histories.

[1550] Step 2: A medical professional interviews the generated virtual patient.

[1551] Step 3: The system evaluates the interview results and provides feedback to the healthcare professional based on the evaluation results.

[1552] "Example 2"

[1553] Step 1: The generative AI generates a virtual patient with diabetes symptoms.

[1554] Step 2: A medical professional interviews the generated virtual patient.

[1555] Step 3: The system evaluates the interview results and provides feedback to healthcare professionals based on the evaluation results, indicating areas for improvement in the interview and how to improve their diabetes interview skills.

[1556] "Example 3"

[1557] Step 1: Generative AI generates virtual patients with specific genetic diseases.

[1558] Step 2: A medical professional interviews the generated virtual patient.

[1559] Step 3: The system evaluates the interview results and provides feedback to the healthcare professional based on the evaluation results.

[1560] Example 1

[1561] Next, a description will be given of Example 1 of Form 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."

[1562] Conventional medical interview training for medical professionals often involves the use of actual patients, which is time-consuming and costly, and training opportunities are limited. Furthermore, evaluation and feedback of medical interview skills tend to be subjective, making objective evaluation difficult. Furthermore, there is a lack of systems for efficiently improving medical interview skills for patients with various symptoms and medical histories.

[1563] 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.

[1564] In this invention, the server includes a means for conducting medical interview training with a virtual patient using generative artificial intelligence, a means for evaluating the results of the medical interview, a means for providing feedback based on the evaluation results, a means for providing an interface for displaying information about the virtual patient, a means for a user to input questions to the virtual patient and for the virtual patient to respond, a means for analyzing the content of the medical interview and evaluating the user's medical interview skills using an evaluation algorithm, and a means for providing the evaluation results as feedback to the user. This allows medical professionals to efficiently improve their medical interview skills without being restricted by time or location. Furthermore, receiving objective evaluations and specific feedback promotes improvement in medical interview skills.

[1565] "Generative AI" is AI that has the ability to generate new information and responses based on various data.

[1566] A "virtual patient" is a fictitious patient generated by a computer system, rather than a real patient.

[1567] "History training" refers to the practice of questions and responses that medical professionals ask patients.

[1568] The "interview results" are a record of questions asked by a medical professional to a virtual patient and their responses.

[1569] An "evaluation algorithm" is a calculation method for analyzing the results of a medical interview and objectively evaluating the medical interview skills of medical professionals.

[1570] "Feedback" refers to information on areas for improvement and advice provided based on the evaluation results.

[1571] An "interface" is an operating environment containing screens and input fields through which a user interacts with a system.

[1572] "User" refers to a medical professional who uses this system to conduct interview training.

[1573] The "means for inputting a question" refers to a text box or input field for the user to input a question to the virtual patient.

[1574] "Means for returning a response" refers to the function of the virtual patient to generate and display an appropriate response to a user's question.

[1575] The "means for analyzing the contents of the medical interview" refers to a function for analyzing the user's questions and the virtual patient's responses, and making an evaluation based on an evaluation algorithm.

[1576] The "means for providing evaluation results" refers to a function for providing the evaluation results obtained by the evaluation algorithm to the user as feedback.

[1577] This invention relates to a virtual patient interview training system that utilizes generative artificial intelligence. This system allows medical professionals to interview virtual patients, evaluate the results, and provide feedback to improve their interview skills.

[1578] System configuration

[1579] 1. Virtual patient generation

[1580] The server uses generative artificial intelligence (e.g., natural language processing models) to generate virtual patients with various symptoms and medical histories. The generated virtual patient information includes age, gender, chief complaint, medical history, current symptoms, etc.

[1581] 2. Providing a medical interview interface

[1582] The terminal provides a user (healthcare professional) with an interface that displays information about the virtual patient, including basic information about the virtual patient and input fields for conducting a medical interview.

[1583] 3. Conducting a medical interview

[1584] The user inputs questions to the virtual patient through a terminal, and the virtual patient uses generative artificial intelligence to provide appropriate responses.

[1585] 4. Evaluation of interview results

[1586] The server analyzes the user's interview content and evaluates the user's interview skills using an evaluation algorithm, including the appropriateness of the questions, the comprehension of the responses, and the flow of the interview.

[1587] 5. Providing Feedback

[1588] The device provides the evaluation results to the user as feedback, which specifically indicates areas for improvement and merits.

[1589] Hardware and software used

[1590] Server: A computer system with high-performance computing power that runs generative AI models and analyzes data.

[1591] Device: A computer, tablet, or smartphone operated by a user, providing an interview interface and accepting user input.

[1592] Generative artificial intelligence models: use natural language processing models (e.g., GPT-4) to generate responses for virtual patients.

[1593] Evaluation algorithm: An algorithm for analyzing the interview results and evaluating the user's interview skills.

[1594] Specific examples

[1595] Example scenario:

[1596] The hypothetical patient is a 35-year-old man who presents with recent headaches and dizziness. He has been previously diagnosed with high blood pressure and is currently taking medication.

[1597] Example prompt sentence:

[1598] "A 35-year-old male patient presents with recent headaches and dizziness. He has been diagnosed with high blood pressure in the past and is currently taking medication. What questions would you ask the patient?"

[1599] Based on this prompt, the user can ask specific questions to the virtual patient, such as "When did your headache start?" or "In what situations does dizziness occur?"

[1600] The server records these questions and the virtual patient's responses, and uses an evaluation algorithm to evaluate the user's interview skills. The evaluation results are fed back as feedback, such as "appropriateness of questions: 90 points," "understandability of responses: 85 points," and "flow of interview: 80 points."

[1601] In this way, users can improve their skills by interviewing virtual patients.

[1602] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1603] Step 1: Creating a virtual patient

[1604] The server generates a virtual patient using a generative AI model. As input, it receives a prompt statement (e.g., "A 35-year-old male patient complains of recent headaches and dizziness. He has been diagnosed with high blood pressure in the past and is currently taking medication."). Based on this prompt statement, the generative AI model generates detailed information about the virtual patient (e.g., age, gender, chief complaint, medical history, current symptoms, etc.). As output, it obtains information about the generated virtual patient.

[1605] Specific behavior:

[1606] The server inputs prompts into the generative artificial intelligence model to generate detailed information about the virtual patient, which is used for subsequent interview training.

[1607] Step 2: Providing a medical interview interface

[1608] The terminal provides the user with an interface that displays information about the virtual patient. As input, the terminal uses the information about the virtual patient received from the server. As output, an interface is displayed that allows the user to check the information about the virtual patient.

[1609] Specific behavior:

[1610] The terminal displays information about the virtual patient (age, gender, chief complaint, etc.) on the screen and provides a text box where the user can enter questions.

[1611] Step 3: Conduct a medical interview

[1612] The user inputs a question to the virtual patient through the terminal. The question entered by the user in the text box is used as input. The server sends this question to a generative artificial intelligence model, which generates a response from the virtual patient. The output is the response of the virtual patient.

[1613] Specific behavior:

[1614] The user types "When did your headache start?" into a text box. The server sends this question to a generative artificial intelligence model, which generates the virtual patient's response: "My headache started a week ago." The device displays this response to the user.

[1615] Step 4: Evaluate the interview results

[1616] The server analyzes the user's interview content and evaluates the user's interview skills using an evaluation algorithm, using the user's questions and the virtual patient's responses as inputs, and obtaining the evaluation results as output.

[1617] Specific behavior:

[1618] The server analyzes the user's questions and the virtual patient's responses and applies an evaluation algorithm to generate results such as "appropriateness of question: 90 points," "understandability of response: 85 points," and "flow of medical interview: 80 points."

[1619] Step 5: Provide feedback

[1620] The terminal provides the evaluation results as feedback to the user. As input, it uses the evaluation results received from the server. As output, it displays feedback that allows the user to check the evaluation results.

[1621] Specific behavior:

[1622] The device displays the evaluation results on the screen, such as "Question appropriateness: 90 points," "Response comprehension: 85 points," and "Interview flow: 80 points," along with specific advice such as "You can get better results by changing the order of the questions."

[1623] (Application example 1)

[1624] Next, a description will be given of Application Example 1 of Form 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."

[1625] In conventional medical interview training systems, medical professionals have limited opportunities to interact with actual patients, making it difficult to provide a realistic training environment. Furthermore, evaluation and feedback of medical interview results are not provided immediately, which slows down the improvement of medical professionals' skills.

[1626] 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.

[1627] In this invention, the server includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the interview, a means for providing feedback based on the evaluation results, a means for a factory robot to act as the virtual patient and for a medical professional to interview the robot, a means for simulating virtual patients with various symptoms and medical histories using generative AI, and a means for evaluating the results of the interview in real time and providing feedback, thereby enabling medical professionals to improve their medical interview skills in a realistic training environment.

[1628] "Generative AI" is an artificial intelligence technology that generates new information and content based on various data.

[1629] A "virtual patient" is a patient that is not a real patient, but is simulated by a computer system.

[1630] "Medical history training" is training to improve medical professionals' techniques and skills in conducting medical interviews with patients.

[1631] "Interview results" refers to the content of the interview conducted by a medical professional with a patient and the information obtained as a result.

[1632] "Evaluation" refers to analyzing the medical professional's interview techniques and skills based on the results of the interview and judging whether they are good or bad.

[1633] "Feedback" refers to information such as areas for improvement and advice provided to medical professionals based on the evaluation results.

[1634] A "factory robot" is an automated mechanical device used in production sites such as factories.

[1635] "Simulate" means to imitate and reproduce actual situations or conditions.

[1636] "Real-time" means that processing and response are immediate and without delay.

[1637] A system for implementing this invention includes means for conducting medical interview training with a virtual patient using generative AI, means for evaluating the results of the interview, means for providing feedback based on the evaluation results, means for a factory robot acting as a virtual patient and for a medical professional to interview the robot, means for simulating virtual patients with various symptoms and medical histories using generative AI, and means for evaluating the results of the interview in real time and providing feedback.

[1638] The server generates a virtual patient using generative AI. The generated virtual patient is simulated through an application installed on a factory robot. Medical personnel interview the factory robot. The interview results are sent to the server in real time, and the server evaluates the interview results. Based on the evaluation results, the server provides feedback to the medical personnel.

[1639] The hardware used includes servers, factory robots, and devices used by medical professionals, while the software used includes generative AI models (e.g., OpenAI API) and algorithms for evaluating the results of medical interviews.

[1640] As a concrete example, a virtual patient can be generated by inputting the following prompt sentences into a generative AI model:

[1641] Virtual patient generation prompt:

[1642] "Generate virtual patients with a variety of symptoms and medical histories."

[1643] After the healthcare professional has completed the interview, they should evaluate the interview using the following prompts:

[1644] Interview assessment prompt:

[1645] Please rate the following questions:\nQuestion 1: What symptoms are you experiencing?\nAnswer 1: I have a headache.\nQuestion 2: How long have you had these symptoms?\nAnswer 2: About three days ago.

[1646] Based on the assessment, provide feedback using the following prompts:

[1647] Feedback prompt:

[1648] Please provide feedback based on the following assessment results:\n Assessment result: The healthcare professional asks appropriate questions and accurately understands the patient's symptoms.

[1649] This allows medical professionals to improve their medical history taking skills in a realistic training environment.

[1650] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1651] Step 1:

[1652] The server generates a virtual patient using a generative AI model. Specifically, the server inputs the following prompt sentences into the generative AI model:

[1653] "Generate virtual patients with a variety of symptoms and medical histories."

[1654] Based on this input, the generative AI model generates virtual patient data and outputs it to a server, including the virtual patient's symptoms and medical history.

[1655] Step 2:

[1656] The server sends the generated virtual patient data to the factory robot, which then acts as the virtual patient based on this data. Specifically, the factory robot simulates the virtual patient's symptoms and medical history and responds in real time to medical interviews with medical professionals.

[1657] Step 3:

[1658] The user (medical worker) conducts a medical interview with the factory robot. The contents of the interview are input by the user through a terminal. For example, the following questions are asked:

[1659] "What symptoms are you experiencing?"

[1660] The factory robot responds based on the virtual patient's data:

[1661] "I have a headache."

[1662] Step 4:

[1663] The terminal sends the details of the user's interview and the factory robot's response to the server, which receives this data and stores it as the interview results.

[1664] Step 5:

[1665] The server inputs the following prompt sentences into the generative AI model to evaluate the interview results:

[1666] Please rate the following questions:\nQuestion 1: What symptoms are you experiencing?\nAnswer 1: I have a headache.\nQuestion 2: How long have you had these symptoms?\nAnswer 2: About three days ago.

[1667] Based on this input, the generative AI model evaluates the interview results and outputs the evaluation results to the server.

[1668] Step 6:

[1669] The server inputs the following prompt sentences into the generative AI model to provide feedback based on the evaluation results:

[1670] Please provide feedback based on the following assessment results:\n Assessment result: The healthcare professional asks appropriate questions and accurately understands the patient's symptoms.

[1671] Based on this input, the generative AI model generates feedback and outputs it to the server.

[1672] Step 7:

[1673] The server transmits the generated feedback to the user's terminal, and the user can check the feedback through the terminal and use it to improve their interview skills.

[1674] Example 2

[1675] Next, a description will be given of Example 2 of Form 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."

[1676] Conventional medical interview training systems have had the problem of being unable to sufficiently improve medical personnel's medical interview skills due to the difficulty of simulating conversations with actual patients. In addition, evaluation of medical interview results and provision of feedback are often done manually, making efficient training difficult.

[1677] 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.

[1678] In this invention, the server includes means for conducting medical interview training with a virtual patient using generative artificial intelligence, means for evaluating the results of the interview, means for providing feedback based on the evaluation results, means for using prompt sentences to generate the virtual patient, means for transmitting the results of the interview to the server, and means for displaying the feedback on a terminal. This allows medical professionals to improve their skills through interviews with realistic virtual patients, thereby achieving efficient and effective training.

[1679] "Generative AI" is an AI technology that uses natural language processing and machine learning to generate data and information based on specific conditions.

[1680] A "virtual patient" is a fictional character simulated by generative artificial intelligence with symptoms and medical history similar to those of a real patient.

[1681] "Medical history training" involves medical professionals asking questions of virtual patients and determining diagnoses and treatment plans based on the answers.

[1682] The "interview results" are a record of questions asked by medical professionals to a virtual patient and the virtual patient's answers to those questions.

[1683] "Evaluation means" refers to algorithms or programs for analyzing the results of the interview and evaluating their content.

[1684] "Feedback" refers to advice provided to medical professionals on areas for improvement and skill development based on the evaluation results obtained through the evaluation tool.

[1685] A "prompt sentence" is a sentence used to input specific conditions or requests to generative artificial intelligence.

[1686] A "server" is a computer system for executing programs such as generative artificial intelligence, evaluation means, and feedback generation modules.

[1687] A "terminal" is a device such as a computer or tablet that medical professionals use to conduct interview training.

[1688] This invention is a system that utilizes generative artificial intelligence to conduct interview training with virtual patients, evaluate the interview results, and provide feedback based on the evaluation results. Specific embodiments of this system are described below.

[1689] Hardware and software used

[1690] Hardware: Servers, devices (PCs, tablets, etc.)

[1691] Software: Generative AI models (e.g., GPT-4), evaluation algorithms, feedback generation modules

[1692] Virtual patient generation

[1693] The server uses the generative AI model to generate a virtual patient with diabetes symptoms. Specifically, the server inputs the following prompt sentence into the generative AI model:

[1694] Example prompt: "Generate a hypothetical patient who is a 50-year-old male with a BMI of 30, HbA1c of 7.5%, and complains of frequent urination and dry mouth."

[1695] Based on this prompt, the generative AI model generates a detailed virtual patient profile, such as a "50-year-old male, BMI 30, HbA1c 7.5%, complaining of frequent urination and dry mouth" with specific symptoms and medical history.

[1696] Conducting medical interviews

[1697] The user (healthcare worker) uses a terminal to interview the generated virtual patient. The user inputs questions such as the following through the terminal interface.

[1698] "How's your diet been lately?"

[1699] "Do you have an exercise routine?"

[1700] The virtual patient provides answers to these questions through a generative AI model, which the user can use to ask more detailed questions.

[1701] Sending and evaluating medical interview results

[1702] The device sends the results of the medical interview entered by the user to the server, which then analyzes the results using an evaluation algorithm. For example, the server may make an evaluation such as, "There are not enough questions about diet" or "More detailed questions about exercise habits are needed."

[1703] Generating and Providing Feedback

[1704] The server generates feedback for the healthcare professional using the feedback generation module based on the evaluation results. Specifically, the following feedback is generated:

[1705] "It would be good to ask more questions about the diet."

[1706] "By asking specific questions about exercise habits, we can get a better understanding of the patient's lifestyle."

[1707] The device then displays the received feedback to the user, allowing the user to receive specific advice on how to improve their interview skills.

[1708] This system is an effective tool for medical professionals to improve their medical interview skills through virtual patients. The server uses a generative AI model to generate realistic virtual patients, evaluates the interview results, and provides feedback to help medical professionals improve their skills.

[1709] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1710] Step 1:

[1711] Virtual patient generation

[1712] The server uses the generative AI model to generate a virtual patient with diabetes symptoms.

[1713] Input: Prompt: "Create a virtual patient who is a 50-year-old male with a BMI of 30, HbA1c of 7.5%, and complains of frequent urination and dry mouth."

[1714] Data processing: The server inputs the prompt sentences into a generative AI model to generate a detailed profile of the virtual patient.

[1715] Output: Virtual patient profile (e.g., "50-year-old male, BMI 30, HbA1c 7.5%, frequent urination, dry mouth")

[1716] Step 2:

[1717] Conducting medical interviews

[1718] The user (medical worker) uses a terminal to interview the generated virtual patient.

[1719] Input: Questions entered by the user through the device interface (e.g., "How is your diet these days?", "Do you exercise regularly?")

[1720] Data processing: The virtual patient generates answers to these questions through a generative AI model.

[1721] Output: The hypothetical patient's response (e.g., "My diet is high in calories and I barely exercise.")

[1722] Step 3:

[1723] Sending medical interview results

[1724] The terminal transmits the results of the medical interview entered by the user to the server.

[1725] Input: User-entered questions and virtual patient answers

[1726] Data processing: The device converts all of this data into a format that can be sent to the server.

[1727] Output: Interview result data sent to the server

[1728] Step 4:

[1729] Evaluation of interview results

[1730] The server analyzes the received interview results using an evaluation algorithm.

[1731] Input: Submitted medical interview result data

[1732] Data processing: The server uses an evaluation algorithm to analyze the interview results and identify areas for improvement and shortcomings.

[1733] Output: Evaluation results (e.g., "There are not enough questions about diet," "More detailed questions about exercise habits are needed")

[1734] Step 5:

[1735] Generating and Providing Feedback

[1736] The server generates feedback for the healthcare professional based on the evaluation results using a feedback generation module.

[1737] Input: Evaluation result

[1738] Data processing: The server uses a feedback generation module to generate specific improvements and advice for skill development.

[1739] Output: Feedback (e.g., "It would be good to ask more questions about diet," "Ask specific questions about exercise habits to get a better understanding of the patient's lifestyle")

[1740] Step 6:

[1741] View Feedback

[1742] The terminal displays the received feedback to the user.

[1743] Input: Feedback sent by the server

[1744] Data processing: The device displays the feedback in a user-friendly format.

[1745] Output: Feedback displayed to the user

[1746] In this way, healthcare professionals can receive specific advice on how to improve their interview skills through virtual patients.

[1747] (Application example 2)

[1748] Next, a description will be given of Application Example 2 of Form 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."

[1749] Conventional training systems for medical professionals make it difficult to practice medical interviews with actual patients, and often do not provide sufficient evaluation or feedback on the results of the interviews. This has led to problems such as delays in improving medical interview skills. Furthermore, there has been a lack of appropriate systems for providing health management training to operators and safety managers working on-site at factories and other facilities. To solve these issues, an effective training system using virtual patients is needed.

[1750] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1751] In this invention, the server includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the medical interview, a means for providing feedback based on the evaluation results, a means for interacting with the virtual patient using a smart device, and a means for generating the virtual patient and evaluating the results of the medical interview via a cloud service. This enables medical professionals and factory operators to effectively improve their medical interview skills and health management skills without having to meet face-to-face with actual patients or workers.

[1752] "Generative AI" refers to artificial intelligence techniques used to generate virtual patients and scenarios.

[1753] "Medical history training" refers to practice in which medical professionals or operators conduct medical history interviews with virtual patients.

[1754] "Evaluation of interview results" is the process of analyzing the results of an interview with a virtual patient and evaluating their quality.

[1755] "Feedback" is information that provides areas for improvement and ways to improve skills based on the evaluation of the interview results.

[1756] "Smart devices" refers to interactive electronic devices such as smartphones, smart glasses, and head-mounted displays.

[1757] "Interaction" refers to the act of a user interacting with a virtual patient.

[1758] "Cloud services" refers to computing resources and services provided over the Internet.

[1759] A system for implementing this invention includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the interview, a means for providing feedback based on the evaluation results, a means for interacting with the virtual patient using a smart device, and a means for generating the virtual patient and evaluating the results of the interview via a cloud service.

[1760] System configuration

[1761] 1. Generative AI

[1762] The server generates virtual patients using a generative AI. The generative AI uses pre-trained models to generate virtual patients with various symptoms and medical histories. For example, to generate a virtual patient with diabetes, the following prompt sentence is used:

[1763] Generate a virtual patient with diabetes symptoms.

[1764] 2. Medical interview training

[1765] The user interacts with the virtual patient using a smart device (e.g., smart glasses), conducts a medical interview with the virtual patient, and inputs the results into the system.

[1766] 3. Evaluation of interview results

[1767] The server uses generative AI to evaluate the questionnaire results entered by the user. An example of an evaluation prompt is as follows:

[1768] Please rate the results of the following interview.

[1769] Interview results: [Virtual patient information]

[1770] evaluation:

[1771] 4. Providing Feedback

[1772] The server generates feedback based on the evaluation results. An example of a feedback prompt is:

[1773] Based on the evaluation below, please indicate areas for improvement in your interview and how you can improve your skills.

[1774] Evaluation: [Evaluation result]

[1775] feedback:

[1776] The generated feedback is displayed on the user's smart device.

[1777] Hardware and software used

[1778] Hardware: Smart devices (smart glasses, smartphones, head-mounted displays, etc.)

[1779] Software: Generative AI models (e.g., OpenAI API), cloud services (e.g., AWS, Google Cloud)

[1780] Specific examples

[1781] For example, consider a scenario in which a factory operator wears smart glasses and interviews a virtual patient. In this case, the operator checks the virtual patient's information through the smart glasses and proceeds with the interview. The interview results are sent to a server via a cloud service, where they are evaluated by generative AI. Based on the evaluation results, the operator can receive feedback on areas where and how to improve their interview skills.

[1782] In this way, medical professionals and factory operators can effectively improve their medical history and health management skills without having to meet actual patients or workers face-to-face.

[1783] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1784] Step 1:

[1785] The server generates a virtual patient using generative AI. The prompt sentence "Please generate a virtual patient with diabetes symptoms" is used as input. The generative AI model generates data for the virtual patient based on this prompt sentence and provides detailed information about the virtual patient as output.

[1786] Step 2:

[1787] The user interacts with the virtual patient using a smart device (e.g., smart glasses). The user interviews the virtual patient and inputs the results into the smart device. The input includes the interview content the user conducted with the virtual patient. The output is the interview results stored in the smart device.

[1788] Step 3:

[1789] The terminal sends the medical interview results to the server via the cloud service. The input includes the medical interview results entered by the user into the smart device. The output is the medical interview results sent to the server.

[1790] Step 4:

[1791] The server uses a generative AI to evaluate the received medical interview results. As input, it uses the medical interview results and the evaluation prompt "Please rate the following medical interview results.\n\nMedical interview results: [Virtual patient information]\n\nEvaluation:". The generative AI model evaluates the medical interview results based on this prompt and provides the evaluation result as output.

[1792] Step 5:

[1793] The server generates feedback based on the evaluation results. As input, it uses the evaluation results and a feedback prompt: "Based on the evaluation below, please indicate areas for improvement in the interview and ways to improve your skills.\n\nEvaluation: [Evaluation result]\n\nFeedback:". The generative AI model generates feedback based on this prompt and provides the feedback information as output.

[1794] Step 6:

[1795] The server sends the generated feedback to the user's smart device through a cloud service. As an input, the feedback information is included. As an output, the feedback information is displayed on the user's smart device.

[1796] Step 7:

[1797] The user can check the feedback through their smart device and use it to improve their medical interview and health management skills. The input includes feedback information sent from the server. The output is the user receiving the feedback and reflecting it in their actual training.

[1798] Example 3

[1799] Next, a third embodiment of the third embodiment will be described. 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."

[1800] With conventional medical interview training systems, it was difficult to gain experience interviewing real patients, especially patients with rare conditions or specific genetic diseases. Furthermore, there was a problem that the improvement of medical interview skills was delayed because the evaluation and feedback of the interview results was not sufficiently carried out.

[1801] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1802] In this invention, the server includes means for a user to input a prompt sentence, means for a terminal to send the prompt sentence to the server, means for the server to receive the prompt sentence and send it to the generative AI model, means for the generative AI model to generate medical condition data of the virtual patient, means for the server to receive the generated medical condition data and return it to the terminal, and means for the terminal to display the medical condition data to the user. This allows the user to gain experience in interviewing virtual patients with rare conditions or specific genetic diseases, and improves their interview skills through evaluation of the interview results and feedback.

[1803] "Generative AI" is a system that uses artificial intelligence technology to generate data and information based on specific conditions.

[1804] A "virtual patient" is not a real patient, but a fictitious patient generated on a computer system.

[1805] "Medical history training" is training for medical professionals to understand the patient's condition and make a diagnosis through dialogue with the patient.

[1806] A "prompt statement" is an instruction statement entered into a generative AI to cause it to generate specific data or information.

[1807] A "terminal" is a device such as a computer or smartphone operated by a user.

[1808] A "server" is a computer system that processes and stores data over a network.

[1809] A "generative AI model" is an algorithm or program that forms the core of generative AI and generates data based on input prompts.

[1810] "Medical condition data" is data that includes information about a patient's symptoms and medical history.

[1811] "Feedback" refers to information such as areas for improvement and advice provided based on the results of the interview.

[1812] MODE FOR CARRYING OUT THE INVENTION

[1813] This invention is a system for conducting interview training with virtual patients using generative AI. The system involves a series of processes: a user inputs a prompt sentence, a generative AI model generates medical condition data for the virtual patient, the interview results are evaluated, and feedback is provided.

[1814] Hardware and software used

[1815] The system uses the following hardware and software:

[1816] Device: A device such as a computer or smartphone operated by a user.

[1817] Server: A computer system that processes and stores data over a network.

[1818] Generative AI model: The core algorithm or program of generative AI. For example, a model that uses natural language processing technology (e.g., GPT-4).

[1819] Specific operation of the system

[1820] The user accesses the terminal interface and inputs a prompt to generate a medical condition for a virtual patient with a specific genetic disease, for example, "Please generate a medical condition for a virtual patient with Fabry disease. The patient is a 30-year-old male whose main symptoms are pain and rash in the hands and feet."

[1821] The device sends the prompt text entered by the user to the server. The server receives the prompt text sent from the device and sends it to the generative AI model. The generative AI model generates medical condition data for the virtual patient based on the prompt text.

[1822] The generated medical condition data is sent back to the server, which then sends the data back to the terminal. The terminal displays the medical condition data received from the server to the user. The user can check the displayed medical condition data and gain experience in interviewing the virtual patient.

[1823] Specific examples

[1824] For example, if a user wants to generate the symptoms of a virtual patient with a specific genetic disease, Fabry disease, he or she would enter the following prompt into the terminal:

[1825] "Generate the symptoms of a hypothetical patient with Fabry disease. The patient is a 30-year-old man whose main symptoms are pain in the hands and feet and a rash."

[1826] The server receives this prompt and sends it to the generative AI model, which generates the following medical condition data:

[1827] "A 30-year-old male patient suffers from Fabry disease, with the main symptoms being severe pain in the hands and feet and a red rash. In addition, the patient complains of frequent fatigue and gastrointestinal upset. Blood tests have confirmed that he has reduced α-galactosidase A activity."

[1828] In this way, the user can gain experience in interviewing patients with conditions that would be difficult to experience in real life. Furthermore, through evaluation of the interview results and feedback, the user's interview skills can be improved. The flow of the identification process in the third embodiment will be described with reference to FIG. 15.

[1829] Program processing flow

[1830] Step 1: The user enters a prompt statement

[1831] The user accesses the terminal interface and inputs a prompt to generate a medical condition for a virtual patient with a specific genetic disease, for example, "Please generate a medical condition for a virtual patient with Fabry disease. The patient is a 30-year-old male whose main symptoms are pain and rash in the hands and feet."

[1832] Input: The prompt text that the user enters

[1833] Output: The prompt text displayed on the terminal

[1834] Step 2: The terminal sends the prompt to the server

[1835] The terminal sends the prompt text entered by the user to the server, converting it into a suitable format so that the server can understand it.

[1836] Input: The prompt text entered by the user

[1837] Output: The prompt sent to the server

[1838] Step 3: The server receives the prompt and sends it to the generative AI model

[1839] The server receives the prompt sentence sent from the terminal. Then, the server sends this prompt sentence to the generative AI model. The server performs the necessary preprocessing so that the generative AI model can process the prompt sentence.

[1840] Input: Prompt sent from the terminal

[1841] Output: The prompt sent to the generative AI model

[1842] Step 4: The generative AI model generates the virtual patient's medical condition data

[1843] The generative AI model generates medical condition data for virtual patients based on prompt sentences sent from the server. The model uses natural language processing technology to create data for virtual patients with detailed medical conditions related to the specified genetic disease.

[1844] Input: Prompt text sent from the server

[1845] Output: Medical condition data of the generated virtual patient

[1846] Step 5: The server receives the generated medical condition data and sends it back to the device.

[1847] The server receives the medical condition data generated by the generative AI model, then sends this data back to the device, where it converts the data into a format that the device can display.

[1848] Input: Medical condition data sent from the generative AI model

[1849] Output: Medical condition data returned to the terminal

[1850] Step 6: The device displays the medical condition data to the user

[1851] The terminal displays the medical condition data received from the server to the user, who can check the displayed medical condition data and gain experience in interviewing the virtual patient.

[1852] Input: Medical condition data sent from the server

[1853] Output: Medical condition data displayed to the user

[1854] (Application example 3)

[1855] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1856] Conventional medical interview training for healthcare professionals is based on common symptoms and medical histories, making it difficult to gain experience interviewing patients with rare conditions. Training using actual patients also poses ethical issues and resource constraints, making it difficult to provide sufficient training opportunities. Furthermore, evaluation and feedback of interview results is insufficient, preventing healthcare professionals from efficiently improving their skills.

[1857] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1858] In this invention, the server includes: means for conducting medical interview training with a virtual patient using generative AI; means for evaluating the results of the medical interview; means for providing feedback based on the evaluation results; means for including a simulator that is installed on a factory robot and enables medical professionals to interact with virtual patients with rare medical conditions and gain medical interview experience; and means for generating the virtual patient's medical condition in real time using a generative AI model and providing appropriate responses to questions from the medical professionals. This allows medical professionals to gain experience in medical interviews with virtual patients with rare medical conditions, efficiently improving their medical interview skills.

[1859] "Generative AI" is a system that uses artificial intelligence technology to generate new data and information.

[1860] A "virtual patient" is a patient that does not actually exist but is recreated through computer simulation.

[1861] "Medical history training" is training to improve medical professionals' skills in listening to patients' symptoms and medical history.

[1862] "Interview results" is a general term for information obtained from patients by medical professionals.

[1863] "Evaluation" means analyzing the skills and responses of medical professionals based on the results of the medical interview and judging whether they are good or bad.

[1864] "Feedback" means providing suggestions for improvement or advice based on the evaluation results.

[1865] A "factory robot" is an automated mechanical device designed to perform tasks in a factory.

[1866] A "simulator" is a device or software that simulates a real-world situation.

[1867] A "generative AI model" is an algorithm or program that forms the core of generative AI and is used to generate new data and information.

[1868] "Real-time" refers to immediate processing or response without delay.

[1869] "Disease condition" refers to the symptoms or disease state exhibited by a patient.

[1870] A "question" is a question that a medical professional asks a patient to obtain information.

[1871] A "response" is an answer or reaction to a question.

[1872] A system for implementing this invention includes a means for conducting medical interview training with a virtual patient using generative AI, a means for evaluating the results of the medical interview, a means for providing feedback based on the evaluation results, a means for including a simulator that is installed on a factory robot and allows medical professionals to interact with virtual patients with rare medical conditions and gain medical interview experience, and a means for generating the virtual patient's medical condition in real time using a generative AI model and providing appropriate responses to questions from medical professionals.

[1873] Hardware and software used

[1874] Hardware: Factory robots

[1875] Software: OpenAI API, Python program

[1876] System processing overview

[1877] 1. Data entry: Medical profession...

Claims

1. A first means for accepting input by a user of a prompt sentence for generating medical condition data of a virtual patient having a specific genetic disease; A second means for inputting the prompt sentence into a generative AI model to generate medical condition data of the virtual patient; a third means for displaying the generated medical condition data of the virtual patient on a terminal of the user; A system including:

2. a fourth means for recognizing an emotional state of the user from the user's tone of voice, facial expression, and word choice using an emotion engine; a fifth means for adjusting a response of the virtual patient based on an emotional state of the user; The system of claim 1 , comprising:

3. The fourth means uses the emotion engine to analyze the emotional state of the user during the interview with the virtual patient from the tone of voice, facial expression, and choice of words of the user; a sixth means for evaluating a relevance of a result of the virtual patient interview and an emotional state of the user at the time of the interview, and generating feedback for the user; The system of claim 2.

Images