System

Abstract

A system is provided.SOLUTION: A system comprising: means for analyzing a patient's declared symptoms, medical image data, and lab results using a generative artificial intelligence model and presenting candidate diagnoses; means for displaying the analysis results to a healthcare professional; and means for collecting and using the final diagnosis results and feedback from the healthcare professional as training data for the generative artificial intelligence model.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 conventional medical diagnoses, doctors make diagnoses based on the patient's reported symptoms, medical images, and test results, which can lead to misdiagnoses or oversights. Furthermore, because the diagnosis depends on the doctor's specialty, it is often difficult to provide an appropriate response for a particular case. There is a need to solve these problems and improve the accuracy and efficiency of diagnoses. [Means for solving the problem]

[0005] The present invention provides a system for analyzing patient-reported symptoms, medical image data, and test results using a generative artificial intelligence model, and presenting candidate diagnoses. The system includes the following means.

[0006] 1. A means of using generative artificial intelligence models to analyze patient-reported symptoms, medical imaging data, and test results to suggest potential diagnoses.

[0007] 2. A means of displaying analysis results to medical professionals.

[0008] 3. A means of collecting final diagnostic results and feedback from healthcare professionals to use as training data for generative artificial intelligence models.

[0009] 4. Means for storing reported symptoms, medical image data and test results in a database (see claim 2).

[0010] 5. Means for pre-processing the collected data (see claim 3).

[0011] This provides doctors with auxiliary information to make more accurate and faster diagnoses, reducing the risk of misdiagnosis, and allows the generative AI model to be continuously improved through collected feedback, thereby increasing diagnostic accuracy.

[0012] "Generative AI models" are artificial intelligence algorithms used to generate potential diagnoses for patients. These algorithms leverage knowledge gained from studying large amounts of medical data to make predictions based on input symptoms and test results.

[0013] "Reported symptoms" are specific complaints about a patient's health condition or poor physical condition that a patient reports to a medical professional.

[0014] "Medical image data" refers to digital image data obtained from medical imaging tests such as X-rays, CT scans, and MRIs.

[0015] "Test results" refers to measurements and diagnostic information obtained from various medical tests such as blood tests, urine tests, and cytology.

[0016] "Candidate Diagnosis" refers to a list of possible diagnoses generated by a generative artificial intelligence model based on a patient's symptoms and test results.

[0017] A "means" refers to an element of an apparatus, method, or system configured to perform a specific function in the present invention.

[0018] "Healthcare professionals" are professionals who are licensed to perform medical procedures, such as doctors, nurses, and technicians.

[0019] "Preprocessing" refers to the process of formatting data and removing unnecessary information before data is input or analyzed.

[0020] "Feedback" refers to evaluations and comments made by healthcare professionals on the system's output, which are collected to help improve the system. [Brief explanation of the drawings]

[0021] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8]FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0022] 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.

[0023] First, the terms used in the following description will be explained.

[0024] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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."

[0029] [First embodiment]

[0030] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0031] 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.

[0032] 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).

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0038] 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.

[0039] 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.

[0040] 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.

[0041] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0042] The embodiments for carrying out the present invention will be described below.

[0043] System Overview

[0044] The present invention relates to a medical diagnosis support system, in particular a system that uses a generative artificial intelligence model to analyze a patient's symptoms, medical image data, and test results, and presents potential diagnoses. The system includes the following main components:

[0045] 1. Terminal: A device where the user (doctor) enters patient information and receives analysis results.

[0046] 2. Server: The central system that collects and analyzes data and uses generative AI models to present potential diagnoses.

[0047] 3. Database: A storage system for storing patient data, analysis results, and feedback.

[0048] Explaining program processing in natural language

[0049] Data Collection Phase

[0050] The terminal receives patient information from the user (doctor) and transmits it to the server. Specifically, this information includes the patient's basic information, reported symptoms, medical image data, and test results.

[0051] Example: A user (doctor) uses a terminal to input basic information about a patient complaining of abdominal pain, CT scan images, and blood test results, and sends them to a server.

[0052] Data analysis phase

[0053] The server stores the received data in a database and performs preprocessing, which includes checking data consistency, cleaning, and format conversion. Once preprocessing is complete, the data is input into a generative AI model for analysis.

[0054] Example: The server preprocesses data received about a patient's abdominal pain and inputs it into a generative AI model. The generative AI model analyzes the data based on past cases of abdominal pain and medical knowledge to generate candidate diagnoses.

[0055] Diagnosis candidate presentation phase

[0056] The server organizes the diagnostic candidates obtained from the generative AI model and sends them to the device used by the user (doctor). The device displays the diagnostic candidates in a format that is easy for the user to understand.

[0057] Example: The server generates diagnostic candidates such as "acute appendicitis" and "intestinal obstruction" and sends them to the terminal so that the user (doctor) can check them.

[0058] Feedback gathering phase

[0059] The device allows the user (doctor) to input the final diagnosis and feedback. This data is sent to the server and stored in a database. This feedback is then used to retrain the generative AI model, contributing to improving the accuracy of the system.

[0060] For example, a user (doctor) diagnoses "acute appendicitis" and provides feedback that the system's diagnosis was appropriate. This information is sent to the server and used as training data for the generative AI model.

[0061] Specific operation example

[0062] When a patient visits the hospital with abdominal pain, the user (doctor) uses a terminal to input the patient's basic information, CT scan images, blood test results, and other test results. This data is sent to the server and stored in a database. The server preprocesses the data and inputs it into a generative AI model for analysis. The generated diagnosis candidates are sent to the terminal and displayed to the user (doctor). After the final diagnosis is made, feedback is sent from the terminal to the server, stored in a database, and used again as training data for the generative AI model.

[0063] The system provides doctors with fast and accurate supplementary information, reducing the risk of misdiagnosis, and continuously improves its accuracy through collected feedback.

[0064] The processing flow will be explained below.

[0065] Specific explanation of the program's processing steps

[0066] Data Collection Phase

[0067] Step 1:

[0068] The terminal receives basic patient information from the user (doctor), including the patient's name, age, sex, and other basic information.

[0069] pseudo

[0070] The terminal receives basic patient information from the user.

[0071] Step 2:

[0072] The terminal receives input from the user (doctor) of the patient's reported symptoms, including the specific symptoms and progress of the patient's awareness.

[0073] pseudo

[0074] The terminal inputs the patient's reported symptoms.

[0075] Step 3:

[0076] The terminal receives input of medical image data (e.g., X-rays, CT scans) and test results (e.g., blood test results, urine test results) from the user (doctor).

[0077] pseudo

[0078] The terminal inputs medical image data and test results

[0079] Step 4:

[0080] The device sends all collected data to the server.

[0081] pseudo

[0082] The device sends the input data to the server

[0083] Data analysis phase

[0084] Step 5:

[0085] The server receives the patient data transmitted from the terminal.

[0086] pseudo

[0087] The server receives data from the device

[0088] Step 6:

[0089] The server stores the received patient data in a database.

[0090] pseudo

[0091] The server stores the data in a database

[0092] Step 7:

[0093] The server checks the integrity of the stored data and performs pre-processing (data cleaning, format conversion, etc.).

[0094] pseudo

[0095] The server performs data integrity checks and preprocessing

[0096] Step 8:

[0097] The server inputs the preprocessed data into the generative AI model.

[0098] pseudo

[0099] The server inputs the preprocessed data into the generative AI model

[0100] Step 9:

[0101] The server obtains the candidate diagnosis results from the generative AI model.

[0102] pseudo

[0103] The server receives the analysis results from the generated AI model

[0104] Diagnosis candidate presentation phase

[0105] Step 10:

[0106] The server formats the acquired diagnosis candidates into a format that is easy for the user (doctor) to understand.

[0107] pseudo

[0108] The server formats the candidate diagnoses

[0109] Step 11:

[0110] The server sends the formatted diagnosis candidates to the terminal and presents them to the user (doctor).

[0111] pseudo

[0112] The server sends the diagnosis candidate to the terminal.

[0113] The device displays possible diagnoses to the user (doctor)

[0114] Feedback gathering phase

[0115] Step 12:

[0116] The terminal allows the user (doctor) to input the final diagnosis result.

[0117] pseudo

[0118] The terminal receives the final diagnosis result from the user.

[0119] Step 13:

[0120] The terminal allows the user (doctor) to input feedback on the system's diagnostic accuracy and areas for improvement.

[0121] pseudo

[0122] The device receives feedback from the user

[0123] Step 14:

[0124] The terminal sends the final diagnosis result and feedback to the server.

[0125] pseudo

[0126] The device sends the final diagnosis results and feedback to the server

[0127] Step 15:

[0128] The server stores the final diagnosis results and feedback in a database.

[0129] pseudo

[0130] The server stores the final diagnosis results and feedback in a database

[0131] Step 16:

[0132] The server uses the collected feedback to retrain the generative AI model.

[0133] pseudo

[0134] The server uses the feedback to retrain the generative AI model

[0135] Through this series of processing flows, the system of the present invention can assist users (doctors) in making diagnoses, improving diagnostic accuracy while reducing the risk of misdiagnosis. Furthermore, through continuous improvements based on feedback, the reliability and effectiveness of the system can be increased.

[0136] Example 1

[0137] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0138] Conventional medical diagnosis support systems have difficulty efficiently analyzing patients' reported symptoms, medical image data, and test results, and presenting appropriate diagnostic candidates. Furthermore, there is a lack of a method for effectively aggregating feedback from medical professionals during the process of making a final diagnosis and improving the accuracy of the diagnosis support system. This increases the risk of misdiagnosis and impacts the quality of medical care.

[0139] 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.

[0140] In this invention, the server includes a means for transmitting patient information input from a terminal to the server, a means for the server to store the received data in a database and perform data preprocessing, a means for analyzing the patient's reported symptoms, medical image data, and test results using a generative AI model and presenting candidate diagnoses, a means for displaying the analysis results to medical professionals, and a means for collecting final diagnoses and feedback from medical professionals and using them as training data for the generative AI model. This enables efficient collection and preprocessing of patient data, presentation of appropriate candidate diagnoses, collection of feedback, and improvement of system accuracy.

[0141] A "generative artificial intelligence model" is a model generated using artificial intelligence technology that analyzes a patient's symptoms and test results based on medical data and generates potential diagnoses.

[0142] "Reported symptoms" are physical discomfort or abnormalities that patients report to medical professionals, and are information that forms the basis of diagnosis.

[0143] "Medical image data" refers to image data obtained using medical equipment, such as CT scans, MRIs, and X-rays, that provide the visual information necessary for diagnosis.

[0144] "Test results" refer to the numerical data and analytical results obtained through various medical tests, such as blood tests and urine tests, and are important information that assists in diagnosis.

[0145] "Candidate diagnoses" are a list of possible diagnoses based on the patient's symptoms and test results, presented as a result of the generative AI model's analysis, and serve as a reference for medical professionals to make a final decision.

[0146] A "terminal" is a device used by a user (doctor) to enter patient information and receive analysis results, and includes PCs, tablets, smartphones, etc.

[0147] The "server" is a central system that receives, stores, and preprocesses patient information, analyzes it using a generative AI model, and presents diagnostic candidates, and also connects to the database.

[0148] A "database" is a storage system for systematically storing patient data, analysis results, and feedback from medical professionals, and is used for subsequent data reference and analysis.

[0149] "Preprocessing" refers to the process of preparing the data to be analyzed, and includes checking data consistency, cleaning, format conversion, etc.

[0150] "Feedback" refers to opinions and evaluations provided by medical professionals based on final diagnosis results, and is data used to retrain generative AI models.

[0151] System Overview

[0152] This invention relates to a medical diagnosis support system that uses a generative AI model to analyze patient-reported symptoms, medical image data, and test results, and presents potential diagnoses. The system includes the following main components:

[0153] 1. Terminal: A device where the user (doctor) enters patient information and receives analysis results. This includes PCs, tablets, smartphones, etc.

[0154] 2. Server: This is the central system that collects and analyzes data, uses generative AI models to present diagnostic candidates, and connects to the database.

[0155] 3. Database: A storage system for storing patient data, analysis results, and feedback.

[0156] Explaining program processing in natural language

[0157] Data Collection Phase

[0158] A user (doctor) uses a terminal to input patient information and sends it to a server. Specifically, this includes the patient's basic information (name, age, gender), reported symptoms, medical image data (CT scans and MRIs), and various test results (blood tests, urine tests, etc.). For example, the user (doctor) inputs the information of a 30-year-old male patient complaining of abdominal pain into the terminal and sends it to the server.

[0159] Data analysis phase

[0160] The server stores the received data in a database and then performs preprocessing on the data. Preprocessing includes checking data integrity, cleaning, and any necessary format conversions. Once preprocessed, the data is input into a generative AI model. The server preprocesses the patient information and inputs it into a generative AI model built using TensorFlow and PyTorch for analysis. The generative AI model performs analysis based on past case data and medical knowledge to generate candidate diagnoses.

[0161] Diagnosis candidate presentation phase

[0162] The server organizes the generated candidate diagnoses and sends them to the terminal. The terminal displays the candidate diagnoses in a format that is easy for the user (doctor) to understand. For example, the terminal displays "acute appendicitis" and "intestinal obstruction" as part of the list of candidate diagnoses generated by the server.

[0163] Feedback gathering phase

[0164] The user (doctor) enters the final diagnosis and feedback into the device and sends it to the server. The feedback is then stored in a database and used to retrain the generative AI model. This feedback continuously improves the accuracy of the system. For example, if the user (doctor) diagnoses "acute appendicitis" and enters feedback that the diagnosis was appropriate, this is sent to the server and used as training data for the generative AI model.

[0165] Prompt Sentence Examples

[0166] "Generate possible diagnoses based on the CT scan images and blood test results of a 30-year-old man complaining of abdominal pain."

[0167] This system will enable doctors to obtain supplementary information quickly and accurately, reducing the risk of misdiagnosis. Furthermore, it is expected that the accuracy of the generative AI model will be improved through collected feedback.

[0168] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0169] Step 1:

[0170] The terminal accepts input from the user (doctor).

[0171] Specific operation: The user (doctor) uses the terminal to input the patient's basic information (name, age, gender), reported symptoms, medical image data (CT scan or MRI), and various test results (blood test, urine test, etc.).

[0172] Input: Patient's basic information, reported symptoms, medical image data, and test results.

[0173] Output: The entered patient information.

[0174] Step 2:

[0175] The terminal transmits the input information to the server.

[0176] Specific operation: The terminal encrypts the entered patient data and sends it to the server using the TLS protocol.

[0177] Input: Patient information entered into the terminal.

[0178] Output: Patient information sent to the server.

[0179] Step 3:

[0180] The server stores the received data in a database.

[0181] Specific operation: The server adds the received patient data to the "Patient Data" table in the database as a new record.

[0182] Input: Patient information received by the server.

[0183] Output: Patient information stored in the database.

[0184] Step 4:

[0185] The server pre-processes the data.

[0186] Specific operation: The server checks the integrity of the received data, removes invalid values, fills in missing data, and converts the format.

[0187] Input: Raw data stored in a database.

[0188] Output: Consistency checked and preprocessed data.

[0189] Step 5:

[0190] The server inputs the preprocessed data into the generative AI model.

[0191] Specific operation: The server inputs the preprocessed data into a generative AI model using TensorFlow or PyTorch.

[0192] Input: Preprocessed data.

[0193] Output: The data that is input into a generative AI model.

[0194] Step 6:

[0195] A generative AI model analyzes the data and generates candidate diagnoses.

[0196] How it works: The generative AI model analyzes the pre-processed data and generates candidate diagnoses based on past case data and medical knowledge.

[0197] Input: Preprocessed data fed into a generative AI model.

[0198] Output: The generated candidate diagnoses.

[0199] Step 7:

[0200] The server transmits the generated candidate diagnoses to the terminal.

[0201] Specific operation: The server organizes the generated diagnosis candidate list and sends it to the terminal.

[0202] Input: A list of diagnostic candidates output from a generative AI model.

[0203] Output: A list of diagnostic candidates sent to the terminal.

[0204] Step 8:

[0205] The terminal displays the possible diagnoses to the user (doctor).

[0206] Specific operation: The terminal displays the list of diagnosis candidates in a format (for example, a list format) that is easy for the user (doctor) to understand.

[0207] Input: A list of diagnostic candidates sent to the terminal.

[0208] Output: A list of possible diagnoses displayed to the user (doctor).

[0209] Step 9:

[0210] The user (doctor) inputs the diagnosis and feedback.

[0211] Specific operation: The user (doctor) inputs the final diagnosis and feedback to the system into the terminal.

[0212] Input: User's (doctor's) final diagnosis and feedback.

[0213] Output: Feedback information entered into the terminal.

[0214] Step 10:

[0215] The terminal sends feedback information to the server.

[0216] Specific operation: The terminal encrypts the input feedback information and sends it to the server.

[0217] Input: Feedback information entered into the device.

[0218] Output: Feedback information sent to the server.

[0219] Step 11:

[0220] The server stores the feedback information in a database.

[0221] Specific operation: The server stores the feedback information in a "diagnostic feedback" table in the database.

[0222] Input: Feedback information received by the server.

[0223] Output: Feedback information stored in a database.

[0224] Step 12:

[0225] The server retrains the generative AI model based on the feedback.

[0226] Specific operation: The server periodically extracts feedback information from the database and retrains the generative AI model.

[0227] Input: Feedback information stored in the database.

[0228] Output: The retrained generative AI model.

[0229] (Application example 1)

[0230] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0231] In modern medical settings, doctors are required to make quick and accurate diagnoses within a limited time frame, but this is often difficult. While selecting the right medicine is also important for pharmacies and drugstores, it can be difficult to accurately understand a patient's symptoms and provide appropriate advice. Therefore, there is a need for a system that can support doctors and pharmacists in the diagnostic process.

[0232] 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.

[0233] In this invention, the server includes means for analyzing patient-reported symptoms, medical image data, and test results using a generative AI model and presenting candidate diagnoses, means for displaying the analysis results to users such as medical professionals or patients, means for collecting final diagnosis results and feedback from users and using them as training data for the generative AI model, and means for using the medical diagnosis support system in physical stores such as drugstores. This allows doctors and pharmacists to quickly and accurately present candidate diagnoses, enabling them to select appropriate medicines and medical procedures.

[0234] A "generative artificial intelligence model" is an artificial intelligence algorithm used to analyze data such as patient symptoms, medical images, and test results to generate potential diagnoses.

[0235] "Reported symptoms" refers to any ill health or specific symptoms reported by the patient.

[0236] "Medical image data" refers to image data taken using medical equipment such as CT scans and MRIs.

[0237] "Test results" refers to data obtained from various medical tests, such as blood tests and urine tests.

[0238] "Analysis results" refer to the diagnostic candidates and related information obtained using the generative artificial intelligence model.

[0239] "Healthcare professionals" refer to professionals involved in diagnosing and treating patients, such as doctors and nurses.

[0240] "Feedback" refers to the final diagnostic results and evaluations and opinions from users.

[0241] "Training Data" means data used to improve and optimize a generative artificial intelligence model.

[0242] A "medical diagnosis support system" is a system that uses a generative artificial intelligence model to present potential diagnoses to medical professionals and patients, helping them make a final diagnosis.

[0243] The embodiments for carrying out the present invention will be described below.

[0244] System Overview

[0245] The present invention relates to a medical diagnosis support system, in particular a system that uses a generative artificial intelligence model to analyze a patient's symptoms, medical image data, and test results, and presents potential diagnoses. The system includes the following main components:

[0246] 1. Terminal: A device where users (doctors and patients) enter symptoms and test results and receive analysis results. Specifically, a smartphone, tablet, or interactive display is used.

[0247] 2. Server: This is the central system that collects, analyzes, and stores data, and uses generative AI models to present diagnostic candidates. A cloud server (e.g., Amazon Web Services or Microsoft Azure) is used.

[0248] 3. Database: A storage system for storing patient data, analysis results, and feedback. SQL / NoSQL databases (MySQL, MongoDB, PostgreSQL) are used.

[0249] Program Processing and Technical Details

[0250] Data Collection Phase

[0251] Users (doctors or patients) use a terminal to input their symptoms, medical images, test results, etc. For example, a patient complaining of abdominal pain inputs their age, gender, CT scan images, and blood test results. This data is then sent to a server via the Internet.

[0252] Data analysis phase

[0253] The server stores the received data in a database and performs preprocessing (integrity checks, cleaning, format conversion) on the data. After preprocessing is complete, the data is input into a generative artificial intelligence model for analysis. This process uses machine learning libraries such as TensorFlow and PyTorch.

[0254] Diagnosis candidate presentation phase

[0255] The candidate diagnoses obtained from the analysis results are organized on the server and sent to the terminal, which then displays them in a format that is easy for the user to understand. For example, candidate diagnoses such as "acute appendicitis" or "intestinal obstruction" are displayed.

[0256] Feedback gathering phase

[0257] The final diagnosis and feedback are entered on the device and sent to the server, where they are stored in a database and used to retrain the generative AI model.

[0258] Specific operation example

[0259] When a patient visits a hospital or drugstore with abdominal pain, the user (doctor or pharmacist) uses a terminal to enter the patient's basic information, CT scan images, blood test results, and other test results. For example, the user might enter, "I have abdominal pain and have had similar symptoms several times in the past. I am 45 years old. I also have CT scan images. Please let me know if you have any diagnostic candidates." This data is sent to a server and stored in a database. The server preprocesses the data and inputs it into a generative AI model for analysis. The generated diagnostic candidates are sent to the terminal and displayed to the user (doctor or patient). After the final diagnosis is made, feedback is sent from the terminal to the server, stored in a database, and used as training data for the generative AI model.

[0260] The system allows users to quickly and accurately identify potential diagnoses, reducing the risk of misdiagnosis, and continuously improves its accuracy through collected feedback.

[0261] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0262] Step 1:

[0263] The terminal receives input from the user (doctor or patient) about symptoms, basic information, medical image data, and test results. The input data includes age, gender, and details of symptoms. This input data is sent to the server via the Internet. An example of a prompt for the user to input specific information is, "I have abdominal pain and have had similar symptoms several times in the past. I am 45 years old. I also have a CT scan image. Please let me know if you have any possible diagnoses."

[0264] Step 2:

[0265] The server receives data sent from the terminal and stores it in a database. The stored data may not be suitable for analysis as is, so it must be preprocessed. Specific preprocessing steps include checking data consistency, filling in missing data, and standardizing formats. This process ensures the availability and reliability of input data.

[0266] Step 3:

[0267] The server inputs the preprocessed data into the artificial intelligence model. Specifically, machine learning libraries such as TensorFlow and PyTorch are used to input the data into the model. The model then performs analysis based on the input data, using past case data and a medical knowledge base. As a result of the analysis, candidate diagnoses are generated.

[0268] Step 4:

[0269] The server organizes the generated candidate diagnoses and sends them to the terminal. The terminal displays the analysis results in an easy-to-read format for the user (doctor or patient). At this stage, an appropriate user interface design is important to present the analysis results in an easy-to-understand format. For example, specific candidate diagnoses such as "acute appendicitis" or "intestinal obstruction" are displayed.

[0270] Step 5:

[0271] The user uses the terminal to input the final diagnosis results and feedback. The user also inputs feedback on the diagnosis performed based on the confirmed diagnosis candidates and their results into the terminal. This includes an evaluation of the accuracy of the diagnosis and the information presented. This feedback data is sent to the server.

[0272] Step 6:

[0273] The server stores the final diagnosis results and feedback sent by the user in a database and uses them as training data for the generative AI model. By retraining the generative AI model based on this feedback data, the accuracy of the model is continuously improved, further increasing the reliability and usefulness of the system.

[0274] Through the above processing steps, this medical diagnosis support system is able to efficiently and accurately analyze medical data and present potential diagnoses.

[0275] 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.

[0276] The embodiments for carrying out the present invention will be described below.

[0277] System Overview

[0278] The present invention relates to a medical diagnosis support system, in particular a system that uses a generative artificial intelligence model and an emotion engine to analyze patient symptoms, medical image data, and test results, and presents potential diagnoses while also taking into account the user's emotional state. The system includes the following main components:

[0279] 1. Terminal: A device where the user (doctor) inputs patient information and receives analysis results and emotions.

[0280] 2. Server: The central system that collects and analyzes data, uses generative AI models to present diagnostic candidates, and analyzes user sentiment.

[0281] 3. Database: A storage system for storing patient data, analysis results, feedback and emotion data.

[0282] 4. Emotion engine: Algorithms for recognizing and analyzing the emotional state of the user (doctor) and patient.

[0283] Explaining program processing in natural language

[0284] Data Collection Phase

[0285] The terminal receives basic patient information from the user (doctor) and sends it to the server. This information includes the patient's basic information, reported symptoms, medical image data, and test results. The emotion engine also recognizes the user's (doctor's) emotional state and captures it as data.

[0286] Example: A user (doctor) uses a terminal to input basic information about a patient complaining of abdominal pain, CT scan images, and blood test results, and at the same time, the emotion engine detects the user's (doctor's) stress level and transmits the data.

[0287] Data analysis phase

[0288] The server stores the received data in a database and performs preprocessing. Preprocessing includes checking data consistency, cleaning, and format conversion. Once preprocessed, the data is input into a generative AI model for analysis. Furthermore, emotional data of users and patients obtained by the emotion engine is also used in the analysis.

[0289] Example: The server preprocesses data received about a patient's abdominal pain and inputs it into a generative AI model. The generative AI model analyzes the data based on past cases of abdominal pain and medical knowledge to generate a candidate diagnosis. At the same time, it evaluates the possibility that the doctor's emotional state may affect the diagnosis based on the results of the emotion engine analysis.

[0290] Diagnosis candidate presentation phase

[0291] The server organizes the acquired diagnosis candidates and sends them to the device used by the user (doctor). The device displays the diagnosis candidates and emotion analysis results in a format that is easy for the user (doctor) to understand. The way in which the diagnosis candidates are presented is also adaptively changed depending on the emotion analysis results.

[0292] For example, the server generates diagnostic candidates such as "acute appendicitis" or "intestinal obstruction" and sends them to the terminal for the user (doctor) to review. At the same time, the emotion engine adjusts the way the diagnostic candidates are displayed based on the emotional state recognized by the server.

[0293] Feedback gathering phase

[0294] The device allows the user (doctor) to input the final diagnosis, feedback, and emotional data. This data is sent to the server and stored in a database. This feedback is then used to retrain the generative AI model, contributing to improving the accuracy of the system.

[0295] Example: A user (doctor) diagnoses "acute appendicitis" and provides feedback that the system's diagnosis was appropriate. This information is sent to the server and used as training data for the generative AI model. The emotional state collected by the emotion engine is also stored.

[0296] Specific operation example

[0297] When a patient visits the hospital with abdominal pain, the user (doctor) uses a terminal to input the patient's basic information, CT scan images, blood test results, and other test results. This data is sent to the server and stored in a database. The server preprocesses the data and inputs it into a generative AI model for analysis. The generated diagnosis candidates are sent to the terminal and displayed to the user (doctor). After the final diagnosis is made, feedback is sent from the terminal to the server, stored in a database, and used again as training data for the generative AI model. At the same time, an emotion engine detects the user's emotional state and uses this data for analysis.

[0298] This system allows doctors to obtain supplementary information quickly and accurately, reducing the risk of misdiagnosis. Furthermore, the use of an emotion engine enables diagnostic support that takes the user's mental state into account, which is expected to improve diagnostic accuracy and the quality of medical care. The accuracy of the system is continuously improved through collected feedback.

[0299] The processing flow will be explained below.

[0300] Specific explanation of the program's processing steps

[0301] Data Collection Phase

[0302] Step 1:

[0303] The terminal receives basic patient information from the user (doctor), including basic information such as the patient's name, age, sex, height, and weight.

[0304] pseudo

[0305] The terminal receives basic patient information from the user.

[0306] Step 2:

[0307] The terminal receives input from the user (doctor) of the patient's reported symptoms, including the specific symptoms and progress of the symptoms that the patient is aware of.

[0308] pseudo

[0309] The terminal inputs the patient's reported symptoms.

[0310] Step 3:

[0311] The terminal receives input of medical image data (e.g., X-rays, CT scans) and test results (e.g., blood test results, urine test results) from the user (doctor).

[0312] pseudo

[0313] The terminal inputs medical image data and test results

[0314] Step 4:

[0315] The emotion engine recognizes the user's (doctor's) emotional state and captures it as data, including the doctor's stress level and fatigue level.

[0316] pseudo

[0317] Emotion engine recognizes the doctor's emotional state

[0318] Step 5:

[0319] The device sends all collected data to the server.

[0320] pseudo

[0321] The device sends the input data to the server

[0322] Data analysis phase

[0323] Step 6:

[0324] The server receives the patient data and emotion data transmitted from the terminal.

[0325] pseudo

[0326] The server receives data from the device

[0327] Step 7:

[0328] The server stores the received patient data in a database.

[0329] pseudo

[0330] The server stores the data in a database

[0331] Step 8:

[0332] The server checks the integrity of the stored data and performs pre-processing (data cleaning, format conversion, etc.).

[0333] pseudo

[0334] The server performs data integrity checks and preprocessing

[0335] Step 9:

[0336] The server inputs the preprocessed data into a generative AI model, including emotional data.

[0337] pseudo

[0338] The server inputs the preprocessed data into the generative AI model

[0339] Step 10:

[0340] The server obtains the candidate diagnosis results from the generative AI model.

[0341] pseudo

[0342] The server receives the analysis results from the generated AI model

[0343] Diagnosis candidate presentation phase

[0344] Step 11:

[0345] The server formats the diagnosis candidates obtained into a format that is easy for the user (doctor) to understand, and adjusts the results by reflecting emotional data.

[0346] pseudo

[0347] The server formats the candidate diagnoses

[0348] The server adjusts the results by taking into account the emotional data.

[0349] Step 12:

[0350] The server sends the formatted diagnosis candidates to the terminal and presents them to the user (doctor).

[0351] pseudo

[0352] The server sends the diagnosis candidate to the terminal.

[0353] The device displays possible diagnoses to the user (doctor)

[0354] Feedback gathering phase

[0355] Step 13:

[0356] The terminal allows the user (doctor) to input the final diagnosis result.

[0357] pseudo

[0358] The terminal receives the final diagnosis result from the user.

[0359] Step 14:

[0360] The device allows the user (doctor) to input feedback on the system's diagnostic accuracy and areas for improvement, along with their emotional state.

[0361] pseudo

[0362] The device receives feedback from the user

[0363] The device also collects emotional state data

[0364] Step 15:

[0365] The terminal sends the final diagnosis result and feedback to the server.

[0366] pseudo

[0367] The device sends the final diagnosis results and feedback to the server

[0368] Step 16:

[0369] The server stores the final diagnosis results and feedback in a database.

[0370] pseudo

[0371] The server stores the final diagnosis results and feedback in a database

[0372] Step 17:

[0373] The server uses the collected feedback to retrain the generative AI model, including the emotional data.

[0374] pseudo

[0375] The server uses the feedback to retrain the generative AI model

[0376] Through this series of processing flows, the system of the present invention provides diagnostic support to the user (doctor), improving diagnostic accuracy while reducing the risk of misdiagnosis. Furthermore, by using an emotion engine, diagnostic support that takes into account the user's mental state is possible, which is expected to improve diagnostic accuracy and the quality of medical care. The accuracy of the system is continuously improved through collected feedback.

[0377] Example 2

[0378] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0379] Traditionally, medical diagnoses have relied primarily on the doctor's experience and knowledge, leading to potential misdiagnoses and oversights. Furthermore, the doctor's emotional state often influences the diagnosis. Furthermore, systems that support diagnosis using generative AI models do not take into account the emotional state of the doctor or patient, limiting the accuracy and adaptability of the diagnosis. Conventional technologies do not incorporate emotional data when collecting feedback or retraining AI models, resulting in inflexible behavior adjustments and making it difficult to contribute to improving the quality of medical care.

[0380] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for recognizing the emotional state of the user and using it for analysis, a means for adjusting the diagnosis presentation method based on the emotional state of the medical professional, and a means for collecting the final diagnosis result and feedback and using them as training data for the generative AI model. This enables diagnostic support that takes the emotional state into consideration, improving the accuracy of diagnosis and the quality of medical care.

[0381] 1. A "generative artificial intelligence model" is a system that uses machine learning algorithms to generate specific outputs from input data.

[0382] 2. "Patient-reported symptoms" refers to the patient's health condition or physical abnormalities reported to a medical professional.

[0383] 3. "Medical Image Data" means image data used for medical diagnosis or treatment, such as CT scans, MRIs, and X-ray images.

[0384] 4. "Test results" means data based on medical tests such as blood tests, urine tests, and tissue tests.

[0385] 5. "Diagnosis candidate" refers to a diagnosis result that the generative artificial intelligence model may present based on the analysis results.

[0386] 6. "Analysis results" refers to the conclusions or insights obtained by analyzing the input data.

[0387] 7. "Healthcare professionals" are doctors, nurses, clinical laboratory technicians, and other professionals who diagnose, treat, and care for patients.

[0388] 8. "User emotional state" means the mental or psychological state of a healthcare professional or patient.

[0389] 9. "Adjusting the diagnosis presentation method based on emotional state" refers to a method of adaptively changing the display and presentation of diagnosis results depending on the user's emotional state.

[0390] 10. "Final diagnosis" means the patient's final diagnosis determined by a medical professional.

[0391] 11. "Feedback" refers to information such as evaluations and opinions of the system and medical professionals' reactions to diagnostic results.

[0392] 12. “Training Data” means a data set used to train a machine learning model.

[0393] MODE FOR CARRYING OUT THE INVENTION

[0394] The present invention relates to a medical diagnosis support system, and in particular to a system that uses a generative artificial intelligence model and an emotion engine to analyze a patient's symptoms, medical image data, and test results, and presents diagnostic candidates while also taking into account the user's emotional state. The following describes in detail an embodiment of the present invention.

[0395] System configuration

[0396] The system of the present invention includes the following major components:

[0397] 1. Terminal: A device where the user (doctor) inputs patient information and receives analysis results and emotions. Specifically, input devices such as a PC or tablet can be used.

[0398] 2. Server: This is the central system that collects and analyzes data, uses generative AI models to present diagnostic candidates, and analyzes user emotions. This can be implemented on a cloud server or an on-premise server.

[0399] 3. Database: A storage system for storing patient data, analysis results, feedback, and emotion data. For example, an SQL database or a NoSQL database can be used.

[0400] 4. Emotion Engine: An algorithm for recognizing and analyzing the emotional state of users (doctors) and patients. This engine analyzes emotions using natural language processing and image recognition technology.

[0401] Example of a system

[0402] The present invention will be described in detail below based on a specific example of the operation of the system.

[0403] Entering patient information

[0404] The user (doctor) uses the terminal to input basic patient information, symptoms, medical image data, and test results. For example, for a patient complaining of abdominal pain, basic information, CT scan images, and blood test results are input. The emotion engine also detects the user's (doctor's) emotional state in real time and acquires data.

[0405] Sending data

[0406] The device sends the collected patient information and emotion data to the server using an HTTP request.

[0407] Data storage and preprocessing

[0408] The server stores the received data in a database, after which pre-processing of the data is performed, which includes data integrity checks, cleaning, and format conversion.

[0409] AI model for generating diagnostic candidates

[0410] The preprocessed data is input into a generative AI model on the server, which generates candidate diagnoses based on the patient's symptoms. At the same time, emotional data obtained from the emotion engine is also included in the analysis. For example, the generative AI model references past case data to generate candidate diagnoses such as "acute appendicitis," "intestinal obstruction," and "gastritis" for a patient complaining of abdominal pain.

[0411] Adjusting the display and presentation of results

[0412] The generated diagnosis candidates are sent from the server to the terminal. The terminal displays the diagnosis candidates in a format that is easy for the user (doctor) to understand. The emotion engine also adaptively adjusts the display method of the diagnosis candidates based on the emotion data acquired. For example, for a user with high stress, the diagnosis candidates are displayed in different colors to make them easier to understand visually.

[0413] Enter and submit feedback

[0414] The user (doctor) inputs the final diagnosis and feedback using a terminal and sends it to the server. This feedback data is used to retrain the generative AI model.

[0415] Prompt Sentence Examples

[0416] "Please list possible diagnoses based on the CT scan images and blood test results of a patient who visits the hospital with abdominal pain. Please suggest diagnoses based on medical evidence. Also, please adjust the display method according to the user's (doctor's) emotional state."

[0417] This invention allows doctors to quickly and accurately obtain supplementary information, reducing the risk of misdiagnosis. Furthermore, the use of an emotion engine enables diagnostic support that takes the user's mental state into account, which is expected to improve diagnostic accuracy and the quality of medical care. It is also expected that the accuracy of the system will continue to improve through collected feedback.

[0418] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0419] Step 1: Enter patient information

[0420] The terminal receives input from the user (doctor) of the patient's basic information, symptoms (e.g., abdominal pain), medical image data, and test results. The input includes the patient's basic information, CT scan images, blood test results, etc. Based on this, a basic information dataset is generated, and the emotion engine detects the user's (doctor's) emotional state (e.g., stress, fatigue) in real time and acquires it as additional data.

[0421] Specific behavior:

[0422] The user (doctor) enters the patient's name, age, gender, symptoms, CT scan images, and blood test results into the device's input form and presses the send button. The emotion engine analyzes the user's facial expressions and tone of voice and records their emotional state.

[0423] Step 2: Sending data

[0424] The terminal sends the collected patient information and emotion data to the server. As input, the reference dataset is sent by the terminal to the server. The data is sent using the HTTP request format.

[0425] Specific behavior:

[0426] The device collects all patient information and emotion data, structures it in JSON format, and sends it to the server using an HTTP POST request.

[0427] Step 3: Storing and Preprocessing Data

[0428] The server stores the received data in a database. The input includes the reference dataset sent from the terminal. Then, the data is preprocessed. The preprocessing includes data integrity check, cleaning, and format conversion. The output is a preprocessed clean dataset.

[0429] Specific behavior:

[0430] The server structures the incoming data and stores it in the database using SQL insert queries. After storing it, it detects missing values ​​in the data, imputes them appropriately, and then converts all data into a unified format.

[0431] Step 4: Generate candidate diagnoses using the AI ​​model

[0432] The server inputs the preprocessed data into a generative AI model to generate candidate diagnoses. The input includes a clean, preprocessed dataset. During the analysis, the AI ​​model references past case data and medical knowledge. The output is multiple candidate diagnoses.

[0433] Specific behavior:

[0434] The server runs a generative AI model using a deep learning framework such as TensorFlow or PyTorch, inputs the preprocessed data into the model, and the model analyzes the data to generate candidate diagnoses such as "acute appendicitis," "intestinal obstruction," and "gastritis."

[0435] Step 5: Adjust the display and presentation of results

[0436] The server organizes the generated candidate diagnoses and sends them to the terminal. The input includes the generated candidate diagnoses. The terminal displays the candidate diagnoses and adjusts the display method based on the emotional state of the user (doctor). The output is the adjusted display of the candidate diagnoses.

[0437] Specific behavior:

[0438] The server sends the generated diagnosis candidates in JSON format to the terminal as an HTTP response. The terminal receives this and displays the diagnosis candidates through a GUI. Based on the results of the emotion engine analysis, if the doctor is in a state of high stress, the diagnosis candidates are color-coded to make them easier to understand visually.

[0439] Step 6: Enter and submit your feedback

[0440] The terminal allows the user (doctor) to input the final diagnosis and feedback, and sends the data to the server. The input includes the final diagnosis and feedback. The output is the feedback data sent to the server.

[0441] Specific behavior:

[0442] The user (doctor) enters the final diagnosis (e.g., acute appendicitis) and feedback about the system's diagnostic accuracy into the input form on the terminal, and presses the submit button. The terminal structures this data in JSON format and sends it to the server using an HTTP POST request.

[0443] Step 7: Save feedback and retrain

[0444] The server stores the received feedback data in a database and uses it to retrain the generative AI model. The feedback data is included as input. The output is an updated generative AI model.

[0445] Specific behavior:

[0446] The server adds and stores the feedback data in a database, then retrains the generative AI model using a new dataset that includes the feedback data, e.g., updating model parameters to improve the model's performance.

[0447] (Application example 2)

[0448] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0449] The problem that this invention aims to solve is to improve the accuracy and efficiency of diagnoses by reducing the risk of misdiagnosis caused by information overload, time constraints, and the doctor's own emotional state when making a diagnosis in the medical field. Another problem is to improve the safety and efficiency of work by analyzing the emotional state of security operators in real time and utilizing that information when generating alerts.

[0450] 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.

[0451] In this invention, the server includes means for analyzing patient-reported symptoms, medical image data, and test results using a generative AI model and presenting diagnosis candidates, means for displaying the analysis results to medical professionals, means for collecting final diagnosis results and feedback from medical professionals and using them as training data for the generative AI model, means for collecting and analyzing biometric data in real time, and means for generating and displaying alerts based on the analysis results and biometric data, thereby enabling improved accuracy and efficiency of diagnoses in medical settings and improved safety and efficiency in security operations.

[0452] A "generative artificial intelligence model" refers to an algorithm or machine learning model that analyzes data collected in medical diagnosis or security analysis and generates diagnostic candidates or anomaly detection results.

[0453] "Patient-reported symptoms" refers to information about specific symptoms such as poor health or pain that patients self-report in medical settings.

[0454] "Medical image data" refers to image information of a patient's inside the body obtained using medical equipment such as CT scans, X-ray images, and MRIs.

[0455] "Test results" refers to numerical data and information indicating whether or not there are abnormalities obtained from blood tests, urine tests, various biochemical tests, etc.

[0456] "Analysis results" refers collectively to the diagnosis candidates and anomaly detection results obtained after analyzing input data using a generative artificial intelligence model.

[0457] "Medical professionals" refers to professionals who perform medical-related tasks, such as doctors, nurses, and radiological technologists.

[0458] "Biometric data" refers to information related to the physical and mental state of medical professionals, security operators, etc., such as heart rate, body temperature, and facial expression data.

[0459] An "alert" refers to a warning or caution message issued by the system based on analysis results or abnormal values ​​in biometric data.

[0460] "Database" refers to a system that organizes and stores large amounts of information, such as reported symptoms, medical image data, test results, and analysis results, and enables quick search and retrieval as needed.

[0461] "Preprocessing" refers to the process of checking the integrity of collected data, cleaning it, and converting its format to make it suitable for subsequent analysis.

[0462] "Feedback" refers to the evaluations and opinions of medical professionals and security operators regarding the final diagnosis results and system proposals.

[0463] The embodiments for carrying out the present invention will be described below.

[0464] System Overview

[0465] The system of the present invention includes the following major components:

[0466] 1. Terminal: A device used by the user (doctor or security operator) to input and display basic information, biometric data, analysis results, and emotional state of the patient.

[0467] 2. Server: A central system that collects and analyzes data, generates diagnostic candidates and anomaly detection results using generative AI models, integrates the analysis results with biometric data, and generates appropriate alerts.

[0468] 3. Database: A storage system for storing reported symptoms, medical image data, test results, analysis results, feedback, emotional data, etc.

[0469] 4. Emotion engine: An algorithm that recognizes and analyzes the emotional state of the user (doctor or security operator).

[0470] Data collection

[0471] The terminal collects basic information about the patient and operator, as well as biometric data (heart rate, body temperature, facial expression data) from the user, and also collects video data and medical image data from security cameras and medical equipment.

[0472] Data analysis

[0473] The server stores the received data in a database and performs preprocessing. This involves checking the data for consistency, cleaning it, and converting its format. Once this is complete, the data is input into the generative AI model for analysis. At the same time, the emotion data obtained by the emotion engine is used for analysis.

[0474] Alert Generation

[0475] Based on the analysis results and emotion data, the server generates appropriate alerts. For example, if an abnormality is detected and an operator is in a high stress state at the same time, an alert is displayed indicating that immediate action is required. The alert is sent to the terminal and displayed to the user.

[0476] Feedback collection

[0477] Feedback on analysis results and alerts is collected from users. This feedback is sent to the server and stored in a database. The collected feedback is used as retraining data for the generative AI model and is used to improve the accuracy of the system.

[0478] Specific examples

[0479] When a doctor uses a device to enter basic patient information, CT scan images, and blood test results, the data is sent to a server. The server preprocesses the data and analyzes it using a generative AI model. As a result, a candidate diagnosis is generated and displayed on the device. At the same time, an emotion engine analyzes the doctor's emotional state and adjusts the way the candidate diagnosis is presented based on that data.

[0480] If a security operator wears smart glasses during the night shift and the system simultaneously detects an abnormality and high stress, an alert message will be displayed stating, "High stress and abnormality detected! Immediate action required!"

[0481] Example prompts to input to the generative AI model

[0482] "If an anomaly is detected, how can we tailor our response to reduce operator burden?"

[0483] "A method for detecting operator stress levels in real time and proposing countermeasures during security operations."

[0484] In this way, the present invention enables efficient information presentation and alert generation that integrates analysis results and emotional data in medical settings and security operations.

[0485] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0486] Step 1:

[0487] Data Collection Phase

[0488] Input: Basic information from the user (doctor or security operator), biometric data (heart rate, body temperature, facial expression data), video data and medical image data from security cameras and medical equipment.

[0489] Output: Sends collected data to a server.

[0490] Specific behavior:

[0491] The user uses the terminal to input basic patient information and biometric data, and data is collected in real time from security cameras and medical devices, which are then transmitted to a server.

[0492] Step 2:

[0493] Database save phase

[0494] Input: Basic information, biometric data, video data, and medical image data sent to the server.

[0495] Output: Various data stored in the database.

[0496] Specific behavior:

[0497] The server stores the received data in a database, where it checks for consistency.

[0498] Step 3:

[0499] Preprocessing Phase

[0500] Input: Basic information, biometric data, video data, and medical image data stored in the database.

[0501] Output: The preprocessed data.

[0502] Specific behavior:

[0503] The server retrieves data from the database, checks its integrity, cleans it, and converts it into a format suitable for generative AI models.

[0504] Step 4:

[0505] Data analysis phase

[0506] Input: Preprocessed basic information, biometric data, video data, and medical image data.

[0507] Output: Diagnosis candidates, anomaly detection results, and sentiment analysis results from the generative AI model.

[0508] Specific behavior:

[0509] The server inputs the preprocessed data into the generative AI model for analysis, while simultaneously analyzing and integrating the user's emotional data obtained by the emotion engine.

[0510] Step 5:

[0511] Alert Generation Phase

[0512] Input: Diagnosis candidates, anomaly detection results, and sentiment analysis results.

[0513] Output: The alert message.

[0514] Specific behavior:

[0515] The server generates appropriate alerts based on the analysis results and biometric data. For example, if an abnormality is detected and an operator is in a high stress state, an alert message requiring immediate action is generated. The generated alert is then sent to the terminal.

[0516] Step 6:

[0517] Display Phase

[0518] Input: Alert message, diagnosis candidates, and sentiment analysis results sent from the server.

[0519] Output: Alert message and potential diagnosis displayed on the terminal.

[0520] Specific behavior:

[0521] The device receives alert messages and diagnostic candidates from the server and displays them to the user, adjusting the display format according to the results of emotion analysis.

[0522] Step 7:

[0523] Feedback gathering phase

[0524] Input: Final diagnostic results and feedback from the user.

[0525] Output: Feedback sent to the server.

[0526] Specific behavior:

[0527] The user (doctor or security operator) inputs feedback on the final diagnosis and the evaluation of the alert through the terminal, and this data is sent to the server and stored in the database.

[0528] Step 8:

[0529] Model retraining phase

[0530] Input: Stored feedback, newly collected data.

[0531] Output: A generative artificial intelligence model with improved accuracy.

[0532] Specific behavior:

[0533] The server retrains the generative AI model based on collected feedback and newly acquired data, thereby continuously improving diagnostic accuracy and anomaly detection performance.

[0534] This concludes the explanation of each processing step of this program and its specific operations.

[0535] 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.

[0536] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0537] 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.

[0538] [Second embodiment]

[0539] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0540] 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.

[0541] 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).

[0542] 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.

[0543] 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.

[0544] 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).

[0545] 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.

[0546] 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.

[0547] 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.

[0548] 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.

[0549] 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.

[0550] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0551] The embodiments for carrying out the present invention will be described below.

[0552] System Overview

[0553] The present invention relates to a medical diagnosis support system, in particular a system that uses a generative artificial intelligence model to analyze a patient's symptoms, medical image data, and test results, and presents potential diagnoses. The system includes the following main components:

[0554] 1. Terminal: A device where the user (doctor) enters patient information and receives analysis results.

[0555] 2. Server: The central system that collects and analyzes data and uses generative AI models to present potential diagnoses.

[0556] 3. Database: A storage system for storing patient data, analysis results, and feedback.

[0557] Explaining program processing in natural language

[0558] Data Collection Phase

[0559] The terminal receives patient information from the user (doctor) and transmits it to the server. Specifically, this information includes the patient's basic information, reported symptoms, medical image data, and test results.

[0560] Example: A user (doctor) uses a terminal to input basic information about a patient complaining of abdominal pain, CT scan images, and blood test results, and sends them to a server.

[0561] Data analysis phase

[0562] The server stores the received data in a database and performs preprocessing, which includes checking data consistency, cleaning, and format conversion. Once preprocessing is complete, the data is input into a generative AI model for analysis.

[0563] Example: The server preprocesses data received about a patient's abdominal pain and inputs it into a generative AI model. The generative AI model analyzes the data based on past cases of abdominal pain and medical knowledge to generate candidate diagnoses.

[0564] Diagnosis candidate presentation phase

[0565] The server organizes the diagnostic candidates obtained from the generative AI model and sends them to the device used by the user (doctor). The device displays the diagnostic candidates in a format that is easy for the user to understand.

[0566] Example: The server generates diagnostic candidates such as "acute appendicitis" and "intestinal obstruction" and sends them to the terminal so that the user (doctor) can check them.

[0567] Feedback gathering phase

[0568] The device allows the user (doctor) to input the final diagnosis and feedback. This data is sent to the server and stored in a database. This feedback is then used to retrain the generative AI model, contributing to improving the accuracy of the system.

[0569] For example, a user (doctor) diagnoses "acute appendicitis" and provides feedback that the system's diagnosis was appropriate. This information is sent to the server and used as training data for the generative AI model.

[0570] Specific operation example

[0571] When a patient visits the hospital with abdominal pain, the user (doctor) uses a terminal to input the patient's basic information, CT scan images, blood test results, and other test results. This data is sent to the server and stored in a database. The server preprocesses the data and inputs it into a generative AI model for analysis. The generated diagnosis candidates are sent to the terminal and displayed to the user (doctor). After the final diagnosis is made, feedback is sent from the terminal to the server, stored in a database, and used again as training data for the generative AI model.

[0572] The system provides doctors with fast and accurate supplementary information, reducing the risk of misdiagnosis, and continuously improves its accuracy through collected feedback.

[0573] The processing flow will be explained below.

[0574] Specific explanation of the program's processing steps

[0575] Data Collection Phase

[0576] Step 1:

[0577] The terminal receives basic patient information from the user (doctor), including the patient's name, age, sex, and other basic information.

[0578] pseudo

[0579] The terminal receives basic patient information from the user.

[0580] Step 2:

[0581] The terminal receives input from the user (doctor) of the patient's reported symptoms, including the specific symptoms and progress of the patient's awareness.

[0582] pseudo

[0583] The terminal inputs the patient's reported symptoms.

[0584] Step 3:

[0585] The terminal receives input of medical image data (e.g., X-rays, CT scans) and test results (e.g., blood test results, urine test results) from the user (doctor).

[0586] pseudo

[0587] The terminal inputs medical image data and test results

[0588] Step 4:

[0589] The device sends all collected data to the server.

[0590] pseudo

[0591] The device sends the input data to the server

[0592] Data analysis phase

[0593] Step 5:

[0594] The server receives the patient data transmitted from the terminal.

[0595] pseudo

[0596] The server receives data from the device

[0597] Step 6:

[0598] The server stores the received patient data in a database.

[0599] pseudo

[0600] The server stores the data in a database

[0601] Step 7:

[0602] The server checks the integrity of the stored data and performs pre-processing (data cleaning, format conversion, etc.).

[0603] pseudo

[0604] The server performs data integrity checks and preprocessing

[0605] Step 8:

[0606] The server inputs the preprocessed data into the generative AI model.

[0607] pseudo

[0608] The server inputs the preprocessed data into the generative AI model

[0609] Step 9:

[0610] The server obtains the candidate diagnosis results from the generative AI model.

[0611] pseudo

[0612] The server receives the analysis results from the generated AI model

[0613] Diagnosis candidate presentation phase

[0614] Step 10:

[0615] The server formats the acquired diagnosis candidates into a format that is easy for the user (doctor) to understand.

[0616] pseudo

[0617] The server formats the candidate diagnoses

[0618] Step 11:

[0619] The server sends the formatted diagnosis candidates to the terminal and presents them to the user (doctor).

[0620] pseudo

[0621] The server sends the diagnosis candidate to the terminal.

[0622] The device displays possible diagnoses to the user (doctor)

[0623] Feedback gathering phase

[0624] Step 12:

[0625] The terminal allows the user (doctor) to input the final diagnosis result.

[0626] pseudo

[0627] The terminal receives the final diagnosis result from the user.

[0628] Step 13:

[0629] The terminal allows the user (doctor) to input feedback on the system's diagnostic accuracy and areas for improvement.

[0630] pseudo

[0631] The device receives feedback from the user

[0632] Step 14:

[0633] The terminal sends the final diagnosis result and feedback to the server.

[0634] pseudo

[0635] The device sends the final diagnosis results and feedback to the server

[0636] Step 15:

[0637] The server stores the final diagnosis results and feedback in a database.

[0638] pseudo

[0639] The server stores the final diagnosis results and feedback in a database

[0640] Step 16:

[0641] The server uses the collected feedback to retrain the generative AI model.

[0642] pseudo

[0643] The server uses the feedback to retrain the generative AI model

[0644] Through this series of processing flows, the system of the present invention can assist users (doctors) in making diagnoses, improving diagnostic accuracy while reducing the risk of misdiagnosis. Furthermore, through continuous improvements based on feedback, the reliability and effectiveness of the system can be increased.

[0645] Example 1

[0646] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0647] Conventional medical diagnosis support systems have difficulty efficiently analyzing patients' reported symptoms, medical image data, and test results, and presenting appropriate diagnostic candidates. Furthermore, there is a lack of a method for effectively aggregating feedback from medical professionals during the process of making a final diagnosis and improving the accuracy of the diagnosis support system. This increases the risk of misdiagnosis and impacts the quality of medical care.

[0648] 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.

[0649] In this invention, the server includes a means for transmitting patient information input from a terminal to the server, a means for the server to store the received data in a database and perform data preprocessing, a means for analyzing the patient's reported symptoms, medical image data, and test results using a generative AI model and presenting candidate diagnoses, a means for displaying the analysis results to medical professionals, and a means for collecting final diagnoses and feedback from medical professionals and using them as training data for the generative AI model. This enables efficient collection and preprocessing of patient data, presentation of appropriate candidate diagnoses, collection of feedback, and improvement of system accuracy.

[0650] A "generative artificial intelligence model" is a model generated using artificial intelligence technology that analyzes a patient's symptoms and test results based on medical data and generates potential diagnoses.

[0651] "Reported symptoms" are physical discomfort or abnormalities that patients report to medical professionals, and are information that forms the basis of diagnosis.

[0652] "Medical image data" refers to image data obtained using medical equipment, such as CT scans, MRIs, and X-rays, that provide the visual information necessary for diagnosis.

[0653] "Test results" refer to the numerical data and analytical results obtained through various medical tests, such as blood tests and urine tests, and are important information that assists in diagnosis.

[0654] "Candidate diagnoses" are a list of possible diagnoses presented by the generative AI model as a result of its analysis, based on the patient's symptoms and test results, and serve as a reference for medical professionals to make a final decision.

[0655] A "terminal" is a device used by a user (doctor) to enter patient information and receive analysis results, and includes PCs, tablets, smartphones, etc.

[0656] The "server" is a central system that receives, stores, and preprocesses patient information, analyzes it using a generative AI model, and presents diagnostic candidates, and also connects to the database.

[0657] A "database" is a storage system for systematically storing patient data, analysis results, and feedback from medical professionals, and is used for subsequent data reference and analysis.

[0658] "Preprocessing" refers to the process of preparing the data to be analyzed, and includes checking data consistency, cleaning, format conversion, etc.

[0659] "Feedback" refers to opinions and evaluations provided by medical professionals based on final diagnosis results, and is data used to retrain generative AI models.

[0660] System Overview

[0661] This invention relates to a medical diagnosis support system that uses a generative AI model to analyze patient-reported symptoms, medical image data, and test results, and presents potential diagnoses. The system includes the following main components:

[0662] 1. Terminal: A device where the user (doctor) enters patient information and receives analysis results. This includes PCs, tablets, smartphones, etc.

[0663] 2. Server: This is the central system that collects and analyzes data, uses generative AI models to present diagnostic candidates, and connects to the database.

[0664] 3. Database: A storage system for storing patient data, analysis results, and feedback.

[0665] Explaining program processing in natural language

[0666] Data Collection Phase

[0667] A user (doctor) uses a terminal to input patient information and sends it to a server. Specifically, this includes the patient's basic information (name, age, gender), reported symptoms, medical image data (CT scans and MRIs), and various test results (blood tests, urine tests, etc.). For example, the user (doctor) inputs the information of a 30-year-old male patient complaining of abdominal pain into the terminal and sends it to the server.

[0668] Data analysis phase

[0669] The server stores the received data in a database and then performs preprocessing on the data. Preprocessing includes checking data integrity, cleaning, and any necessary format conversions. Once preprocessed, the data is input into a generative AI model. The server preprocesses the patient information and inputs it into a generative AI model built using TensorFlow and PyTorch for analysis. The generative AI model performs analysis based on past case data and medical knowledge to generate candidate diagnoses.

[0670] Diagnosis candidate presentation phase

[0671] The server organizes the generated candidate diagnoses and sends them to the terminal. The terminal displays the candidate diagnoses in a format that is easy for the user (doctor) to understand. For example, the terminal displays "acute appendicitis" and "intestinal obstruction" as part of the list of candidate diagnoses generated by the server.

[0672] Feedback gathering phase

[0673] The user (doctor) enters the final diagnosis and feedback into the device and sends it to the server. The feedback is then stored in a database and used to retrain the generative AI model. This feedback continuously improves the accuracy of the system. For example, if the user (doctor) diagnoses "acute appendicitis" and enters feedback that the diagnosis was appropriate, this is sent to the server and used as training data for the generative AI model.

[0674] Prompt Sentence Examples

[0675] "Generate possible diagnoses based on the CT scan images and blood test results of a 30-year-old man complaining of abdominal pain."

[0676] This system will enable doctors to obtain supplementary information quickly and accurately, reducing the risk of misdiagnosis. Furthermore, it is expected that the accuracy of the generative AI model will be improved through collected feedback.

[0677] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0678] Step 1:

[0679] The terminal accepts input from the user (doctor).

[0680] Specific operation: The user (doctor) uses the terminal to enter the patient's basic information (name, age, gender), reported symptoms, medical image data (CT scan or MRI), and various test results (blood test, urine test, etc.).

[0681] Input: Patient's basic information, reported symptoms, medical image data, and test results.

[0682] Output: The entered patient information.

[0683] Step 2:

[0684] The terminal transmits the input information to the server.

[0685] Specific operation: The terminal encrypts the entered patient data and sends it to the server using the TLS protocol.

[0686] Input: Patient information entered into the terminal.

[0687] Output: Patient information sent to the server.

[0688] Step 3:

[0689] The server stores the received data in a database.

[0690] Specific operation: The server adds the received patient data to the "Patient Data" table in the database as a new record.

[0691] Input: Patient information received by the server.

[0692] Output: Patient information stored in the database.

[0693] Step 4:

[0694] The server pre-processes the data.

[0695] Specific operation: The server checks the integrity of the received data, removes invalid values, fills in missing data, and converts the format.

[0696] Input: Raw data stored in a database.

[0697] Output: Consistency checked and preprocessed data.

[0698] Step 5:

[0699] The server inputs the preprocessed data into the generative AI model.

[0700] Specific operation: The server inputs the preprocessed data into a generative AI model using TensorFlow or PyTorch.

[0701] Input: Preprocessed data.

[0702] Output: The data that is input into a generative AI model.

[0703] Step 6:

[0704] A generative AI model analyzes the data and generates candidate diagnoses.

[0705] How it works: The generative AI model analyzes the pre-processed data and generates candidate diagnoses based on past case data and medical knowledge.

[0706] Input: Preprocessed data fed into a generative AI model.

[0707] Output: The generated candidate diagnoses.

[0708] Step 7:

[0709] The server transmits the generated candidate diagnoses to the terminal.

[0710] Specific operation: The server organizes the generated diagnosis candidate list and sends it to the terminal.

[0711] Input: A list of diagnostic candidates output from a generative AI model.

[0712] Output: A list of diagnostic candidates sent to the terminal.

[0713] Step 8:

[0714] The terminal displays the possible diagnoses to the user (doctor).

[0715] Specific operation: The terminal displays the list of diagnosis candidates in a format (for example, a list format) that is easy for the user (doctor) to understand.

[0716] Input: A list of diagnostic candidates sent to the terminal.

[0717] Output: A list of possible diagnoses displayed to the user (doctor).

[0718] Step 9:

[0719] The user (doctor) inputs the diagnosis and feedback.

[0720] Specific operation: The user (doctor) inputs the final diagnosis and feedback to the system into the terminal.

[0721] Input: User's (doctor's) final diagnosis and feedback.

[0722] Output: Feedback information entered into the terminal.

[0723] Step 10:

[0724] The terminal sends feedback information to the server.

[0725] Specific operation: The terminal encrypts the input feedback information and sends it to the server.

[0726] Input: Feedback information entered into the device.

[0727] Output: Feedback information sent to the server.

[0728] Step 11:

[0729] The server stores the feedback information in a database.

[0730] Specific operation: The server stores the feedback information in a "diagnostic feedback" table in the database.

[0731] Input: Feedback information received by the server.

[0732] Output: Feedback information stored in a database.

[0733] Step 12:

[0734] The server retrains the generative AI model based on the feedback.

[0735] Specific operation: The server periodically extracts feedback information from the database and retrains the generative AI model.

[0736] Input: Feedback information stored in the database.

[0737] Output: The retrained generative AI model.

[0738] (Application example 1)

[0739] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0740] In modern medical settings, doctors are required to make quick and accurate diagnoses within a limited time frame, but this is often difficult. While selecting the right medicine is also important for pharmacies and drugstores, it can be difficult to accurately understand a patient's symptoms and provide appropriate advice. Therefore, there is a need for a system that can support doctors and pharmacists in the diagnostic process.

[0741] 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.

[0742] In this invention, the server includes means for analyzing patient-reported symptoms, medical image data, and test results using a generative AI model and presenting candidate diagnoses, means for displaying the analysis results to users such as medical professionals or patients, means for collecting final diagnosis results and feedback from users and using them as training data for the generative AI model, and means for using the medical diagnosis support system in physical stores such as drugstores. This allows doctors and pharmacists to quickly and accurately present candidate diagnoses, enabling them to select appropriate medicines and medical procedures.

[0743] A "generative artificial intelligence model" is an artificial intelligence algorithm used to analyze data such as patient symptoms, medical images, and test results to generate potential diagnoses.

[0744] "Reported symptoms" refers to any ill health or specific symptoms reported by the patient.

[0745] "Medical image data" refers to image data taken using medical equipment such as CT scans and MRIs.

[0746] "Test results" refers to data obtained from various medical tests, such as blood tests and urine tests.

[0747] "Analysis results" refer to the diagnostic candidates and related information obtained using the generative artificial intelligence model.

[0748] "Healthcare professionals" refer to professionals involved in diagnosing and treating patients, such as doctors and nurses.

[0749] "Feedback" refers to the final diagnostic results and evaluations and opinions from users.

[0750] "Training Data" means data used to improve and optimize a generative artificial intelligence model.

[0751] A "medical diagnosis support system" is a system that uses a generative artificial intelligence model to present potential diagnoses to medical professionals and patients, helping them make a final diagnosis.

[0752] The embodiments for carrying out the present invention will be described below.

[0753] System Overview

[0754] The present invention relates to a medical diagnosis support system, in particular a system that uses a generative artificial intelligence model to analyze a patient's symptoms, medical image data, and test results, and presents potential diagnoses. The system includes the following main components:

[0755] 1. Terminal: A device where users (doctors and patients) enter symptoms and test results and receive analysis results. Specifically, a smartphone, tablet, or interactive display is used.

[0756] 2. Server: This is the central system that collects, analyzes, and stores data, and uses generative AI models to present diagnostic candidates. A cloud server (e.g., Amazon Web Services or Microsoft Azure) is used.

[0757] 3. Database: A storage system for storing patient data, analysis results, and feedback. SQL / NoSQL databases (MySQL, MongoDB, PostgreSQL) are used.

[0758] Program Processing and Technical Details

[0759] Data Collection Phase

[0760] Users (doctors or patients) use a terminal to input their symptoms, medical images, test results, etc. For example, a patient complaining of abdominal pain inputs their age, gender, CT scan images, and blood test results. This data is then sent to a server via the Internet.

[0761] Data analysis phase

[0762] The server stores the received data in a database and performs preprocessing (integrity checks, cleaning, format conversion) on the data. After preprocessing is complete, the data is input into a generative artificial intelligence model for analysis. This process uses machine learning libraries such as TensorFlow and PyTorch.

[0763] Diagnosis candidate presentation phase

[0764] The candidate diagnoses obtained from the analysis results are organized on the server and sent to the terminal, which then displays them in a format that is easy for the user to understand. For example, candidate diagnoses such as "acute appendicitis" and "intestinal obstruction" are displayed.

[0765] Feedback gathering phase

[0766] The final diagnosis and feedback are entered on the device and sent to the server, where they are stored in a database and used to retrain the generative AI model.

[0767] Specific operation example

[0768] When a patient visits a hospital or drugstore with abdominal pain, the user (doctor or pharmacist) uses a terminal to enter the patient's basic information, CT scan images, blood test results, and other test results. For example, the user might enter, "I have abdominal pain and have had similar symptoms several times in the past. I am 45 years old. I also have CT scan images. Please let me know if you have any diagnostic candidates." This data is sent to a server and stored in a database. The server preprocesses the data and inputs it into a generative AI model for analysis. The generated diagnostic candidates are sent to the terminal and displayed to the user (doctor or patient). After the final diagnosis is made, feedback is sent from the terminal to the server, stored in a database, and used as training data for the generative AI model.

[0769] The system allows users to quickly and accurately identify potential diagnoses, reducing the risk of misdiagnosis, and continuously improves its accuracy through collected feedback.

[0770] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0771] Step 1:

[0772] The terminal receives input from the user (doctor or patient) about symptoms, basic information, medical image data, and test results. The input data includes age, gender, and details of symptoms. This input data is sent to the server via the Internet. An example of a prompt for the user to input specific information is, "I have abdominal pain and have had similar symptoms several times in the past. I am 45 years old. I also have a CT scan image. Please let me know if you have any possible diagnoses."

[0773] Step 2:

[0774] The server receives data sent from the terminal and stores it in a database. The stored data may not be suitable for analysis as is, so it must be preprocessed. Specific preprocessing steps include checking data consistency, filling in missing data, and standardizing formats. This process ensures the availability and reliability of input data.

[0775] Step 3:

[0776] The server inputs the preprocessed data into the artificial intelligence model. Specifically, machine learning libraries such as TensorFlow and PyTorch are used to input the data into the model. The model then performs analysis based on the input data, using past case data and a medical knowledge base. As a result of the analysis, candidate diagnoses are generated.

[0777] Step 4:

[0778] The server organizes the generated candidate diagnoses and sends them to the terminal. The terminal displays the analysis results in an easy-to-read format for the user (doctor or patient). At this stage, an appropriate user interface design is important to present the analysis results in an easy-to-understand format. For example, specific candidate diagnoses such as "acute appendicitis" or "intestinal obstruction" are displayed.

[0779] Step 5:

[0780] The user uses the terminal to input the final diagnosis results and feedback. The user also inputs feedback on the diagnosis performed based on the confirmed diagnosis candidates and their results into the terminal. This includes an evaluation of the accuracy of the diagnosis and the information presented. This feedback data is sent to the server.

[0781] Step 6:

[0782] The server stores the final diagnosis results and feedback sent by the user in a database and uses them as training data for the generative AI model. By retraining the generative AI model based on this feedback data, the accuracy of the model is continuously improved, further increasing the reliability and usefulness of the system.

[0783] Through the above processing steps, this medical diagnosis support system is able to efficiently and accurately analyze medical data and present potential diagnoses.

[0784] 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.

[0785] The embodiments for carrying out the present invention will be described below.

[0786] System Overview

[0787] The present invention relates to a medical diagnosis support system, in particular a system that uses a generative artificial intelligence model and an emotion engine to analyze patient symptoms, medical image data, and test results, and presents potential diagnoses while also taking into account the user's emotional state. The system includes the following main components:

[0788] 1. Terminal: A device where the user (doctor) inputs patient information and receives analysis results and emotions.

[0789] 2. Server: The central system that collects and analyzes data, uses generative AI models to present diagnostic candidates, and analyzes user sentiment.

[0790] 3. Database: A storage system for storing patient data, analysis results, feedback and emotion data.

[0791] 4. Emotion engine: Algorithms for recognizing and analyzing the emotional state of the user (doctor) and patient.

[0792] Explaining program processing in natural language

[0793] Data Collection Phase

[0794] The terminal receives basic patient information from the user (doctor) and sends it to the server. This information includes the patient's basic information, reported symptoms, medical image data, and test results. The emotion engine also recognizes the user's (doctor's) emotional state and captures it as data.

[0795] Example: A user (doctor) uses a terminal to input basic information about a patient complaining of abdominal pain, CT scan images, and blood test results, and at the same time, the emotion engine detects the user's (doctor's) stress level and transmits the data.

[0796] Data analysis phase

[0797] The server stores the received data in a database and performs preprocessing. Preprocessing includes checking data consistency, cleaning, and format conversion. Once preprocessed, the data is input into a generative AI model for analysis. Furthermore, emotional data of users and patients obtained by the emotion engine is also used in the analysis.

[0798] Example: The server preprocesses data received about a patient's abdominal pain and inputs it into a generative AI model. The generative AI model analyzes the data based on past cases of abdominal pain and medical knowledge to generate a candidate diagnosis. At the same time, it evaluates the possibility that the doctor's emotional state may affect the diagnosis based on the results of the emotion engine analysis.

[0799] Diagnosis candidate presentation phase

[0800] The server organizes the acquired diagnosis candidates and sends them to the device used by the user (doctor). The device displays the diagnosis candidates and emotion analysis results in a format that is easy for the user (doctor) to understand. The presentation method of the diagnosis candidates is also adaptively changed depending on the emotion analysis results.

[0801] For example, the server generates diagnostic candidates such as "acute appendicitis" or "intestinal obstruction" and sends them to the terminal for the user (doctor) to review. At the same time, the emotion engine adjusts the way the diagnostic candidates are displayed based on the emotional state recognized by the server.

[0802] Feedback gathering phase

[0803] The device allows the user (doctor) to input the final diagnosis, feedback, and emotional data. This data is sent to the server and stored in a database. This feedback is then used to retrain the generative AI model, contributing to improving the accuracy of the system.

[0804] Example: A user (doctor) diagnoses "acute appendicitis" and provides feedback that the system's diagnosis was appropriate. This information is sent to the server and used as training data for the generative AI model. The emotional state collected by the emotion engine is also stored.

[0805] Specific operation example

[0806] When a patient visits the hospital with abdominal pain, the user (doctor) uses a terminal to input the patient's basic information, CT scan images, blood test results, and other test results. This data is sent to the server and stored in a database. The server preprocesses the data and inputs it into a generative AI model for analysis. The generated diagnosis candidates are sent to the terminal and displayed to the user (doctor). After the final diagnosis is made, feedback is sent from the terminal to the server, stored in a database, and used again as training data for the generative AI model. At the same time, an emotion engine detects the user's emotional state and uses this data for analysis.

[0807] This system allows doctors to obtain supplementary information quickly and accurately, reducing the risk of misdiagnosis. Furthermore, the use of an emotion engine enables diagnostic support that takes the user's mental state into account, which is expected to improve diagnostic accuracy and the quality of medical care. The accuracy of the system is continuously improved through collected feedback.

[0808] The processing flow will be explained below.

[0809] Specific explanation of the program's processing steps

[0810] Data Collection Phase

[0811] Step 1:

[0812] The terminal receives basic patient information from the user (doctor), including basic information such as the patient's name, age, sex, height, and weight.

[0813] pseudo

[0814] The terminal receives basic patient information from the user.

[0815] Step 2:

[0816] The terminal receives input from the user (doctor) of the patient's reported symptoms, including the specific symptoms and progress of the symptoms that the patient is aware of.

[0817] pseudo

[0818] The terminal inputs the patient's reported symptoms.

[0819] Step 3:

[0820] The terminal receives input of medical image data (e.g., X-rays, CT scans) and test results (e.g., blood test results, urine test results) from the user (doctor).

[0821] pseudo

[0822] The terminal inputs medical image data and test results

[0823] Step 4:

[0824] The emotion engine recognizes the user's (doctor's) emotional state and captures it as data, including the doctor's stress level and fatigue level.

[0825] pseudo

[0826] Emotion engine recognizes the doctor's emotional state

[0827] Step 5:

[0828] The device sends all collected data to the server.

[0829] pseudo

[0830] The device sends the input data to the server

[0831] Data analysis phase

[0832] Step 6:

[0833] The server receives the patient data and emotion data transmitted from the terminal.

[0834] pseudo

[0835] The server receives data from the device

[0836] Step 7:

[0837] The server stores the received patient data in a database.

[0838] pseudo

[0839] The server stores the data in a database

[0840] Step 8:

[0841] The server checks the integrity of the stored data and performs pre-processing (data cleaning, format conversion, etc.).

[0842] pseudo

[0843] The server performs data integrity checks and preprocessing

[0844] Step 9:

[0845] The server inputs the preprocessed data into a generative AI model, including emotional data.

[0846] pseudo

[0847] The server inputs the preprocessed data into the generative AI model

[0848] Step 10:

[0849] The server obtains the candidate diagnosis results from the generative AI model.

[0850] pseudo

[0851] The server receives the analysis results from the generated AI model

[0852] Diagnosis candidate presentation phase

[0853] Step 11:

[0854] The server formats the diagnosis candidates obtained into a format that is easy for the user (doctor) to understand, and adjusts the results by reflecting emotional data.

[0855] pseudo

[0856] The server formats the candidate diagnoses

[0857] The server adjusts the results by taking into account the emotional data.

[0858] Step 12:

[0859] The server sends the formatted diagnosis candidates to the terminal and presents them to the user (doctor).

[0860] pseudo

[0861] The server sends the diagnosis candidate to the terminal.

[0862] The device displays possible diagnoses to the user (doctor)

[0863] Feedback gathering phase

[0864] Step 13:

[0865] The terminal allows the user (doctor) to input the final diagnosis result.

[0866] pseudo

[0867] The terminal receives the final diagnosis result from the user.

[0868] Step 14:

[0869] The device allows the user (doctor) to input feedback on the system's diagnostic accuracy and areas for improvement, along with their emotional state.

[0870] pseudo

[0871] The device receives feedback from the user

[0872] The device also collects emotional state data

[0873] Step 15:

[0874] The terminal sends the final diagnosis result and feedback to the server.

[0875] pseudo

[0876] The device sends the final diagnosis results and feedback to the server

[0877] Step 16:

[0878] The server stores the final diagnosis results and feedback in a database.

[0879] pseudo

[0880] The server stores the final diagnosis results and feedback in a database

[0881] Step 17:

[0882] The server uses the collected feedback to retrain the generative AI model, including the emotional data.

[0883] pseudo

[0884] The server uses the feedback to retrain the generative AI model

[0885] Through this series of processing flows, the system of the present invention provides diagnostic support to the user (doctor), improving diagnostic accuracy while reducing the risk of misdiagnosis. Furthermore, by using an emotion engine, diagnostic support that takes into account the user's mental state is possible, which is expected to improve diagnostic accuracy and the quality of medical care. The accuracy of the system is continuously improved through collected feedback.

[0886] Example 2

[0887] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0888] Traditionally, medical diagnoses have relied primarily on the doctor's experience and knowledge, leading to potential misdiagnoses and oversights. Furthermore, the doctor's emotional state often influences the diagnosis. Furthermore, systems that support diagnosis using generative AI models do not take into account the emotional state of the doctor or patient, limiting the accuracy and adaptability of the diagnosis. Conventional technologies do not incorporate emotional data when collecting feedback or retraining AI models, resulting in inflexible behavior adjustments and making it difficult to contribute to improving the quality of medical care.

[0889] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for recognizing the emotional state of the user and using it for analysis, a means for adjusting the diagnosis presentation method based on the emotional state of the medical professional, and a means for collecting the final diagnosis result and feedback and using them as training data for the generative AI model. This enables diagnostic support that takes the emotional state into consideration, improving the accuracy of diagnosis and the quality of medical care.

[0890] 1. A "generative artificial intelligence model" is a system that uses machine learning algorithms to generate specific outputs from input data.

[0891] 2. "Patient-reported symptoms" refers to the patient's health condition or physical abnormalities reported to a medical professional.

[0892] 3. "Medical Image Data" means image data used for medical diagnosis or treatment, such as CT scans, MRIs, and X-ray images.

[0893] 4. "Test results" means data based on medical tests such as blood tests, urine tests, and tissue tests.

[0894] 5. "Diagnosis candidate" refers to a diagnosis result that the generative artificial intelligence model may present based on the analysis results.

[0895] 6. "Analysis results" refers to the conclusions or insights obtained by analyzing the input data.

[0896] 7. "Healthcare professionals" are doctors, nurses, clinical laboratory technicians, and other professionals who diagnose, treat, and care for patients.

[0897] 8. "User emotional state" means the mental or psychological state of a healthcare professional or patient.

[0898] 9. "Adjusting the diagnosis presentation method based on emotional state" refers to a method of adaptively changing the display and presentation of diagnosis results depending on the user's emotional state.

[0899] 10. "Final diagnosis" means the patient's final diagnosis determined by a medical professional.

[0900] 11. "Feedback" refers to information such as evaluations and opinions of the system and medical professionals' reactions to diagnostic results.

[0901] 12. “Training Data” means a data set used to train a machine learning model.

[0902] MODE FOR CARRYING OUT THE INVENTION

[0903] The present invention relates to a medical diagnosis support system, and in particular to a system that uses a generative artificial intelligence model and an emotion engine to analyze a patient's symptoms, medical image data, and test results, and presents diagnostic candidates while also taking into account the user's emotional state. The following describes in detail an embodiment of the present invention.

[0904] System configuration

[0905] The system of the present invention includes the following major components:

[0906] 1. Terminal: A device where the user (doctor) inputs patient information and receives analysis results and emotions. Specifically, input devices such as a PC or tablet can be used.

[0907] 2. Server: This is the central system that collects and analyzes data, uses generative AI models to present diagnostic candidates, and analyzes user emotions. This can be implemented on a cloud server or an on-premise server.

[0908] 3. Database: A storage system for storing patient data, analysis results, feedback, and emotion data. For example, an SQL database or a NoSQL database can be used.

[0909] 4. Emotion Engine: An algorithm for recognizing and analyzing the emotional state of users (doctors) and patients. This engine analyzes emotions using natural language processing and image recognition technology.

[0910] Example of a system

[0911] The present invention will be described in detail below based on a specific example of the operation of the system.

[0912] Entering patient information

[0913] The user (doctor) uses the terminal to input basic patient information, symptoms, medical image data, and test results. For example, for a patient complaining of abdominal pain, basic information, CT scan images, and blood test results are input. The emotion engine also detects the user's (doctor's) emotional state in real time and acquires the data.

[0914] Sending data

[0915] The device sends the collected patient information and emotion data to the server using an HTTP request.

[0916] Data storage and preprocessing

[0917] The server stores the received data in a database, after which pre-processing of the data is performed, which includes data integrity checks, cleaning, and format conversion.

[0918] AI model for generating diagnostic candidates

[0919] The preprocessed data is input into a generative AI model on the server, which generates candidate diagnoses based on the patient's symptoms. At the same time, emotional data obtained from the emotion engine is also included in the analysis. For example, the generative AI model references past case data to generate candidate diagnoses such as "acute appendicitis," "intestinal obstruction," and "gastritis" for a patient complaining of abdominal pain.

[0920] Adjusting the display and presentation of results

[0921] The generated diagnosis candidates are sent from the server to the terminal. The terminal displays the diagnosis candidates in a format that is easy for the user (doctor) to understand. The emotion engine also adaptively adjusts the display method of the diagnosis candidates based on the emotion data acquired. For example, for a user with high stress, the diagnosis candidates are displayed in different colors to make them easier to understand visually.

[0922] Enter and submit feedback

[0923] The user (doctor) inputs the final diagnosis and feedback using a terminal and sends it to the server. This feedback data is used to retrain the generative AI model.

[0924] Prompt Sentence Examples

[0925] "Please list possible diagnoses based on the CT scan images and blood test results of a patient who visits the hospital with abdominal pain. Please suggest diagnoses based on medical evidence. Also, please adjust the display method according to the user's (doctor's) emotional state."

[0926] This invention allows doctors to quickly and accurately obtain supplementary information, reducing the risk of misdiagnosis. Furthermore, the use of an emotion engine enables diagnostic support that takes the user's mental state into account, which is expected to improve diagnostic accuracy and the quality of medical care. It is also expected that the accuracy of the system will continue to improve through collected feedback.

[0927] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0928] Step 1: Enter patient information

[0929] The terminal receives input from the user (doctor) of the patient's basic information, symptoms (e.g., abdominal pain), medical image data, and test results. The input includes the patient's basic information, CT scan images, blood test results, etc. Based on this, a basic information dataset is generated, and the emotion engine detects the user's (doctor's) emotional state (e.g., stress, fatigue) in real time and acquires it as additional data.

[0930] Specific behavior:

[0931] The user (doctor) enters the patient's name, age, gender, symptoms, CT scan images, and blood test results into the device's input form and presses the send button. The emotion engine analyzes the user's facial expressions and tone of voice and records their emotional state.

[0932] Step 2: Sending data

[0933] The terminal sends the collected patient information and emotion data to the server. As input, the reference dataset is sent by the terminal to the server. The data is sent using the HTTP request format.

[0934] Specific behavior:

[0935] The device collects all patient information and emotion data, structures it in JSON format, and sends it to the server using an HTTP POST request.

[0936] Step 3: Storing and Preprocessing Data

[0937] The server stores the received data in a database. The input includes the reference dataset sent from the terminal. Then, the data is preprocessed. The preprocessing includes data integrity check, cleaning, and format conversion. The output is a preprocessed clean dataset.

[0938] Specific behavior:

[0939] The server structures the incoming data and stores it in the database using SQL insert queries. After storing it, it detects missing values ​​in the data, imputes them appropriately, and then converts all data into a unified format.

[0940] Step 4: Generate candidate diagnoses using the AI ​​model

[0941] The server inputs the preprocessed data into a generative AI model to generate candidate diagnoses. The input includes a clean, preprocessed dataset. During the analysis, the AI ​​model references past case data and medical knowledge. The output is multiple candidate diagnoses.

[0942] Specific behavior:

[0943] The server runs a generative AI model using a deep learning framework such as TensorFlow or PyTorch, inputs the preprocessed data into the model, and the model analyzes the data to generate candidate diagnoses such as "acute appendicitis," "intestinal obstruction," and "gastritis."

[0944] Step 5: Adjust the display and presentation of results

[0945] The server organizes the generated candidate diagnoses and sends them to the terminal. The input includes the generated candidate diagnoses. The terminal displays the candidate diagnoses and adjusts the display method based on the emotional state of the user (doctor). The output is the adjusted display of the candidate diagnoses.

[0946] Specific behavior:

[0947] The server sends the generated diagnosis candidates in JSON format to the terminal as an HTTP response. The terminal receives this and displays the diagnosis candidates through a GUI. Based on the results of the emotion engine analysis, if the doctor is in a state of high stress, the diagnosis candidates are color-coded to make them easier to understand visually.

[0948] Step 6: Enter and submit your feedback

[0949] The terminal allows the user (doctor) to input the final diagnosis and feedback, and sends the data to the server. The input includes the final diagnosis and feedback. The output is the feedback data sent to the server.

[0950] Specific behavior:

[0951] The user (doctor) enters the final diagnosis (e.g., acute appendicitis) and feedback about the system's diagnostic accuracy into the input form on the terminal, and presses the submit button. The terminal structures this data in JSON format and sends it to the server using an HTTP POST request.

[0952] Step 7: Save feedback and retrain

[0953] The server stores the received feedback data in a database and uses it to retrain the generative AI model. The feedback data is included as input. The output is an updated generative AI model.

[0954] Specific behavior:

[0955] The server adds and stores the feedback data in a database, then retrains the generative AI model using a new dataset that includes the feedback data, e.g., updating model parameters to improve the model's performance.

[0956] (Application example 2)

[0957] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0958] The problem that this invention aims to solve is to improve the accuracy and efficiency of diagnoses by reducing the risk of misdiagnosis caused by information overload, time constraints, and the doctor's own emotional state when making a diagnosis in the medical field. Another problem is to improve the safety and efficiency of work by analyzing the emotional state of security operators in real time and utilizing that information when generating alerts.

[0959] 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.

[0960] In this invention, the server includes means for analyzing patient-reported symptoms, medical image data, and test results using a generative AI model and presenting diagnosis candidates, means for displaying the analysis results to medical professionals, means for collecting final diagnosis results and feedback from medical professionals and using them as training data for the generative AI model, means for collecting and analyzing biometric data in real time, and means for generating and displaying alerts based on the analysis results and biometric data, thereby enabling improved accuracy and efficiency of diagnoses in medical settings and improved safety and efficiency in security operations.

[0961] A "generative artificial intelligence model" refers to an algorithm or machine learning model that analyzes data collected in medical diagnosis or security analysis and generates diagnostic candidates or anomaly detection results.

[0962] "Patient-reported symptoms" refers to information about specific symptoms such as poor health or pain that patients self-report in medical settings.

[0963] "Medical image data" refers to image information of a patient's inside the body obtained using medical equipment such as CT scans, X-ray images, and MRIs.

[0964] "Test results" refers to numerical data and information indicating whether or not there are abnormalities obtained from blood tests, urine tests, various biochemical tests, etc.

[0965] "Analysis results" refers collectively to the diagnosis candidates and anomaly detection results obtained after analyzing input data using a generative artificial intelligence model.

[0966] "Medical professionals" refers to professionals who perform medical-related tasks, such as doctors, nurses, and radiological technologists.

[0967] "Biometric data" refers to information related to the physical and mental state of medical professionals, security operators, etc., such as heart rate, body temperature, and facial expression data.

[0968] An "alert" refers to a warning or caution message issued by the system based on analysis results or abnormal values ​​in biometric data.

[0969] "Database" refers to a system that organizes and stores large amounts of information, such as reported symptoms, medical image data, test results, and analysis results, and enables quick search and retrieval as needed.

[0970] "Preprocessing" refers to the process of checking the integrity of collected data, cleaning it, and converting its format to make it suitable for subsequent analysis.

[0971] "Feedback" refers to the evaluations and opinions of medical professionals and security operators regarding the final diagnosis results and system proposals.

[0972] The embodiments for carrying out the present invention will be described below.

[0973] System Overview

[0974] The system of the present invention includes the following major components:

[0975] 1. Terminal: A device used by the user (doctor or security operator) to input and display basic information, biometric data, analysis results, and emotional state of the patient.

[0976] 2. Server: A central system that collects and analyzes data, generates diagnostic candidates and anomaly detection results using generative AI models, integrates the analysis results with biometric data, and generates appropriate alerts.

[0977] 3. Database: A storage system for storing reported symptoms, medical image data, test results, analysis results, feedback, emotional data, etc.

[0978] 4. Emotion engine: An algorithm that recognizes and analyzes the emotional state of the user (doctor or security operator).

[0979] Data collection

[0980] The terminal collects basic information about the patient and operator, as well as biometric data (heart rate, body temperature, facial expression data) from the user, and also collects video data and medical image data from security cameras and medical equipment.

[0981] Data analysis

[0982] The server stores the received data in a database and performs preprocessing. This involves checking the data for consistency, cleaning it, and converting its format. Once this is complete, the data is input into the generative AI model for analysis. At the same time, the emotion data obtained by the emotion engine is used for analysis.

[0983] Alert Generation

[0984] Based on the analysis results and emotion data, the server generates appropriate alerts. For example, if an abnormality is detected and an operator is in a high stress state at the same time, an alert is displayed indicating that immediate action is required. The alert is sent to the terminal and displayed to the user.

[0985] Feedback collection

[0986] Feedback on analysis results and alerts is collected from users. This feedback is sent to the server and stored in a database. The collected feedback is used as retraining data for the generative AI model and is used to improve the accuracy of the system.

[0987] Specific examples

[0988] When a doctor uses a device to enter basic patient information, CT scan images, and blood test results, the data is sent to a server. The server preprocesses the data and analyzes it using a generative AI model. As a result, a candidate diagnosis is generated and displayed on the device. At the same time, an emotion engine analyzes the doctor's emotional state and adjusts the way the candidate diagnosis is presented based on that data.

[0989] If a security operator wears smart glasses during the night shift and the system simultaneously detects an abnormality and high stress, an alert message will be displayed stating, "High stress and abnormality detected! Immediate action required!"

[0990] Example prompts to input to the generative AI model

[0991] "If an anomaly is detected, how can we tailor our response to reduce operator burden?"

[0992] "A method for detecting operator stress levels in real time and proposing countermeasures during security operations."

[0993] In this way, the present invention enables efficient information presentation and alert generation that integrates analysis results and emotional data in medical settings and security operations.

[0994] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0995] Step 1:

[0996] Data Collection Phase

[0997] Input: Basic information from the user (doctor or security operator), biometric data (heart rate, body temperature, facial expression data), video data and medical image data from security cameras and medical equipment.

[0998] Output: Sends collected data to a server.

[0999] Specific behavior:

[1000] The user uses the terminal to input basic patient information and biometric data, and data is collected in real time from security cameras and medical devices, which are then transmitted to a server.

[1001] Step 2:

[1002] Database save phase

[1003] Input: Basic information, biometric data, video data, and medical image data sent to the server.

[1004] Output: Various data stored in the database.

[1005] Specific behavior:

[1006] The server stores the received data in a database, where it checks for consistency.

[1007] Step 3:

[1008] Preprocessing Phase

[1009] Input: Basic information, biometric data, video data, and medical image data stored in the database.

[1010] Output: The preprocessed data.

[1011] Specific behavior:

[1012] The server retrieves data from the database, checks its integrity, cleans it, and converts it into a format suitable for generative AI models.

[1013] Step 4:

[1014] Data analysis phase

[1015] Input: Preprocessed basic information, biometric data, video data, and medical image data.

[1016] Output: Diagnosis candidates, anomaly detection results, and sentiment analysis results from the generative AI model.

[1017] Specific behavior:

[1018] The server inputs the preprocessed data into the generative AI model for analysis, while simultaneously analyzing and integrating the user's emotional data obtained by the emotion engine.

[1019] Step 5:

[1020] Alert Generation Phase

[1021] Input: Diagnosis candidates, anomaly detection results, and sentiment analysis results.

[1022] Output: The alert message.

[1023] Specific behavior:

[1024] The server generates appropriate alerts based on the analysis results and biometric data. For example, if an abnormality is detected and an operator is in a high stress state, an alert message requiring immediate action is generated. The generated alert is then sent to the terminal.

[1025] Step 6:

[1026] Display Phase

[1027] Input: Alert message, diagnosis candidates, and sentiment analysis results sent from the server.

[1028] Output: Alert message and potential diagnosis displayed on the terminal.

[1029] Specific behavior:

[1030] The device receives alert messages and diagnostic candidates from the server and displays them to the user, adjusting the display format according to the results of emotion analysis.

[1031] Step 7:

[1032] Feedback gathering phase

[1033] Input: Final diagnostic results and feedback from the user.

[1034] Output: Feedback sent to the server.

[1035] Specific behavior:

[1036] The user (doctor or security operator) inputs feedback on the final diagnosis and the evaluation of the alert through the terminal, and this data is sent to the server and stored in the database.

[1037] Step 8:

[1038] Model retraining phase

[1039] Input: Stored feedback, newly collected data.

[1040] Output: A generative artificial intelligence model with improved accuracy.

[1041] Specific behavior:

[1042] The server retrains the generative AI model based on collected feedback and newly acquired data, thereby continuously improving diagnostic accuracy and anomaly detection performance.

[1043] This concludes the explanation of each processing step of this program and its specific operations.

[1044] 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.

[1045] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1046] 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.

[1047] [Third embodiment]

[1048] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1049] 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.

[1050] 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).

[1051] 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.

[1052] 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.

[1053] 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).

[1054] 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.

[1055] 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.

[1056] 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.

[1057] 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.

[1058] 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.

[1059] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1060] The embodiments for carrying out the present invention will be described below.

[1061] System Overview

[1062] The present invention relates to a medical diagnosis support system, in particular a system that uses a generative artificial intelligence model to analyze a patient's symptoms, medical image data, and test results, and presents potential diagnoses. The system includes the following main components:

[1063] 1. Terminal: A device where the user (doctor) enters patient information and receives analysis results.

[1064] 2. Server: The central system that collects and analyzes data and uses generative AI models to present potential diagnoses.

[1065] 3. Database: A storage system for storing patient data, analysis results, and feedback.

[1066] Explaining program processing in natural language

[1067] Data Collection Phase

[1068] The terminal receives patient information from the user (doctor) and transmits it to the server. Specifically, this information includes the patient's basic information, reported symptoms, medical image data, and test results.

[1069] Example: A user (doctor) uses a terminal to input basic information about a patient complaining of abdominal pain, CT scan images, and blood test results, and sends them to a server.

[1070] Data analysis phase

[1071] The server stores the received data in a database and performs preprocessing, which includes checking data consistency, cleaning, and format conversion. Once preprocessing is complete, the data is input into a generative AI model for analysis.

[1072] Example: The server preprocesses data received about a patient's abdominal pain and inputs it into a generative AI model. The generative AI model analyzes the data based on past cases of abdominal pain and medical knowledge to generate candidate diagnoses.

[1073] Diagnosis candidate presentation phase

[1074] The server organizes the diagnostic candidates obtained from the generative AI model and sends them to the device used by the user (doctor). The device displays the diagnostic candidates in a format that is easy for the user to understand.

[1075] Example: The server generates diagnostic candidates such as "acute appendicitis" and "intestinal obstruction" and sends them to the terminal so that the user (doctor) can check them.

[1076] Feedback gathering phase

[1077] The device allows the user (doctor) to input the final diagnosis and feedback. This data is sent to the server and stored in a database. This feedback is then used to retrain the generative AI model, contributing to improving the accuracy of the system.

[1078] For example, a user (doctor) diagnoses "acute appendicitis" and provides feedback that the system's diagnosis was appropriate. This information is sent to the server and used as training data for the generative AI model.

[1079] Specific operation example

[1080] When a patient visits the hospital with abdominal pain, the user (doctor) uses a terminal to input the patient's basic information, CT scan images, blood test results, and other test results. This data is sent to the server and stored in a database. The server preprocesses the data and inputs it into a generative AI model for analysis. The generated diagnosis candidates are sent to the terminal and displayed to the user (doctor). After the final diagnosis is made, feedback is sent from the terminal to the server, stored in a database, and used again as training data for the generative AI model.

[1081] The system provides doctors with fast and accurate supplementary information, reducing the risk of misdiagnosis, and continuously improves its accuracy through collected feedback.

[1082] The processing flow will be explained below.

[1083] Specific explanation of the program's processing steps

[1084] Data Collection Phase

[1085] Step 1:

[1086] The terminal receives basic patient information from the user (doctor), including the patient's name, age, sex, and other basic information.

[1087] pseudo

[1088] The terminal receives basic patient information from the user.

[1089] Step 2:

[1090] The terminal receives input from the user (doctor) of the patient's reported symptoms, including the specific symptoms and progress of the patient's awareness.

[1091] pseudo

[1092] The terminal inputs the patient's reported symptoms.

[1093] Step 3:

[1094] The terminal receives input of medical image data (e.g., X-rays, CT scans) and test results (e.g., blood test results, urine test results) from the user (doctor).

[1095] pseudo

[1096] The terminal inputs medical image data and test results

[1097] Step 4:

[1098] The device sends all collected data to the server.

[1099] pseudo

[1100] The device sends the input data to the server

[1101] Data analysis phase

[1102] Step 5:

[1103] The server receives the patient data transmitted from the terminal.

[1104] pseudo

[1105] The server receives data from the device

[1106] Step 6:

[1107] The server stores the received patient data in a database.

[1108] pseudo

[1109] The server stores the data in a database

[1110] Step 7:

[1111] The server checks the integrity of the stored data and performs pre-processing (data cleaning, format conversion, etc.).

[1112] pseudo

[1113] The server performs data integrity checks and preprocessing

[1114] Step 8:

[1115] The server inputs the preprocessed data into the generative AI model.

[1116] pseudo

[1117] The server inputs the preprocessed data into the generative AI model

[1118] Step 9:

[1119] The server obtains the candidate diagnosis results from the generative AI model.

[1120] pseudo

[1121] The server receives the analysis results from the generated AI model

[1122] Diagnosis candidate presentation phase

[1123] Step 10:

[1124] The server formats the acquired diagnosis candidates into a format that is easy for the user (doctor) to understand.

[1125] pseudo

[1126] The server formats the candidate diagnoses

[1127] Step 11:

[1128] The server sends the formatted diagnosis candidates to the terminal and presents them to the user (doctor).

[1129] pseudo

[1130] The server sends the diagnosis candidate to the terminal.

[1131] The device displays possible diagnoses to the user (doctor)

[1132] Feedback gathering phase

[1133] Step 12:

[1134] The terminal allows the user (doctor) to input the final diagnosis result.

[1135] pseudo

[1136] The terminal receives the final diagnosis result from the user.

[1137] Step 13:

[1138] The terminal allows the user (doctor) to input feedback on the system's diagnostic accuracy and areas for improvement.

[1139] pseudo

[1140] The device receives feedback from the user

[1141] Step 14:

[1142] The terminal sends the final diagnosis result and feedback to the server.

[1143] pseudo

[1144] The device sends the final diagnosis results and feedback to the server

[1145] Step 15:

[1146] The server stores the final diagnosis results and feedback in a database.

[1147] pseudo

[1148] The server stores the final diagnosis results and feedback in a database

[1149] Step 16:

[1150] The server uses the collected feedback to retrain the generative AI model.

[1151] pseudo

[1152] The server uses the feedback to retrain the generative AI model

[1153] Through this series of processing flows, the system of the present invention can assist users (doctors) in making diagnoses, improving diagnostic accuracy while reducing the risk of misdiagnosis. Furthermore, through continuous improvements based on feedback, the reliability and effectiveness of the system can be increased.

[1154] Example 1

[1155] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1156] Conventional medical diagnosis support systems have difficulty efficiently analyzing patients' reported symptoms, medical image data, and test results, and presenting appropriate diagnostic candidates. Furthermore, there is a lack of a method for effectively aggregating feedback from medical professionals during the process of making a final diagnosis and improving the accuracy of the diagnosis support system. This increases the risk of misdiagnosis and impacts the quality of medical care.

[1157] 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.

[1158] In this invention, the server includes a means for transmitting patient information input from a terminal to the server, a means for the server to store the received data in a database and perform data preprocessing, a means for analyzing the patient's reported symptoms, medical image data, and test results using a generative AI model and presenting candidate diagnoses, a means for displaying the analysis results to medical professionals, and a means for collecting final diagnoses and feedback from medical professionals and using them as training data for the generative AI model. This enables efficient collection and preprocessing of patient data, presentation of appropriate candidate diagnoses, collection of feedback, and improvement of system accuracy.

[1159] A "generative artificial intelligence model" is a model generated using artificial intelligence technology that analyzes a patient's symptoms and test results based on medical data and generates potential diagnoses.

[1160] "Reported symptoms" are physical discomfort or abnormalities that patients report to medical professionals, and are information that forms the basis of diagnosis.

[1161] "Medical image data" refers to image data obtained using medical equipment, such as CT scans, MRIs, and X-rays, that provide the visual information necessary for diagnosis.

[1162] "Test results" refer to the numerical data and analytical results obtained through various medical tests, such as blood tests and urine tests, and are important information that assists in diagnosis.

[1163] "Candidate diagnoses" are a list of possible diagnoses presented by the generative AI model as a result of its analysis, based on the patient's symptoms and test results, and serve as a reference for medical professionals to make a final decision.

[1164] A "terminal" is a device used by a user (doctor) to enter patient information and receive analysis results, and includes PCs, tablets, smartphones, etc.

[1165] The "server" is a central system that receives, stores, and preprocesses patient information, analyzes it using a generative AI model, and presents diagnostic candidates, and also connects to the database.

[1166] A "database" is a storage system for systematically storing patient data, analysis results, and feedback from medical professionals, and is used for subsequent data reference and analysis.

[1167] "Preprocessing" refers to the process of preparing the data to be analyzed, and includes checking data consistency, cleaning, format conversion, etc.

[1168] "Feedback" refers to opinions and evaluations provided by medical professionals based on final diagnosis results, and is data used to retrain generative AI models.

[1169] System Overview

[1170] This invention relates to a medical diagnosis support system that uses a generative AI model to analyze patient-reported symptoms, medical image data, and test results, and presents potential diagnoses. The system includes the following main components:

[1171] 1. Terminal: A device where the user (doctor) enters patient information and receives analysis results. This includes PCs, tablets, smartphones, etc.

[1172] 2. Server: This is the central system that collects and analyzes data, uses generative AI models to present diagnostic candidates, and connects to the database.

[1173] 3. Database: A storage system for storing patient data, analysis results, and feedback.

[1174] Explaining program processing in natural language

[1175] Data Collection Phase

[1176] A user (doctor) uses a terminal to input patient information and sends it to a server. Specifically, this includes the patient's basic information (name, age, gender), reported symptoms, medical image data (CT scans and MRIs), and various test results (blood tests, urine tests, etc.). For example, the user (doctor) inputs the information of a 30-year-old male patient complaining of abdominal pain into the terminal and sends it to the server.

[1177] Data analysis phase

[1178] The server stores the received data in a database and then performs preprocessing on the data. Preprocessing includes checking data integrity, cleaning, and any necessary format conversions. Once preprocessed, the data is input into a generative AI model. The server preprocesses the patient information and inputs it into a generative AI model built using TensorFlow and PyTorch for analysis. The generative AI model performs analysis based on past case data and medical knowledge to generate candidate diagnoses.

[1179] Diagnosis candidate presentation phase

[1180] The server organizes the generated candidate diagnoses and sends them to the terminal. The terminal displays the candidate diagnoses in a format that is easy for the user (doctor) to understand. For example, the terminal displays "acute appendicitis" and "intestinal obstruction" as part of the list of candidate diagnoses generated by the server.

[1181] Feedback gathering phase

[1182] The user (doctor) enters the final diagnosis and feedback into the device and sends it to the server. The feedback is then stored in a database and used to retrain the generative AI model. This feedback continuously improves the accuracy of the system. For example, if the user (doctor) diagnoses "acute appendicitis" and enters feedback that the diagnosis was appropriate, this is sent to the server and used as training data for the generative AI model.

[1183] Prompt Sentence Examples

[1184] "Generate possible diagnoses based on the CT scan images and blood test results of a 30-year-old man complaining of abdominal pain."

[1185] This system will enable doctors to obtain supplementary information quickly and accurately, reducing the risk of misdiagnosis. Furthermore, it is expected that the accuracy of the generative AI model will be improved through collected feedback.

[1186] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1187] Step 1:

[1188] The terminal accepts input from the user (doctor).

[1189] Specific operation: The user (doctor) uses the terminal to enter the patient's basic information (name, age, gender), reported symptoms, medical image data (CT scan or MRI), and various test results (blood test, urine test, etc.).

[1190] Input: Patient's basic information, reported symptoms, medical image data, and test results.

[1191] Output: The entered patient information.

[1192] Step 2:

[1193] The terminal transmits the input information to the server.

[1194] Specific operation: The terminal encrypts the entered patient data and sends it to the server using the TLS protocol.

[1195] Input: Patient information entered into the terminal.

[1196] Output: Patient information sent to the server.

[1197] Step 3:

[1198] The server stores the received data in a database.

[1199] Specific operation: The server adds the received patient data to the "Patient Data" table in the database as a new record.

[1200] Input: Patient information received by the server.

[1201] Output: Patient information stored in the database.

[1202] Step 4:

[1203] The server pre-processes the data.

[1204] Specific operation: The server checks the integrity of the received data, removes invalid values, fills in missing data, and converts the format.

[1205] Input: Raw data stored in a database.

[1206] Output: Consistency checked and preprocessed data.

[1207] Step 5:

[1208] The server inputs the preprocessed data into the generative AI model.

[1209] Specific operation: The server inputs the preprocessed data into a generative AI model using TensorFlow or PyTorch.

[1210] Input: Preprocessed data.

[1211] Output: The data that is input into a generative AI model.

[1212] Step 6:

[1213] A generative AI model analyzes the data and generates candidate diagnoses.

[1214] How it works: The generative AI model analyzes the pre-processed data and generates candidate diagnoses based on past case data and medical knowledge.

[1215] Input: Preprocessed data fed into a generative AI model.

[1216] Output: The generated candidate diagnoses.

[1217] Step 7:

[1218] The server transmits the generated candidate diagnoses to the terminal.

[1219] Specific operation: The server organizes the generated diagnosis candidate list and sends it to the terminal.

[1220] Input: A list of diagnostic candidates output from a generative AI model.

[1221] Output: A list of diagnostic candidates sent to the terminal.

[1222] Step 8:

[1223] The terminal displays the possible diagnoses to the user (doctor).

[1224] Specific operation: The terminal displays the list of diagnosis candidates in a format (for example, a list format) that is easy for the user (doctor) to understand.

[1225] Input: A list of diagnostic candidates sent to the terminal.

[1226] Output: A list of possible diagnoses displayed to the user (doctor).

[1227] Step 9:

[1228] The user (doctor) inputs the diagnosis and feedback.

[1229] Specific operation: The user (doctor) inputs the final diagnosis and feedback to the system into the terminal.

[1230] Input: User's (doctor's) final diagnosis and feedback.

[1231] Output: Feedback information entered into the terminal.

[1232] Step 10:

[1233] The terminal sends feedback information to the server.

[1234] Specific operation: The terminal encrypts the input feedback information and sends it to the server.

[1235] Input: Feedback information entered into the device.

[1236] Output: Feedback information sent to the server.

[1237] Step 11:

[1238] The server stores the feedback information in a database.

[1239] Specific operation: The server stores the feedback information in a "diagnostic feedback" table in the database.

[1240] Input: Feedback information received by the server.

[1241] Output: Feedback information stored in a database.

[1242] Step 12:

[1243] The server retrains the generative AI model based on the feedback.

[1244] Specific operation: The server periodically extracts feedback information from the database and retrains the generative AI model.

[1245] Input: Feedback information stored in the database.

[1246] Output: The retrained generative AI model.

[1247] (Application example 1)

[1248] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1249] In modern medical settings, doctors are required to make quick and accurate diagnoses within a limited time frame, but this is often difficult. While selecting the right medicine is also important for pharmacies and drugstores, it can be difficult to accurately understand a patient's symptoms and provide appropriate advice. Therefore, there is a need for a system that can support doctors and pharmacists in the diagnostic process.

[1250] 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.

[1251] In this invention, the server includes means for analyzing patient-reported symptoms, medical image data, and test results using a generative AI model and presenting candidate diagnoses, means for displaying the analysis results to users such as medical professionals or patients, means for collecting final diagnosis results and feedback from users and using them as training data for the generative AI model, and means for using the medical diagnosis support system in physical stores such as drugstores. This allows doctors and pharmacists to quickly and accurately present candidate diagnoses, enabling them to select appropriate medicines and medical procedures.

[1252] A "generative artificial intelligence model" is an artificial intelligence algorithm used to analyze data such as patient symptoms, medical images, and test results to generate potential diagnoses.

[1253] "Reported symptoms" refers to any ill health or specific symptoms reported by the patient.

[1254] "Medical image data" refers to image data taken using medical equipment such as CT scans and MRIs.

[1255] "Test results" refers to data obtained from various medical tests, such as blood tests and urine tests.

[1256] "Analysis results" refer to the diagnostic candidates and related information obtained using the generative artificial intelligence model.

[1257] "Healthcare professionals" refer to professionals involved in diagnosing and treating patients, such as doctors and nurses.

[1258] "Feedback" refers to the final diagnostic results and evaluations and opinions from users.

[1259] "Training Data" means data used to improve and optimize a generative artificial intelligence model.

[1260] A "medical diagnosis support system" is a system that uses a generative artificial intelligence model to present potential diagnoses to medical professionals and patients, helping them make a final diagnosis.

[1261] The embodiments for carrying out the present invention will be described below.

[1262] System Overview

[1263] The present invention relates to a medical diagnosis support system, in particular a system that uses a generative artificial intelligence model to analyze a patient's symptoms, medical image data, and test results, and presents potential diagnoses. The system includes the following main components:

[1264] 1. Terminal: A device where users (doctors and patients) enter symptoms and test results and receive analysis results. Specifically, a smartphone, tablet, or interactive display is used.

[1265] 2. Server: This is the central system that collects, analyzes, and stores data, and uses generative AI models to present diagnostic candidates. A cloud server (e.g., Amazon Web Services or Microsoft Azure) is used.

[1266] 3. Database: A storage system for storing patient data, analysis results, and feedback. SQL / NoSQL databases (MySQL, MongoDB, PostgreSQL) are used.

[1267] Program Processing and Technical Details

[1268] Data Collection Phase

[1269] Users (doctors or patients) use a terminal to input their symptoms, medical images, test results, etc. For example, a patient complaining of abdominal pain inputs their age, gender, CT scan images, and blood test results. This data is then sent to a server via the Internet.

[1270] Data analysis phase

[1271] The server stores the received data in a database and performs preprocessing (integrity checks, cleaning, format conversion) on the data. After preprocessing is complete, the data is input into a generative artificial intelligence model for analysis. This process uses machine learning libraries such as TensorFlow and PyTorch.

[1272] Diagnosis candidate presentation phase

[1273] The candidate diagnoses obtained from the analysis results are organized on the server and sent to the terminal, which then displays them in a format that is easy for the user to understand. For example, candidate diagnoses such as "acute appendicitis" and "intestinal obstruction" are displayed.

[1274] Feedback gathering phase

[1275] The final diagnosis and feedback are entered on the device and sent to the server, where they are stored in a database and used to retrain the generative AI model.

[1276] Specific operation example

[1277] When a patient visits a hospital or drugstore with abdominal pain, the user (doctor or pharmacist) uses a terminal to enter the patient's basic information, CT scan images, blood test results, and other test results. For example, the user might enter, "I have abdominal pain and have had similar symptoms several times in the past. I am 45 years old. I also have CT scan images. Please let me know if you have any diagnostic candidates." This data is sent to a server and stored in a database. The server preprocesses the data and inputs it into a generative AI model for analysis. The generated diagnostic candidates are sent to the terminal and displayed to the user (doctor or patient). After the final diagnosis is made, feedback is sent from the terminal to the server, stored in a database, and used as training data for the generative AI model.

[1278] The system allows users to quickly and accurately identify potential diagnoses, reducing the risk of misdiagnosis, and continuously improves its accuracy through collected feedback.

[1279] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1280] Step 1:

[1281] The terminal receives input from the user (doctor or patient) about symptoms, basic information, medical image data, and test results. The input data includes age, gender, and details of symptoms. This input data is sent to the server via the Internet. An example of a prompt for the user to input specific information is, "I have abdominal pain and have had similar symptoms several times in the past. I am 45 years old. I also have a CT scan image. Please let me know if you have any possible diagnoses."

[1282] Step 2:

[1283] The server receives data sent from the terminal and stores it in a database. The stored data may not be suitable for analysis as is, so it must be preprocessed. Specific preprocessing steps include checking data consistency, filling in missing data, and standardizing formats. This process ensures the availability and reliability of input data.

[1284] Step 3:

[1285] The server inputs the preprocessed data into the artificial intelligence model. Specifically, machine learning libraries such as TensorFlow and PyTorch are used to input the data into the model. The model then performs analysis based on the input data, using past case data and a medical knowledge base. As a result of the analysis, candidate diagnoses are generated.

[1286] Step 4:

[1287] The server organizes the generated candidate diagnoses and sends them to the terminal. The terminal displays the analysis results in an easy-to-read format for the user (doctor or patient). At this stage, an appropriate user interface design is important to present the analysis results in an easy-to-understand format. For example, specific candidate diagnoses such as "acute appendicitis" or "intestinal obstruction" are displayed.

[1288] Step 5:

[1289] The user uses the terminal to input the final diagnosis results and feedback. The user also inputs feedback on the diagnosis performed based on the confirmed diagnosis candidates and their results into the terminal. This includes an evaluation of the accuracy of the diagnosis and the information presented. This feedback data is sent to the server.

[1290] Step 6:

[1291] The server stores the final diagnosis results and feedback sent by the user in a database and uses them as training data for the generative AI model. By retraining the generative AI model based on this feedback data, the accuracy of the model is continuously improved, further increasing the reliability and usefulness of the system.

[1292] Through the above processing steps, this medical diagnosis support system is able to efficiently and accurately analyze medical data and present potential diagnoses.

[1293] 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.

[1294] The embodiments for carrying out the present invention will be described below.

[1295] System Overview

[1296] The present invention relates to a medical diagnosis support system, in particular a system that uses a generative artificial intelligence model and an emotion engine to analyze patient symptoms, medical image data, and test results, and presents potential diagnoses while also taking into account the user's emotional state. The system includes the following main components:

[1297] 1. Terminal: A device where the user (doctor) inputs patient information and receives analysis results and emotions.

[1298] 2. Server: The central system that collects and analyzes data, uses generative AI models to present diagnostic candidates, and analyzes user sentiment.

[1299] 3. Database: A storage system for storing patient data, analysis results, feedback and emotion data.

[1300] 4. Emotion engine: Algorithms for recognizing and analyzing the emotional state of the user (doctor) and patient.

[1301] Explaining program processing in natural language

[1302] Data Collection Phase

[1303] The terminal receives basic patient information from the user (doctor) and sends it to the server. This information includes the patient's basic information, reported symptoms, medical image data, and test results. The emotion engine also recognizes the user's (doctor's) emotional state and captures it as data.

[1304] Example: A user (doctor) uses a terminal to input basic information about a patient complaining of abdominal pain, CT scan images, and blood test results, and at the same time, the emotion engine detects the user's (doctor's) stress level and transmits the data.

[1305] Data analysis phase

[1306] The server stores the received data in a database and performs preprocessing. Preprocessing includes checking data consistency, cleaning, and format conversion. Once preprocessed, the data is input into a generative AI model for analysis. Furthermore, emotional data of users and patients obtained by the emotion engine is also used in the analysis.

[1307] Example: The server preprocesses data received about a patient's abdominal pain and inputs it into a generative AI model. The generative AI model analyzes the data based on past cases of abdominal pain and medical knowledge to generate a candidate diagnosis. At the same time, it evaluates the possibility that the doctor's emotional state may affect the diagnosis based on the results of the emotion engine analysis.

[1308] Diagnosis candidate presentation phase

[1309] The server organizes the acquired diagnosis candidates and sends them to the device used by the user (doctor). The device displays the diagnosis candidates and emotion analysis results in a format that is easy for the user (doctor) to understand. The presentation method of the diagnosis candidates is also adaptively changed depending on the emotion analysis results.

[1310] For example, the server generates diagnostic candidates such as "acute appendicitis" or "intestinal obstruction" and sends them to the terminal for the user (doctor) to review. At the same time, the emotion engine adjusts the way the diagnostic candidates are displayed based on the emotional state recognized by the server.

[1311] Feedback gathering phase

[1312] The device allows the user (doctor) to input the final diagnosis, feedback, and emotional data. This data is sent to the server and stored in a database. This feedback is then used to retrain the generative AI model, contributing to improving the accuracy of the system.

[1313] Example: A user (doctor) diagnoses "acute appendicitis" and provides feedback that the system's diagnosis was appropriate. This information is sent to the server and used as training data for the generative AI model. The emotional state collected by the emotion engine is also stored.

[1314] Specific operation example

[1315] When a patient visits the hospital with abdominal pain, the user (doctor) uses a terminal to input the patient's basic information, CT scan images, blood test results, and other test results. This data is sent to the server and stored in a database. The server preprocesses the data and inputs it into a generative AI model for analysis. The generated diagnosis candidates are sent to the terminal and displayed to the user (doctor). After the final diagnosis is made, feedback is sent from the terminal to the server, stored in a database, and used again as training data for the generative AI model. At the same time, an emotion engine detects the user's emotional state and uses this data for analysis.

[1316] This system allows doctors to obtain supplementary information quickly and accurately, reducing the risk of misdiagnosis. Furthermore, the use of an emotion engine enables diagnostic support that takes the user's mental state into account, which is expected to improve diagnostic accuracy and the quality of medical care. The accuracy of the system is continuously improved through collected feedback.

[1317] The processing flow will be explained below.

[1318] Specific explanation of the program's processing steps

[1319] Data Collection Phase

[1320] Step 1:

[1321] The terminal receives basic patient information from the user (doctor), including basic information such as the patient's name, age, sex, height, and weight.

[1322] pseudo

[1323] The terminal receives basic patient information from the user.

[1324] Step 2:

[1325] The terminal receives input from the user (doctor) of the patient's reported symptoms, including the specific symptoms and progress of the symptoms that the patient is aware of.

[1326] pseudo

[1327] The terminal inputs the patient's reported symptoms.

[1328] Step 3:

[1329] The terminal receives input of medical image data (e.g., X-rays, CT scans) and test results (e.g., blood test results, urine test results) from the user (doctor).

[1330] pseudo

[1331] The terminal inputs medical image data and test results

[1332] Step 4:

[1333] The emotion engine recognizes the user's (doctor's) emotional state and captures it as data, including the doctor's stress level and fatigue level.

[1334] pseudo

[1335] Emotion engine recognizes the doctor's emotional state

[1336] Step 5:

[1337] The device sends all collected data to the server.

[1338] pseudo

[1339] The device sends the input data to the server

[1340] Data analysis phase

[1341] Step 6:

[1342] The server receives the patient data and emotion data transmitted from the terminal.

[1343] pseudo

[1344] The server receives data from the device

[1345] Step 7:

[1346] The server stores the received patient data in a database.

[1347] pseudo

[1348] The server stores the data in a database

[1349] Step 8:

[1350] The server checks the integrity of the stored data and performs pre-processing (data cleaning, format conversion, etc.).

[1351] pseudo

[1352] The server performs data integrity checks and preprocessing

[1353] Step 9:

[1354] The server inputs the preprocessed data into a generative AI model, including emotional data.

[1355] pseudo

[1356] The server inputs the preprocessed data into the generative AI model

[1357] Step 10:

[1358] The server obtains the candidate diagnosis results from the generative AI model.

[1359] pseudo

[1360] The server receives the analysis results from the generated AI model

[1361] Diagnosis candidate presentation phase

[1362] Step 11:

[1363] The server formats the diagnosis candidates obtained into a format that is easy for the user (doctor) to understand, and adjusts the results by reflecting emotional data.

[1364] pseudo

[1365] The server formats the candidate diagnoses

[1366] The server adjusts the results by taking into account the emotional data.

[1367] Step 12:

[1368] The server sends the formatted diagnosis candidates to the terminal and presents them to the user (doctor).

[1369] pseudo

[1370] The server sends the diagnosis candidate to the terminal.

[1371] The device displays possible diagnoses to the user (doctor)

[1372] Feedback gathering phase

[1373] Step 13:

[1374] The terminal allows the user (doctor) to input the final diagnosis result.

[1375] pseudo

[1376] The terminal receives the final diagnosis result from the user.

[1377] Step 14:

[1378] The device allows the user (doctor) to input feedback on the system's diagnostic accuracy and areas for improvement, along with their emotional state.

[1379] pseudo

[1380] The device receives feedback from the user

[1381] The device also collects emotional state data

[1382] Step 15:

[1383] The terminal sends the final diagnosis result and feedback to the server.

[1384] pseudo

[1385] The device sends the final diagnosis results and feedback to the server

[1386] Step 16:

[1387] The server stores the final diagnosis results and feedback in a database.

[1388] pseudo

[1389] The server stores the final diagnosis results and feedback in a database

[1390] Step 17:

[1391] The server uses the collected feedback to retrain the generative AI model, including the emotional data.

[1392] pseudo

[1393] The server uses the feedback to retrain the generative AI model

[1394] Through this series of processing flows, the system of the present invention provides diagnostic support to the user (doctor), improving diagnostic accuracy while reducing the risk of misdiagnosis. Furthermore, by using an emotion engine, diagnostic support that takes into account the user's mental state is possible, which is expected to improve diagnostic accuracy and the quality of medical care. The accuracy of the system is continuously improved through collected feedback.

[1395] Example 2

[1396] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1397] Traditionally, medical diagnoses have relied primarily on the doctor's experience and knowledge, leading to potential misdiagnoses and oversights. Furthermore, the doctor's emotional state often influences the diagnosis. Furthermore, systems that support diagnosis using generative AI models do not take into account the emotional state of the doctor or patient, limiting the accuracy and adaptability of the diagnosis. Conventional technologies do not incorporate emotional data when collecting feedback or retraining AI models, resulting in inflexible behavior adjustments and making it difficult to contribute to improving the quality of medical care.

[1398] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for recognizing the emotional state of the user and using it for analysis, a means for adjusting the diagnosis presentation method based on the emotional state of the medical professional, and a means for collecting the final diagnosis result and feedback and using them as training data for the generative AI model. This enables diagnostic support that takes the emotional state into consideration, improving the accuracy of diagnosis and the quality of medical care.

[1399] 1. A "generative artificial intelligence model" is a system that uses machine learning algorithms to generate specific outputs from input data.

[1400] 2. "Patient-reported symptoms" refers to the patient's health condition or physical abnormalities reported to a medical professional.

[1401] 3. "Medical Image Data" means image data used for medical diagnosis or treatment, such as CT scans, MRIs, and X-ray images.

[1402] 4. "Test results" means data based on medical tests such as blood tests, urine tests, and tissue tests.

[1403] 5. "Diagnosis candidate" refers to a diagnosis result that the generative artificial intelligence model may present based on the analysis results.

[1404] 6. "Analysis results" refers to the conclusions or insights obtained by analyzing the input data.

[1405] 7. "Healthcare professionals" are doctors, nurses, clinical laboratory technicians, and other professionals who diagnose, treat, and care for patients.

[1406] 8. "User emotional state" means the mental or psychological state of a healthcare professional or patient.

[1407] 9. "Adjusting the diagnosis presentation method based on emotional state" refers to a method of adaptively changing the display and presentation of diagnosis results depending on the user's emotional state.

[1408] 10. "Final diagnosis" means the patient's final diagnosis determined by a medical professional.

[1409] 11. "Feedback" refers to information such as evaluations and opinions of the system and medical professionals' reactions to diagnostic results.

[1410] 12. “Training Data” means a data set used to train a machine learning model.

[1411] MODE FOR CARRYING OUT THE INVENTION

[1412] The present invention relates to a medical diagnosis support system, and in particular to a system that uses a generative artificial intelligence model and an emotion engine to analyze a patient's symptoms, medical image data, and test results, and presents diagnostic candidates while also taking into account the user's emotional state. The following describes in detail an embodiment of the present invention.

[1413] System configuration

[1414] The system of the present invention includes the following major components:

[1415] 1. Terminal: A device where the user (doctor) inputs patient information and receives analysis results and emotions. Specifically, input devices such as a PC or tablet can be used.

[1416] 2. Server: This is the central system that collects and analyzes data, uses generative AI models to present diagnostic candidates, and analyzes user emotions. This can be implemented on a cloud server or an on-premise server.

[1417] 3. Database: A storage system for storing patient data, analysis results, feedback, and emotion data. For example, an SQL database or a NoSQL database can be used.

[1418] 4. Emotion Engine: An algorithm for recognizing and analyzing the emotional state of users (doctors) and patients. This engine analyzes emotions using natural language processing and image recognition technology.

[1419] Example of a system

[1420] The present invention will be described in detail below based on a specific example of the operation of the system.

[1421] Entering patient information

[1422] The user (doctor) uses the terminal to input basic patient information, symptoms, medical image data, and test results. For example, for a patient complaining of abdominal pain, basic information, CT scan images, and blood test results are input. The emotion engine also detects the user's (doctor's) emotional state in real time and acquires the data.

[1423] Sending data

[1424] The device sends the collected patient information and emotion data to the server using an HTTP request.

[1425] Data storage and preprocessing

[1426] The server stores the received data in a database, after which pre-processing of the data is performed, which includes data integrity checks, cleaning, and format conversion.

[1427] AI model for generating diagnostic candidates

[1428] The preprocessed data is input into a generative AI model on the server, which generates candidate diagnoses based on the patient's symptoms. At the same time, emotional data obtained from the emotion engine is also included in the analysis. For example, the generative AI model references past case data to generate candidate diagnoses such as "acute appendicitis," "intestinal obstruction," and "gastritis" for a patient complaining of abdominal pain.

[1429] Adjusting the display and presentation of results

[1430] The generated diagnosis candidates are sent from the server to the terminal. The terminal displays the diagnosis candidates in a format that is easy for the user (doctor) to understand. The emotion engine also adaptively adjusts the display method of the diagnosis candidates based on the emotion data acquired. For example, for a user with high stress, the diagnosis candidates are displayed in different colors to make them easier to understand visually.

[1431] Enter and submit feedback

[1432] The user (doctor) inputs the final diagnosis and feedback using a terminal and sends it to the server. This feedback data is used to retrain the generative AI model.

[1433] Prompt Sentence Examples

[1434] "Please list possible diagnoses based on the CT scan images and blood test results of a patient who visits the hospital with abdominal pain. Please suggest diagnoses based on medical evidence. Also, please adjust the display method according to the user's (doctor's) emotional state."

[1435] This invention allows doctors to quickly and accurately obtain supplementary information, reducing the risk of misdiagnosis. Furthermore, the use of an emotion engine enables diagnostic support that takes the user's mental state into account, which is expected to improve diagnostic accuracy and the quality of medical care. It is also expected that the accuracy of the system will continue to improve through collected feedback.

[1436] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1437] Step 1: Enter patient information

[1438] The terminal receives input from the user (doctor) of the patient's basic information, symptoms (e.g., abdominal pain), medical image data, and test results. The input includes the patient's basic information, CT scan images, blood test results, etc. Based on this, a basic information dataset is generated, and the emotion engine detects the user's (doctor's) emotional state (e.g., stress, fatigue) in real time and acquires it as additional data.

[1439] Specific behavior:

[1440] The user (doctor) enters the patient's name, age, gender, symptoms, CT scan images, and blood test results into the device's input form and presses the send button. The emotion engine analyzes the user's facial expressions and tone of voice and records their emotional state.

[1441] Step 2: Sending data

[1442] The terminal sends the collected patient information and emotion data to the server. As input, the reference dataset is sent by the terminal to the server. The data is sent using the HTTP request format.

[1443] Specific behavior:

[1444] The device collects all patient information and emotion data, structures it in JSON format, and sends it to the server using an HTTP POST request.

[1445] Step 3: Storing and Preprocessing Data

[1446] The server stores the received data in a database. The input includes the reference dataset sent from the terminal. Then, the data is preprocessed. The preprocessing includes data integrity check, cleaning, and format conversion. The output is a preprocessed clean dataset.

[1447] Specific behavior:

[1448] The server structures the incoming data and stores it in the database using SQL insert queries. After storing it, it detects missing values ​​in the data, imputes them appropriately, and then converts all data into a unified format.

[1449] Step 4: Generate candidate diagnoses using the AI ​​model

[1450] The server inputs the preprocessed data into a generative AI model to generate candidate diagnoses. The input includes a clean, preprocessed dataset. During the analysis, the AI ​​model references past case data and medical knowledge. The output is multiple candidate diagnoses.

[1451] Specific behavior:

[1452] The server runs a generative AI model using a deep learning framework such as TensorFlow or PyTorch, inputs the preprocessed data into the model, and the model analyzes the data to generate candidate diagnoses such as "acute appendicitis," "intestinal obstruction," and "gastritis."

[1453] Step 5: Adjust the display and presentation of results

[1454] The server organizes the generated candidate diagnoses and sends them to the terminal. The input includes the generated candidate diagnoses. The terminal displays the candidate diagnoses and adjusts the display method based on the emotional state of the user (doctor). The output is the adjusted display of the candidate diagnoses.

[1455] Specific behavior:

[1456] The server sends the generated diagnosis candidates in JSON format to the terminal as an HTTP response. The terminal receives this and displays the diagnosis candidates through a GUI. Based on the results of the emotion engine analysis, if the doctor is in a state of high stress, the diagnosis candidates are color-coded to make them easier to understand visually.

[1457] Step 6: Enter and submit your feedback

[1458] The terminal allows the user (doctor) to input the final diagnosis and feedback, and sends the data to the server. The input includes the final diagnosis and feedback. The output is the feedback data sent to the server.

[1459] Specific behavior:

[1460] The user (doctor) enters the final diagnosis (e.g., acute appendicitis) and feedback about the system's diagnostic accuracy into the input form on the terminal, and presses the submit button. The terminal structures this data in JSON format and sends it to the server using an HTTP POST request.

[1461] Step 7: Save feedback and retrain

[1462] The server stores the received feedback data in a database and uses it to retrain the generative AI model. The feedback data is included as input. The output is an updated generative AI model.

[1463] Specific behavior:

[1464] The server adds and stores the feedback data in a database, then retrains the generative AI model using a new dataset that includes the feedback data, e.g., updating model parameters to improve the model's performance.

[1465] (Application example 2)

[1466] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1467] The problem that this invention aims to solve is to improve the accuracy and efficiency of diagnoses by reducing the risk of misdiagnosis caused by information overload, time constraints, and the doctor's own emotional state when making a diagnosis in the medical field. Another problem is to improve the safety and efficiency of work by analyzing the emotional state of security operators in real time and utilizing that information when generating alerts.

[1468] 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.

[1469] In this invention, the server includes means for analyzing patient-reported symptoms, medical image data, and test results using a generative AI model and presenting diagnosis candidates, means for displaying the analysis results to medical professionals, means for collecting final diagnosis results and feedback from medical professionals and using them as training data for the generative AI model, means for collecting and analyzing biometric data in real time, and means for generating and displaying alerts based on the analysis results and biometric data, thereby enabling improved accuracy and efficiency of diagnoses in medical settings and improved safety and efficiency in security operations.

[1470] A "generative artificial intelligence model" refers to an algorithm or machine learning model that analyzes data collected in medical diagnosis or security analysis and generates diagnostic candidates or anomaly detection results.

[1471] "Patient-reported symptoms" refers to information about specific symptoms such as poor health or pain that patients self-report in medical settings.

[1472] "Medical image data" refers to image information of a patient's inside the body obtained using medical equipment such as CT scans, X-ray images, and MRIs.

[1473] "Test results" refers to numerical data and information indicating whether or not there are abnormalities obtained from blood tests, urine tests, various biochemical tests, etc.

[1474] "Analysis results" refers collectively to the diagnosis candidates and anomaly detection results obtained after analyzing input data using a generative artificial intelligence model.

[1475] "Medical professionals" refers to professionals who perform medical-related tasks, such as doctors, nurses, and radiological technologists.

[1476] "Biometric data" refers to information related to the physical and mental state of medical professionals, security operators, etc., such as heart rate, body temperature, and facial expression data.

[1477] An "alert" refers to a warning or caution message issued by the system based on analysis results or abnormal values ​​in biometric data.

[1478] "Database" refers to a system that organizes and stores large amounts of information, such as reported symptoms, medical image data, test results, and analysis results, and enables quick search and retrieval as needed.

[1479] "Preprocessing" refers to the process of checking the integrity of collected data, cleaning it, and converting its format to make it suitable for subsequent analysis.

[1480] "Feedback" refers to the evaluations and opinions of medical professionals and security operators regarding the final diagnosis results and system proposals.

[1481] The embodiments for carrying out the present invention will be described below.

[1482] System Overview

[1483] The system of the present invention includes the following major components:

[1484] 1. Terminal: A device used by the user (doctor or security operator) to input and display basic information, biometric data, analysis results, and emotional state of the patient.

[1485] 2. Server: A central system that collects and analyzes data, generates diagnostic candidates and anomaly detection results using generative AI models, integrates the analysis results with biometric data, and generates appropriate alerts.

[1486] 3. Database: A storage system for storing reported symptoms, medical image data, test results, analysis results, feedback, emotional data, etc.

[1487] 4. Emotion engine: An algorithm that recognizes and analyzes the emotional state of the user (doctor or security operator).

[1488] Data collection

[1489] The terminal collects basic information about the patient and operator, as well as biometric data (heart rate, body temperature, facial expression data) from the user, and also collects video data and medical image data from security cameras and medical equipment.

[1490] Data analysis

[1491] The server stores the received data in a database and performs preprocessing. This involves checking the data for consistency, cleaning it, and converting its format. Once this is complete, the data is input into the generative AI model for analysis. At the same time, the emotion data obtained by the emotion engine is used for analysis.

[1492] Alert Generation

[1493] Based on the analysis results and emotion data, the server generates appropriate alerts. For example, if an abnormality is detected and an operator is in a high stress state at the same time, an alert is displayed indicating that immediate action is required. The alert is sent to the terminal and displayed to the user.

[1494] Feedback collection

[1495] Feedback on analysis results and alerts is collected from users. This feedback is sent to the server and stored in a database. The collected feedback is used as retraining data for the generative AI model and is used to improve the accuracy of the system.

[1496] Specific examples

[1497] When a doctor uses a device to enter basic patient information, CT scan images, and blood test results, the data is sent to a server. The server preprocesses the data and analyzes it using a generative AI model. As a result, a candidate diagnosis is generated and displayed on the device. At the same time, an emotion engine analyzes the doctor's emotional state and adjusts the way the candidate diagnosis is presented based on that data.

[1498] If a security operator wears smart glasses during the night shift and the system simultaneously detects an abnormality and high stress, an alert message will be displayed stating, "High stress and abnormality detected! Immediate action required!"

[1499] Example prompts to input to the generative AI model

[1500] "If an anomaly is detected, how can we tailor our response to reduce operator burden?"

[1501] "A method for detecting operator stress levels in real time and proposing countermeasures during security operations."

[1502] In this way, the present invention enables efficient information presentation and alert generation that integrates analysis results and emotional data in medical settings and security operations.

[1503] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1504] Step 1:

[1505] Data Collection Phase

[1506] Input: Basic information from the user (doctor or security operator), biometric data (heart rate, body temperature, facial expression data), video data and medical image data from security cameras and medical equipment.

[1507] Output: Sends collected data to a server.

[1508] Specific behavior:

[1509] The user uses the terminal to input basic patient information and biometric data, and data is collected in real time from security cameras and medical devices, which are then transmitted to a server.

[1510] Step 2:

[1511] Database save phase

[1512] Input: Basic information, biometric data, video data, and medical image data sent to the server.

[1513] Output: Various data stored in the database.

[1514] Specific behavior:

[1515] The server stores the received data in a database, where it checks for consistency.

[1516] Step 3:

[1517] Preprocessing Phase

[1518] Input: Basic information, biometric data, video data, and medical image data stored in the database.

[1519] Output: The preprocessed data.

[1520] Specific behavior:

[1521] The server retrieves data from the database, checks its integrity, cleans it, and converts it into a format suitable for generative AI models.

[1522] Step 4:

[1523] Data analysis phase

[1524] Input: Preprocessed basic information, biometric data, video data, and medical image data.

[1525] Output: Diagnosis candidates, anomaly detection results, and sentiment analysis results from the generative AI model.

[1526] Specific behavior:

[1527] The server inputs the preprocessed data into the generative AI model for analysis, while simultaneously analyzing and integrating the user's emotional data obtained by the emotion engine.

[1528] Step 5:

[1529] Alert Generation Phase

[1530] Input: Diagnosis candidates, anomaly detection results, and sentiment analysis results.

[1531] Output: The alert message.

[1532] Specific behavior:

[1533] The server generates appropriate alerts based on the analysis results and biometric data. For example, if an abnormality is detected and an operator is in a high stress state, an alert message requiring immediate action is generated. The generated alert is then sent to the terminal.

[1534] Step 6:

[1535] Display Phase

[1536] Input: Alert message, diagnosis candidates, and sentiment analysis results sent from the server.

[1537] Output: Alert message and potential diagnosis displayed on the terminal.

[1538] Specific behavior:

[1539] The device receives alert messages and diagnostic candidates from the server and displays them to the user, adjusting the display format according to the results of emotion analysis.

[1540] Step 7:

[1541] Feedback gathering phase

[1542] Input: Final diagnostic results and feedback from the user.

[1543] Output: Feedback sent to the server.

[1544] Specific behavior:

[1545] The user (doctor or security operator) inputs feedback on the final diagnosis and the evaluation of the alert through the terminal, and this data is sent to the server and stored in the database.

[1546] Step 8:

[1547] Model retraining phase

[1548] Input: Stored feedback, newly collected data.

[1549] Output: A generative artificial intelligence model with improved accuracy.

[1550] Specific behavior:

[1551] The server retrains the generative AI model based on collected feedback and newly acquired data, thereby continuously improving diagnostic accuracy and anomaly detection performance.

[1552] This concludes the explanation of each processing step of this program and its specific operations.

[1553] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1554] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1555] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1556] [Fourth embodiment]

[1557] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1558] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1559] 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).

[1560] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1561] 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.

[1562] 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).

[1563] 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.

[1564] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1565] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1566] 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.

[1567] 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.

[1568] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1569] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1570] The embodiments for carrying out the present invention will be described below.

[1571] System Overview

[1572] The present invention relates to a medical diagnosis support system, in particular a system that uses a generative artificial intelligence model to analyze a patient's symptoms, medical image data, and test results, and presents potential diagnoses. The system includes the following main components:

[1573] 1. Terminal: A device where the user (doctor) enters patient information and receives analysis results.

[1574] 2. Server: The central system that collects and analyzes data and uses generative AI models to present potential diagnoses.

[1575] 3. Database: A storage system for storing patient data, analysis results, and feedback.

[1576] Explaining program processing in natural language

[1577] Data Collection Phase

[1578] The terminal receives patient information from the user (doctor) and transmits it to the server. Specifically, this information includes the patient's basic information, reported symptoms, medical image data, and test results.

[1579] Example: A user (doctor) uses a terminal to input basic information about a patient complaining of abdominal pain, CT scan images, and blood test results, and sends them to a server.

[1580] Data analysis phase

[1581] The server stores the received data in a database and performs preprocessing, which includes checking data consistency, cleaning, and format conversion. Once preprocessing is complete, the data is input into a generative AI model for analysis.

[1582] Example: The server preprocesses data received about a patient's abdominal pain and inputs it into a generative AI model. The generative AI model analyzes the data based on past cases of abdominal pain and medical knowledge to generate candidate diagnoses.

[1583] Diagnosis candidate presentation phase

[1584] The server organizes the diagnostic candidates obtained from the generative AI model and sends them to the device used by the user (doctor). The device displays the diagnostic candidates in a format that is easy for the user to understand.

[1585] Example: The server generates diagnostic candidates such as "acute appendicitis" and "intestinal obstruction" and sends them to the terminal so that the user (doctor) can check them.

[1586] Feedback gathering phase

[1587] The device allows the user (doctor) to input the final diagnosis and feedback. This data is sent to the server and stored in a database. This feedback is then used to retrain the generative AI model, contributing to improving the accuracy of the system.

[1588] For example, a user (doctor) diagnoses "acute appendicitis" and provides feedback that the system's diagnosis was appropriate. This information is sent to the server and used as training data for the generative AI model.

[1589] Specific operation example

[1590] When a patient visits the hospital with abdominal pain, the user (doctor) uses a terminal to input the patient's basic information, CT scan images, blood test results, and other test results. This data is sent to the server and stored in a database. The server preprocesses the data and inputs it into a generative AI model for analysis. The generated diagnosis candidates are sent to the terminal and displayed to the user (doctor). After the final diagnosis is made, feedback is sent from the terminal to the server, stored in a database, and used again as training data for the generative AI model.

[1591] The system provides doctors with fast and accurate supplementary information, reducing the risk of misdiagnosis, and continuously improves its accuracy through collected feedback.

[1592] The processing flow will be explained below.

[1593] Specific explanation of the program's processing steps

[1594] Data Collection Phase

[1595] Step 1:

[1596] The terminal receives basic patient information from the user (doctor), including the patient's name, age, sex, and other basic information.

[1597] pseudo

[1598] The terminal receives basic patient information from the user.

[1599] Step 2:

[1600] The terminal receives input from the user (doctor) of the patient's reported symptoms, including the specific symptoms and progress of the patient's awareness.

[1601] pseudo

[1602] The terminal inputs the patient's reported symptoms.

[1603] Step 3:

[1604] The terminal receives input of medical image data (e.g., X-rays, CT scans) and test results (e.g., blood test results, urine test results) from the user (doctor).

[1605] pseudo

[1606] The terminal inputs medical image data and test results

[1607] Step 4:

[1608] The device sends all collected data to the server.

[1609] pseudo

[1610] The device sends the input data to the server

[1611] Data analysis phase

[1612] Step 5:

[1613] The server receives the patient data transmitted from the terminal.

[1614] pseudo

[1615] The server receives data from the device

[1616] Step 6:

[1617] The server stores the received patient data in a database.

[1618] pseudo

[1619] The server stores the data in a database

[1620] Step 7:

[1621] The server checks the integrity of the stored data and performs pre-processing (data cleaning, format conversion, etc.).

[1622] pseudo

[1623] The server performs data integrity checks and preprocessing

[1624] Step 8:

[1625] The server inputs the preprocessed data into the generative AI model.

[1626] pseudo

[1627] The server inputs the preprocessed data into the generative AI model

[1628] Step 9:

[1629] The server obtains the candidate diagnosis results from the generative AI model.

[1630] pseudo

[1631] The server receives the analysis results from the generated AI model

[1632] Diagnosis candidate presentation phase

[1633] Step 10:

[1634] The server formats the acquired diagnosis candidates into a format that is easy for the user (doctor) to understand.

[1635] pseudo

[1636] The server formats the candidate diagnoses

[1637] Step 11:

[1638] The server sends the formatted diagnosis candidates to the terminal and presents them to the user (doctor).

[1639] pseudo

[1640] The server sends the diagnosis candidate to the terminal.

[1641] The device displays possible diagnoses to the user (doctor)

[1642] Feedback gathering phase

[1643] Step 12:

[1644] The terminal allows the user (doctor) to input the final diagnosis result.

[1645] pseudo

[1646] The terminal receives the final diagnosis result from the user.

[1647] Step 13:

[1648] The terminal allows the user (doctor) to input feedback on the system's diagnostic accuracy and areas for improvement.

[1649] pseudo

[1650] The device receives feedback from the user

[1651] Step 14:

[1652] The terminal sends the final diagnosis result and feedback to the server.

[1653] pseudo

[1654] The device sends the final diagnosis results and feedback to the server

[1655] Step 15:

[1656] The server stores the final diagnosis results and feedback in a database.

[1657] pseudo

[1658] The server stores the final diagnosis results and feedback in a database

[1659] Step 16:

[1660] The server uses the collected feedback to retrain the generative AI model.

[1661] pseudo

[1662] The server uses the feedback to retrain the generative AI model

[1663] Through this series of processing flows, the system of the present invention can assist users (doctors) in making diagnoses, improving diagnostic accuracy while reducing the risk of misdiagnosis. Furthermore, through continuous improvements based on feedback, the reliability and effectiveness of the system can be increased.

[1664] Example 1

[1665] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1666] Conventional medical diagnosis support systems have difficulty efficiently analyzing patients' reported symptoms, medical image data, and test results, and presenting appropriate diagnostic candidates. Furthermore, there is a lack of a method for effectively aggregating feedback from medical professionals during the process of making a final diagnosis and improving the accuracy of the diagnosis support system. This increases the risk of misdiagnosis and impacts the quality of medical care.

[1667] 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.

[1668] In this invention, the server includes a means for transmitting patient information input from a terminal to the server, a means for the server to store the received data in a database and perform data preprocessing, a means for analyzing the patient's reported symptoms, medical image data, and test results using a generative AI model and presenting candidate diagnoses, a means for displaying the analysis results to medical professionals, and a means for collecting final diagnoses and feedback from medical professionals and using them as training data for the generative AI model. This enables efficient collection and preprocessing of patient data, presentation of appropriate candidate diagnoses, collection of feedback, and improvement of system accuracy.

[1669] A "generative artificial intelligence model" is a model generated using artificial intelligence technology that analyzes a patient's symptoms and test results based on medical data and generates potential diagnoses.

[1670] "Reported symptoms" are physical discomfort or abnormalities that patients report to medical professionals, and are information that forms the basis of diagnosis.

[1671] "Medical image data" refers to image data obtained using medical equipment, such as CT scans, MRIs, and X-rays, that provide the visual information necessary for diagnosis.

[1672] "Test results" refer to the numerical data and analytical results obtained through various medical tests, such as blood tests and urine tests, and are important information that assists in diagnosis.

[1673] "Candidate diagnoses" are a list of possible diagnoses based on the patient's symptoms and test results, presented as a result of the generative AI model's analysis, and serve as a reference for medical professionals to make a final decision.

[1674] A "terminal" is a device used by a user (doctor) to enter patient information and receive analysis results, and includes PCs, tablets, smartphones, etc.

[1675] The "server" is a central system that receives, stores, and preprocesses patient information, analyzes it using a generative AI model, and presents diagnostic candidates, and also connects to the database.

[1676] A "database" is a storage system for systematically storing patient data, analysis results, and feedback from medical professionals, and is used for subsequent data reference and analysis.

[1677] "Preprocessing" refers to the process of preparing the data to be analyzed, and includes checking data consistency, cleaning, format conversion, etc.

[1678] "Feedback" refers to opinions and evaluations provided by medical professionals based on final diagnosis results, and is data used to retrain generative AI models.

[1679] System Overview

[1680] This invention relates to a medical diagnosis support system that uses a generative AI model to analyze patient-reported symptoms, medical image data, and test results, and presents potential diagnoses. The system includes the following main components:

[1681] 1. Terminal: A device where the user (doctor) enters patient information and receives analysis results. This includes PCs, tablets, smartphones, etc.

[1682] 2. Server: This is the central system that collects and analyzes data, uses generative AI models to present diagnostic candidates, and connects to the database.

[1683] 3. Database: A storage system for storing patient data, analysis results, and feedback.

[1684] Explaining program processing in natural language

[1685] Data Collection Phase

[1686] A user (doctor) uses a terminal to input patient information and sends it to a server. Specifically, this includes the patient's basic information (name, age, gender), reported symptoms, medical image data (CT scans and MRIs), and various test results (blood tests, urine tests, etc.). For example, the user (doctor) inputs the information of a 30-year-old male patient complaining of abdominal pain into the terminal and sends it to the server.

[1687] Data analysis phase

[1688] The server stores the received data in a database and then performs preprocessing on the data. Preprocessing includes checking data integrity, cleaning, and any necessary format conversions. Once preprocessed, the data is input into a generative AI model. The server preprocesses the patient information and inputs it into a generative AI model built using TensorFlow and PyTorch for analysis. The generative AI model performs analysis based on past case data and medical knowledge to generate candidate diagnoses.

[1689] Diagnosis candidate presentation phase

[1690] The server organizes the generated candidate diagnoses and sends them to the terminal. The terminal displays the candidate diagnoses in a format that is easy for the user (doctor) to understand. For example, the terminal displays "acute appendicitis" and "intestinal obstruction" as part of the list of candidate diagnoses generated by the server.

[1691] Feedback gathering phase

[1692] The user (doctor) enters the final diagnosis and feedback into the device and sends it to the server. The feedback is then stored in a database and used to retrain the generative AI model. This feedback continuously improves the accuracy of the system. For example, if the user (doctor) diagnoses "acute appendicitis" and enters feedback that the diagnosis was appropriate, this is sent to the server and used as training data for the generative AI model.

[1693] Prompt Sentence Examples

[1694] "Generate possible diagnoses based on the CT scan images and blood test results of a 30-year-old man complaining of abdominal pain."

[1695] This system will enable doctors to obtain supplementary information quickly and accurately, reducing the risk of misdiagnosis. Furthermore, it is expected that the accuracy of the generative AI model will be improved through collected feedback.

[1696] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1697] Step 1:

[1698] The terminal accepts input from the user (doctor).

[1699] Specific operation: The user (doctor) uses the terminal to input the patient's basic information (name, age, gender), reported symptoms, medical image data (CT scan or MRI), and various test results (blood test, urine test, etc.).

[1700] Input: Patient's basic information, reported symptoms, medical image data, and test results.

[1701] Output: The entered patient information.

[1702] Step 2:

[1703] The terminal transmits the input information to the server.

[1704] Specific operation: The terminal encrypts the entered patient data and sends it to the server using the TLS protocol.

[1705] Input: Patient information entered into the terminal.

[1706] Output: Patient information sent to the server.

[1707] Step 3:

[1708] The server stores the received data in a database.

[1709] Specific operation: The server adds the received patient data to the "Patient Data" table in the database as a new record.

[1710] Input: Patient information received by the server.

[1711] Output: Patient information stored in the database.

[1712] Step 4:

[1713] The server pre-processes the data.

[1714] Specific operation: The server checks the integrity of the received data, removes invalid values, fills in missing data, and converts the format.

[1715] Input: Raw data stored in a database.

[1716] Output: Consistency checked and preprocessed data.

[1717] Step 5:

[1718] The server inputs the preprocessed data into the generative AI model.

[1719] Specific operation: The server inputs the preprocessed data into a generative AI model using TensorFlow or PyTorch.

[1720] Input: Preprocessed data.

[1721] Output: The data that is input into a generative AI model.

[1722] Step 6:

[1723] A generative AI model analyzes the data and generates candidate diagnoses.

[1724] How it works: The generative AI model analyzes the pre-processed data and generates candidate diagnoses based on past case data and medical knowledge.

[1725] Input: Preprocessed data fed into a generative AI model.

[1726] Output: The generated candidate diagnoses.

[1727] Step 7:

[1728] The server transmits the generated candidate diagnoses to the terminal.

[1729] Specific operation: The server organizes the generated diagnosis candidate list and sends it to the terminal.

[1730] Input: A list of diagnostic candidates output from a generative AI model.

[1731] Output: A list of diagnostic candidates sent to the terminal.

[1732] Step 8:

[1733] The terminal displays the possible diagnoses to the user (doctor).

[1734] Specific operation: The terminal displays the list of diagnosis candidates in a format (for example, a list format) that is easy for the user (doctor) to understand.

[1735] Input: A list of diagnostic candidates sent to the terminal.

[1736] Output: A list of possible diagnoses displayed to the user (doctor).

[1737] Step 9:

[1738] The user (doctor) inputs the diagnosis and feedback.

[1739] Specific operation: The user (doctor) inputs the final diagnosis and feedback to the system into the terminal.

[1740] Input: User's (doctor's) final diagnosis and feedback.

[1741] Output: Feedback information entered into the terminal.

[1742] Step 10:

[1743] The terminal sends feedback information to the server.

[1744] Specific operation: The terminal encrypts the input feedback information and sends it to the server.

[1745] Input: Feedback information entered into the device.

[1746] Output: Feedback information sent to the server.

[1747] Step 11:

[1748] The server stores the feedback information in a database.

[1749] Specific operation: The server stores the feedback information in a "diagnostic feedback" table in the database.

[1750] Input: Feedback information received by the server.

[1751] Output: Feedback information stored in a database.

[1752] Step 12:

[1753] The server retrains the generative AI model based on the feedback.

[1754] Specific operation: The server periodically extracts feedback information from the database and retrains the generative AI model.

[1755] Input: Feedback information stored in the database.

[1756] Output: The retrained generative AI model.

[1757] (Application example 1)

[1758] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1759] In modern medical settings, doctors are required to make quick and accurate diagnoses within a limited time frame, but this is often difficult. While selecting the right medicine is also important for pharmacies and drugstores, it can be difficult to accurately understand a patient's symptoms and provide appropriate advice. Therefore, there is a need for a system that can support doctors and pharmacists in the diagnostic process.

[1760] 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.

[1761] In this invention, the server includes means for analyzing patient-reported symptoms, medical image data, and test results using a generative AI model and presenting candidate diagnoses, means for displaying the analysis results to users such as medical professionals or patients, means for collecting final diagnosis results and feedback from users and using them as training data for the generative AI model, and means for using the medical diagnosis support system in physical stores such as drugstores. This allows doctors and pharmacists to quickly and accurately present candidate diagnoses, enabling them to select appropriate medicines and medical procedures.

[1762] A "generative artificial intelligence model" is an artificial intelligence algorithm used to analyze data such as patient symptoms, medical images, and test results to generate potential diagnoses.

[1763] "Reported symptoms" refers to any ill health or specific symptoms reported by the patient.

[1764] "Medical image data" refers to image data taken using medical equipment such as CT scans and MRIs.

[1765] "Test results" refers to data obtained from various medical tests, such as blood tests and urine tests.

[1766] "Analysis results" refer to the diagnostic candidates and related information obtained using the generative artificial intelligence model.

[1767] "Healthcare professionals" refer to professionals involved in diagnosing and treating patients, such as doctors and nurses.

[1768] "Feedback" refers to the final diagnostic results and evaluations and opinions from users.

[1769] "Training Data" means data used to improve and optimize a generative artificial intelligence model.

[1770] A "medical diagnosis support system" is a system that uses a generative artificial intelligence model to present potential diagnoses to medical professionals and patients, helping them make a final diagnosis.

[1771] The embodiments for carrying out the present invention will be described below.

[1772] System Overview

[1773] The present invention relates to a medical diagnosis support system, in particular a system that uses a generative artificial intelligence model to analyze a patient's symptoms, medical image data, and test results, and presents potential diagnoses. The system includes the following main components:

[1774] 1. Terminal: A device where users (doctors and patients) enter symptoms and test results and receive analysis results. Specifically, a smartphone, tablet, or interactive display is used.

[1775] 2. Server: This is the central system that collects, analyzes, and stores data, and uses generative AI models to present diagnostic candidates. A cloud server (e.g., Amazon Web Services or Microsoft Azure) is used.

[1776] 3. Database: A storage system for storing patient data, analysis results, and feedback. SQL / NoSQL databases (MySQL, MongoDB, PostgreSQL) are used.

[1777] Program Processing and Technical Details

[1778] Data Collection Phase

[1779] Users (doctors or patients) use a terminal to input their symptoms, medical images, test results, etc. For example, a patient complaining of abdominal pain inputs their age, gender, CT scan images, and blood test results. This data is then sent to a server via the Internet.

[1780] Data analysis phase

[1781] The server stores the received data in a database and performs preprocessing (integrity checks, cleaning, format conversion) on the data. After preprocessing is complete, the data is input into a generative artificial intelligence model for analysis. This process uses machine learning libraries such as TensorFlow and PyTorch.

[1782] Diagnosis candidate presentation phase

[1783] The candidate diagnoses obtained from the analysis results are organized on the server and sent to the terminal, which then displays them in a format that is easy for the user to understand. For example, candidate diagnoses such as "acute appendicitis" or "intestinal obstruction" are displayed.

[1784] Feedback gathering phase

[1785] The final diagnosis and feedback are entered on the device and sent to the server, where they are stored in a database and used to retrain the generative AI model.

[1786] Specific operation example

[1787] When a patient visits a hospital or drugstore with abdominal pain, the user (doctor or pharmacist) uses a terminal to enter the patient's basic information, CT scan images, blood test results, and other test results. For example, the user might enter, "I have abdominal pain and have had similar symptoms several times in the past. I am 45 years old. I also have CT scan images. Please let me know if you have any diagnostic candidates." This data is sent to a server and stored in a database. The server preprocesses the data and inputs it into a generative AI model for analysis. The generated diagnostic candidates are sent to the terminal and displayed to the user (doctor or patient). After the final diagnosis is made, feedback is sent from the terminal to the server, stored in a database, and used as training data for the generative AI model.

[1788] The system allows users to quickly and accurately identify potential diagnoses, reducing the risk of misdiagnosis, and continuously improves its accuracy through collected feedback.

[1789] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1790] Step 1:

[1791] The terminal receives input from the user (doctor or patient) about symptoms, basic information, medical image data, and test results. The input data includes age, gender, and details of symptoms. This input data is sent to the server via the Internet. An example of a prompt for the user to input specific information is, "I have abdominal pain and have had similar symptoms several times in the past. I am 45 years old. I also have a CT scan image. Please let me know if you have any possible diagnoses."

[1792] Step 2:

[1793] The server receives data sent from the terminal and stores it in a database. The stored data may not be suitable for analysis as is, so it must be preprocessed. Specific preprocessing steps include checking data consistency, filling in missing data, and standardizing formats. This process ensures the availability and reliability of input data.

[1794] Step 3:

[1795] The server inputs the preprocessed data into the artificial intelligence model. Specifically, machine learning libraries such as TensorFlow and PyTorch are used to input the data into the model. The model then performs analysis based on the input data, using past case data and a medical knowledge base. As a result of the analysis, candidate diagnoses are generated.

[1796] Step 4:

[1797] The server organizes the generated candidate diagnoses and sends them to the terminal. The terminal displays the analysis results in an easy-to-read format for the user (doctor or patient). At this stage, an appropriate user interface design is important to present the analysis results in an easy-to-understand format. For example, specific candidate diagnoses such as "acute appendicitis" or "intestinal obstruction" are displayed.

[1798] Step 5:

[1799] The user uses the terminal to input the final diagnosis results and feedback. The user also inputs feedback on the diagnosis performed based on the confirmed diagnosis candidates and their results into the terminal. This includes an evaluation of the accuracy of the diagnosis and the information presented. This feedback data is sent to the server.

[1800] Step 6:

[1801] The server stores the final diagnosis results and feedback sent by the user in a database and uses them as training data for the generative AI model. By retraining the generative AI model based on this feedback data, the accuracy of the model is continuously improved, further increasing the reliability and usefulness of the system.

[1802] Through the above processing steps, this medical diagnosis support system is able to efficiently and accurately analyze medical data and present potential diagnoses.

[1803] 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.

[1804] The embodiments for carrying out the present invention will be described below.

[1805] System Overview

[1806] The present invention relates to a medical diagnosis support system, in particular a system that uses a generative artificial intelligence model and an emotion engine to analyze patient symptoms, medical image data, and test results, and presents potential diagnoses while also taking into account the user's emotional state. The system includes the following main components:

[1807] 1. Terminal: A device where the user (doctor) inputs patient information and receives analysis results and emotions.

[1808] 2. Server: The central system that collects and analyzes data, uses generative AI models to present diagnostic candidates, and analyzes user sentiment.

[1809] 3. Database: A storage system for storing patient data, analysis results, feedback and emotion data.

[1810] 4. Emotion engine: Algorithms for recognizing and analyzing the emotional state of the user (doctor) and patient.

[1811] Explaining program processing in natural language

[1812] Data Collection Phase

[1813] The terminal receives basic patient information from the user (doctor) and sends it to the server. This information includes the patient's basic information, reported symptoms, medical image data, and test results. The emotion engine also recognizes the user's (doctor's) emotional state and captures it as data.

[1814] Example: A user (doctor) uses a terminal to input basic information about a patient complaining of abdominal pain, CT scan images, and blood test results, and at the same time, the emotion engine detects the user's (doctor's) stress level and transmits the data.

[1815] Data analysis phase

[1816] The server stores the received data in a database and performs preprocessing. Preprocessing includes checking data consistency, cleaning, and format conversion. Once preprocessed, the data is input into a generative AI model for analysis. Furthermore, emotional data of users and patients obtained by the emotion engine is also used in the analysis.

[1817] Example: The server preprocesses data received about a patient's abdominal pain and inputs it into a generative AI model. The generative AI model analyzes the data based on past cases of abdominal pain and medical knowledge to generate a candidate diagnosis. At the same time, it evaluates the possibility that the doctor's emotional state may affect the diagnosis based on the results of the emotion engine analysis.

[1818] Diagnosis candidate presentation phase

[1819] The server organizes the acquired diagnosis candidates and sends them to the device used by the user (doctor). The device displays the diagnosis candidates and emotion analysis results in a format that is easy for the user (doctor) to understand. The way in which the diagnosis candidates are presented is also adaptively changed depending on the emotion analysis results.

[1820] For example, the server generates diagnostic candidates such as "acute appendicitis" or "intestinal obstruction" and sends them to the terminal for the user (doctor) to review. At the same time, the emotion engine adjusts the way the diagnostic candidates are displayed based on the emotional state recognized by the server.

[1821] Feedback gathering phase

[1822] The device allows the user (doctor) to input the final diagnosis, feedback, and emotional data. This data is sent to the server and stored in a database. This feedback is then used to retrain the generative AI model, contributing to improving the accuracy of the system.

[1823] Example: A user (doctor) diagnoses "acute appendicitis" and provides feedback that the system's diagnosis was appropriate. This information is sent to the server and used as training data for the generative AI model. The emotional state collected by the emotion engine is also stored.

[1824] Specific operation example

[1825] When a patient visits the hospital with abdominal pain, the user (doctor) uses a terminal to input the patient's basic information, CT scan images, blood test results, and other test results. This data is sent to the server and stored in a database. The server preprocesses the data and inputs it into a generative AI model for analysis. The generated diagnosis candidates are sent to the terminal and displayed to the user (doctor). After the final diagnosis is made, feedback is sent from the terminal to the server, stored in a database, and used again as training data for the generative AI model. At the same time, an emotion engine detects the user's emotional state and uses this data for analysis.

[1826] This system allows doctors to obtain supplementary information quickly and accurately, reducing the risk of misdiagnosis. Furthermore, the use of an emotion engine enables diagnostic support that takes the user's mental state into account, which is expected to improve diagnostic accuracy and the quality of medical care. The accuracy of the system is continuously improved through collected feedback.

[1827] The processing flow will be explained below.

[1828] Specific explanation of the program's processing steps

[1829] Data Collection Phase

[1830] Step 1:

[1831] The terminal receives basic patient information from the user (doctor), including basic information such as the patient's name, age, sex, height, and weight.

[1832] pseudo

[1833] The terminal receives basic patient information from the user.

[1834] Step 2:

[1835] The terminal receives input from the user (doctor) of the patient's reported symptoms, including the specific symptoms and progress of the symptoms that the patient is aware of.

[1836] pseudo

[1837] The terminal inputs the patient's reported symptoms.

[1838] Step 3:

[1839] The terminal receives input of medical image data (e.g., X-rays, CT scans) and test results (e.g., blood test results, urine test results) from the user (doctor).

[1840] pseudo

[1841] The terminal inputs medical image data and test results

[1842] Step 4:

[1843] The emotion engine recognizes the user's (doctor's) emotional state and captures it as data, including the doctor's stress level and fatigue level.

[1844] pseudo

[1845] Emotion engine recognizes the doctor's emotional state

[1846] Step 5:

[1847] The device sends all collected data to the server.

[1848] pseudo

[1849] The device sends the input data to the server

[1850] Data analysis phase

[1851] Step 6:

[1852] The server receives the patient data and emotion data transmitted from the terminal.

[1853] pseudo

[1854] The server receives data from the device

[1855] Step 7:

[1856] The server stores the received patient data in a database.

[1857] pseudo

[1858] The server stores the data in a database

[1859] Step 8:

[1860] The server checks the integrity of the stored data and performs pre-processing (data cleaning, format conversion, etc.).

[1861] pseudo

[1862] The server performs data integrity checks and preprocessing

[1863] Step 9:

[1864] The server inputs the preprocessed data into a generative AI model, including emotional data.

[1865] pseudo

[1866] The server inputs the preprocessed data into the generative AI model

[1867] Step 10:

[1868] The server obtains the candidate diagnosis results from the generative AI model.

[1869] pseudo

[1870] The server receives the analysis results from the generated AI model

[1871] Diagnosis candidate presentation phase

[1872] Step 11:

[1873] The server formats the diagnosis candidates obtained into a format that is easy for the user (doctor) to understand, and adjusts the results by reflecting emotional data.

[1874] pseudo

[1875] The server formats the candidate diagnoses

[1876] The server adjusts the results by taking into account the emotional data.

[1877] Step 12:

[1878] The server sends the formatted diagnosis candidates to the terminal and presents them to the user (doctor).

[1879] pseudo

[1880] The server sends the diagnosis candidate to the terminal.

[1881] The device displays possible diagnoses to the user (doctor)

[1882] Feedback gathering phase

[1883] Step 13:

[1884] The terminal allows the user (doctor) to input the final diagnosis result.

[1885] pseudo

[1886] The terminal receives the final diagnosis result from the user.

[1887] Step 14:

[1888] The device allows the user (doctor) to input feedback on the system's diagnostic accuracy and areas for improvement, along with their emotional state.

[1889] pseudo

[1890] The device receives feedback from the user

[1891] The device also collects emotional state data

[1892] Step 15:

[1893] The terminal sends the final diagnosis result and feedback to the server.

[1894] pseudo

[1895] The device sends the final diagnosis results and feedback to the server

[1896] Step 16:

[1897] The server stores the final diagnosis results and feedback in a database.

[1898] pseudo

[1899] The server stores the final diagnosis results and feedback in a database

[1900] Step 17:

[1901] The server uses the collected feedback to retrain the generative AI model, including the emotional data.

[1902] pseudo

[1903] The server uses the feedback to retrain the generative AI model

[1904] Through this series of processing flows, the system of the present invention provides diagnostic support to the user (doctor), improving diagnostic accuracy while reducing the risk of misdiagnosis. Furthermore, by using an emotion engine, diagnostic support that takes into account the user's mental state is possible, which is expected to improve diagnostic accuracy and the quality of medical care. The accuracy of the system is continuously improved through collected feedback.

[1905] Example 2

[1906] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1907] Traditionally, medical diagnoses have relied primarily on the doctor's experience and knowledge, leading to potential misdiagnoses and oversights. Furthermore, the doctor's emotional state often influences the diagnosis. Furthermore, systems that support diagnosis using generative AI models do not take into account the emotional state of the doctor or patient, limiting the accuracy and adaptability of the diagnosis. Conventional technologies do not incorporate emotional data when collecting feedback or retraining AI models, resulting in inflexible behavior adjustments and making it difficult to contribute to improving the quality of medical care.

[1908] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for recognizing the emotional state of the user and using it for analysis, a means for adjusting the diagnosis presentation method based on the emotional state of the medical professional, and a means for collecting the final diagnosis result and feedback and using them as training data for the generative AI model. This enables diagnostic support that takes the emotional state into consideration, improving the accuracy of diagnosis and the quality of medical care.

[1909] 1. A "generative artificial intelligence model" is a system that uses machine learning algorithms to generate specific outputs from input data.

[1910] 2. "Patient-reported symptoms" refers to the patient's health condition or physical abnormalities reported to a medical professional.

[1911] 3. "Medical Image Data" means image data used for medical diagnosis or treatment, such as CT scans, MRIs, and X-ray images.

[1912] 4. "Test results" means data based on medical tests such as blood tests, urine tests, and tissue tests.

[1913] 5. "Diagnosis candidate" refers to a diagnosis result that the generative artificial intelligence model may present based on the analysis results.

[1914] 6. "Analysis results" refers to the conclusions or insights obtained by analyzing the input data.

[1915] 7. "Healthcare professionals" are doctors, nurses, clinical laboratory technicians, and other professionals who diagnose, treat, and care for patients.

[1916] 8. "User emotional state" means the mental or psychological state of a healthcare professional or patient.

[1917] 9. "Adjusting the diagnosis presentation method based on emotional state" refers to a method of adaptively changing the display and presentation of diagnosis results depending on the user's emotional state.

[1918] 10. "Final diagnosis" means the patient's final diagnosis determined by a medical professional.

[1919] 11. "Feedback" refers to information such as evaluations and opinions of the system and medical professionals' reactions to diagnostic results.

[1920] 12. “Training Data” means a data set used to train a machine learning model.

[1921] MODE FOR CARRYING OUT THE INVENTION

[1922] The present invention relates to a medical diagnosis support system, and in particular to a system that uses a generative artificial intelligence model and an emotion engine to analyze a patient's symptoms, medical image data, and test results, and presents diagnostic candidates while also taking into account the user's emotional state. The following describes in detail an embodiment of the present invention.

[1923] System configuration

[1924] The system of the present invention includes the following major components:

[1925] 1. Terminal: A device where the user (doctor) inputs patient information and receives analysis results and emotions. Specifically, input devices such as a PC or tablet can be used.

[1926] 2. Server: This is the central system that collects and analyzes data, uses generative AI models to present diagnostic candidates, and analyzes user emotions. This can be implemented on a cloud server or an on-premise server.

[1927] 3. Database: A storage system for storing patient data, analysis results, feedback, and emotion data. For example, an SQL database or a NoSQL database can be used.

[1928] 4. Emotion Engine: An algorithm for recognizing and analyzing the emotional state of users (doctors) and patients. This engine analyzes emotions using natural language processing and image recognition technology.

[1929] Example of a system

[1930] The present invention will be described in detail below based on a specific example of the operation of the system.

[1931] Entering patient information

[1932] The user (doctor) uses the terminal to input basic patient information, symptoms, medical image data, and test results. For example, for a patient complaining of abdominal pain, basic information, CT scan images, and blood test results are input. The emotion engine also detects the user's (doctor's) emotional state in real time and acquires data.

[1933] Sending data

[1934] The device sends the collected patient information and emotion data to the server using an HTTP request.

[1935] Data storage and preprocessing

[1936] The server stores the received data in a database, after which pre-processing of the data is performed, which includes data integrity checks, cleaning, and format conversion.

[1937] AI model for generating diagnostic candidates

[1938] The preprocessed data is input into a generative AI model on the server, which generates candidate diagnoses based on the patient's symptoms. At the same time, emotional data obtained from the emotion engine is also included in the analysis. For example, the generative AI model references past case data to generate candidate diagnoses such as "acute appendicitis," "intestinal obstruction," and "gastritis" for a patient complaining of abdominal pain.

[1939] Adjusting the display and presentation of results

[1940] The generated diagnosis candidates are sent from the server to the terminal. The terminal displays the diagnosis candidates in a format that is easy for the user (doctor) to understand. The emotion engine also adaptively adjusts the display method of the diagnosis candidates based on the emotion data acquired. For example, for a user with high stress, the diagnosis candidates are displayed in different colors to make them easier to understand visually.

[1941] Enter and submit feedback

[1942] The user (doctor) inputs the final diagnosis and feedback using a terminal and sends it to the server. This feedback data is used to retrain the generative AI model.

[1943] Prompt Sentence Examples

[1944] "Please list possible diagnoses based on the CT scan images and blood test results of a patient who visits the hospital with abdominal pain. Please suggest diagnoses based on medical evidence. Also, please adjust the display method according to the user's (doctor's) emotional state."

[1945] This invention allows doctors to quickly and accurately obtain supplementary information, reducing the risk of misdiagnosis. Furthermore, the use of an emotion engine enables diagnostic support that takes the user's mental state into account, which is expected to improve diagnostic accuracy and the quality of medical care. It is also expected that the accuracy of the system will continue to improve through collected feedback.

[1946] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1947] Step 1: Enter patient information

[1948] The terminal receives input from the user (doctor) of the patient's basic information, symptoms (e.g., abdominal pain), medical image data, and test results. The input includes the patient's basic information, CT scan images, blood test results, etc. Based on this, a basic information dataset is generated, and the emotion engine detects the user's (doctor's) emotional state (e.g., stress, fatigue) in real time and acquires it as additional data.

[1949] Specific behavior:

[1950] The user (doctor) enters the patient's name, age, gender, symptoms, CT scan images, and blood test results into the device's input form and presses the send button. The emotion engine analyzes the user's facial expressions and tone of voice and records their emotional state.

[1951] Step 2: Sending data

[1952] The terminal sends the collected patient information and emotion data to the server. As input, the reference dataset is sent by the terminal to the server. The data is sent using the HTTP request format.

[1953] Specific behavior:

[1954] The device collects all patient information and emotion data, structures it in JSON format, and sends it to the server using an HTTP POST request.

[1955] Step 3: Storing and Preprocessing Data

[1956] The server stores the received data in a database. The input includes the reference dataset sent from the terminal. Then, the data is preprocessed. The preprocessing includes data integrity check, cleaning, and format conversion. The output is a preprocessed clean dataset.

[1957] Specific behavior:

[1958] The server structures the incoming data and stores it in the database using SQL insert queries. After storing it, it detects missing values ​​in the data, imputes them appropriately, and then converts all data into a unified format.

[1959] Step 4: Generate candidate diagnoses using the AI ​​model

[1960] The server inputs the preprocessed data into a generative AI model to generate candidate diagnoses. The input includes a clean, preprocessed dataset. During the analysis, the AI ​​model references past case data and medical knowledge. The output is multiple candidate diagnoses.

[1961] Specific behavior:

[1962] The server runs a generative AI model using a deep learning framework such as TensorFlow or PyTorch, inputs the preprocessed data into the model, and the model analyzes the data to generate candidate diagnoses such as "acute appendicitis," "intestinal obstruction," and "gastritis."

[1963] Step 5: Adjust the display and presentation of results

[1964] The server organizes the generated candidate diagnoses and sends them to the terminal. The input includes the generated candidate diagnoses. The terminal displays the candidate diagnoses and adjusts the display method based on the emotional state of the user (doctor). The output is the adjusted display of the candidate diagnoses.

[1965] Specific behavior:

[1966] The server sends the generated diagnosis candidates in JSON format to the terminal as an HTTP response. The terminal receives this and displays the diagnosis candidates through a GUI. Based on the results of the emotion engine analysis, if the doctor is in a state of high stress, the diagnosis candidates are color-coded to make them easier to understand visually.

[1967] Step 6: Enter and submit your feedback

[1968] The terminal allows the user (doctor) to input the final diagnosis and feedback, and sends the data to the server. The input includes the final diagnosis and feedback. The output is the feedback data sent to the server.

[1969] Specific behavior:

[1970] The user (doctor) enters the final diagnosis (e.g., acute appendicitis) and feedback about the system's diagnostic accuracy into the input form on the terminal, and presses the submit button. The terminal structures this data in JSON format and sends it to the server using an HTTP POST request.

[1971] Step 7: Save feedback and retrain

[1972] The server stores the received feedback data in a database and uses it to retrain the generative AI model. The feedback data is included as input. The output is an updated generative AI model.

[1973] Specific behavior:

[1974] The server adds and stores the feedback data in a database, then retrains the generative AI model using a new dataset that includes the feedback data, e.g., updating model parameters to improve the model's performance.

[1975] (Application example 2)

[1976] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1977] The problem that this invention aims to solve is to improve the accuracy and efficiency of diagnoses by reducing the risk of misdiagnosis caused by information overload, time constraints, and the doctor's own emotional state when making a diagnosis in the medical field. Another problem is to improve the safety and efficiency of work by analyzing the emotional state of security operators in real time and utilizing that information when generating alerts.

[1978] 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.

[1979] In this invention, the server includes means for analyzing patient-reported symptoms, medical image data, and test results using a generative AI model and presenting diagnosis candidates, means for displaying the analysis results to medical professionals, means for collecting final diagnosis results and feedback from medical professionals and using them as training data for the generative AI model, means for collecting and analyzing biometric data in real time, and means for generating and displaying alerts based on the analysis results and biometric data, thereby enabling improved accuracy and efficiency of diagnoses in medical settings and improved safety and efficiency in security operations.

[1980] A "generative artificial intelligence model" refers to an algorithm or machine learning model that analyzes data collected in medical diagnosis or security analysis and generates diagnostic candidates or anomaly detection results.

[1981] "Patient-reported symptoms" refers to information about specific symptoms such as poor health or pain that patients self-report in medical settings.

[1982] "Medical image data" refers to image information of a patient's inside the body obtained using medical equipment such as CT scans, X-ray images, and MRIs.

[1983] "Test results" refers to numerical data and information indicating whether or not there are abnormalities obtained from blood tests, urine tests, various biochemical tests, etc.

[1984] "Analysis results" refers collectively to the diagnosis candidates and anomaly detection results obtained after analyzing input data using a generative artificial intelligence model.

[1985] "Medical professionals" refers to professionals who perform medical-related tasks, such as doctors, nurses, and radiological technologists.

[1986] "Biometric data" refers to information related to the physical and mental state of medical professionals, security operators, etc., such as heart rate, body temperature, and facial expression data.

[1987] An "alert" refers to a warning or caution message issued by the system based on analysis results or abnormal values ​​in biometric data.

[1988] "Database" refers to a system that organizes and stores large amounts of information, such as reported symptoms, medical image data, test results, and analysis results, and enables quick search and retrieval as needed.

[1989] "Preprocessing" refers to the process of checking the integrity of collected data, cleaning it, and converting its format to make it suitable for subsequent analysis.

[1990] "Feedback" refers to the evaluations and opinions of medical professionals and security operators regarding the final diagnosis results and system proposals.

[1991] The embodiments for carrying out the present invention will be described below.

[1992] System Overview

[1993] The system of the present invention includes the following major components:

[1994] 1. Terminal: A device used by the user (doctor or security operator) to input and display basic information, biometric data, analysis results, and emotional state of the patient.

[1995] 2. Server: A central system that collects and analyzes data, generates diagnostic candidates and anomaly detection results using generative AI models, integrates the analysis results with biometric data, and generates appropriate alerts.

[1996] 3. Database: A storage system for storing reported symptoms, medical image data, test results, analysis results, feedback, emotional data, etc.

[1997] 4. Emotion engine: An algorithm that recognizes and analyzes the emotional state of the user (doctor or security operator).

[1998] Data collection

[1999] The terminal collects basic information about the patient and operator, as well as biometric data (heart rate, body temperature, facial expression data) from the user, and also collects video data and medical image data from security cameras and medical equipment.

[2000] Data analysis

[2001] The server stores the received data in a database and performs preprocessing. This involves checking the data for consistency, cleaning it, and converting its format. Once this is complete, the data is input into the generative AI model for analysis. At the same time, the emotion data obtained by the emotion engine is used for analysis.

[2002] Alert Generation

[2003] Based on the analysis results and emotion data, the server generates appropriate alerts. For example, if an abnormality is detected and an operator is in a high stress state at the same time, an alert is displayed indicating that immediate action is required. The alert is sent to the terminal and displayed to the user.

[2004] Feedback collection

[2005] Feedback on analysis results and alerts is collected from users. This feedback is sent to the server and stored in a database. The collected feedback is used as retraining data for the generative AI model and is used to improve the accuracy of the system.

[2006] Specific examples

[2007] When a doctor uses a device to enter basic patient information, CT scan images, and blood test results, the data is sent to a server. The server preprocesses the data and analyzes it using a generative AI model. As a result, a candidate diagnosis is generated and displayed on the device. At the same time, an emotion engine analyzes the doctor's emotional state and adjusts the way the candidate diagnosis is presented based on that data.

[2008] If a security operator wears smart glasses during the night shift and the system simultaneously detects an abnormality and high stress, an alert message will be displayed stating, "High stress and abnormality detected! Immediate action required!"

[2009] Example prompts to input to the generative AI model

[2010] "If an anomaly is detected, how can we tailor our response to reduce operator burden?"

[2011] "A method for detecting operator stress levels in real time and proposing countermeasures during security operations."

[2012] In this way, the present invention enables efficient information presentation and alert generation that integrates analysis results and emotional data in medical settings and security operations.

[2013] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2014] Step 1:

[2015] Data Collection Phase

[2016] Input: Basic information from the user (doctor or security operator), biometric data (heart rate, body temperature, facial expression data), video data and medical image data from security cameras and medical equipment.

[2017] Output: Sends collected data to a server.

[2018] Specific behavior:

[2019] The user uses the terminal to input basic patient information and biometric data, and data is collected in real time from security cameras and medical devices, which are then transmitted to a server.

[2020] Step 2:

[2021] Database save phase

[2022] Input: Basic information, biometric data, video data, and medical image data sent to the server.

[2023] Output: Various data stored in the database.

[2024] Specific behavior:

[2025] The server stores the received data in a database, where it checks for consistency.

[2026] Step 3:

[2027] Preprocessing Phase

[2028] Input: Basic information, biometric data, video data, and medical image data stored in the database.

[2029] Output: The preprocessed data.

[2030] Specific behavior:

[2031] The server retrieves data from the database, checks its integrity, cleans it, and converts it into a format suitable for generative AI models.

[2032] Step 4:

[2033] Data analysis phase

[2034] Input: Preprocessed basic information, biometric data, video data, and medical image data.

[2035] Output: Diagnosis candidates, anomaly detection results, and sentiment analysis results from the generative AI model.

[2036] Specific behavior:

[2037] The server inputs the preprocessed data into the generative AI model for analysis, while simultaneously analyzing and integrating the user's emotional data obtained by the emotion engine.

[2038] Step 5:

[2039] Alert Generation Phase

[2040] Input: Diagnosis candidates, anomaly detection results, and sentiment analysis results.

[2041] Output: The alert message.

[2042] Specific behavior:

[2043] The server generates appropriate alerts based on the analysis results and biometric data. For example, if an abnormality is detected and an operator is in a high stress state, an alert message requiring immediate action is generated. The generated alert is then sent to the terminal.

[2044] Step 6:

[2045] Display Phase

[2046] Input: Alert message, diagnosis candidates, and sentiment analysis results sent from the server.

[2047] Output: Alert message and potential diagnosis displayed on the terminal.

[2048] Specific behavior:

[2049] The device receives alert messages and diagnostic candidates from the server and displays them to the user, adjusting the display format according to the results of emotion analysis.

[2050] Step 7:

[2051] Feedback gathering phase

[2052] Input: Final diagnostic results and feedback from the user.

[2053] Output: Feedback sent to the server.

[2054] Specific behavior:

[2055] The user (doctor or security operator) inputs feedback on the final diagnosis and the evaluation of the alert through the terminal, and this data is sent to the server and stored in the database.

[2056] Step 8:

[2057] Model retraining phase

[2058] Input: Stored feedback, newly collected data.

[2059] Output: A generative artificial intelligence model with improved accuracy.

[2060] Specific behavior:

[2061] The server retrains the generative AI model based on collected feedback and newly acquired data, thereby continuously improving diagnostic accuracy and anomaly detection performance.

[2062] This concludes the explanation of each processing step of this program and its specific operations.

[2063] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[2064] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2065] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[2066] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2067] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[2068] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[2069] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[2070] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, motorcycles, and other devices, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[2071] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[2072] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[2073] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[2074] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[2075] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[2076] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[2077] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[2078] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[2079] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[2080] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[2081] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[2082] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[2083] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[2084] The following is further disclosed regarding the above embodiment.

[2085] (Claim 1)

[2086] A means for analyzing patient-reported symptoms, medical image data, and test results using a generative artificial intelligence model and presenting candidate diagnoses;

[2087] a means for displaying the analysis results to a medical professional;

[2088] a means for collecting the final diagnosis results and feedback from medical professionals and using them as training data for a generative artificial intelligence model;

[2089] A system including:

[2090] (Claim 2)

[2091] 10. The system according to claim 1, further comprising means for storing the reported symptoms, medical image data and test results in a database.

[2092] (Claim 3)

[2093] 10. The system of claim 1, further comprising means for pre-processing the collected data.

[2094] "Example 1"

[2095] (Claim 1)

[2096] A means for analyzing patient-reported symptoms, medical image data, and test results using a generative artificial intelligence model and presenting candidate diagnoses;

[2097] means for transmitting the input patient information to a server by the terminal;

[2098] a means for the server to store the received data in a database and perform pre-processing of the data;

[2099] a means for displaying the analysis results to a medical professional;

[2100] a means for collecting the final diagnosis results and feedback from medical professionals and using them as training data for a generative artificial intelligence model;

[2101] A system including:

[2102] (Claim 2)

[2103] 10. The system according to claim 1, further comprising means for storing the reported symptoms, medical image data and test results in a database.

[2104] (Claim 3)

[2105] 10. The system of claim 1, further comprising means for pre-processing the collected data.

[2106] "Application Example 1"

[2107] (Claim 1)

[2108] A means for analyzing patient-reported symptoms, medical image data, and test results using a generative artificial intelligence model and presenting candidate diagnoses;

[2109] A means for displaying the analysis results to users such as medical professionals or patients;

[2110] A means for collecting the final diagnosis results and feedback from users and using them as training data for a generative artificial intelligence model;

[2111] A means for using the medical diagnostic support system in physical stores such as drugstores, and

[2112] A system including:

[2113] (Claim 2)

[2114] 10. The system according to claim 1, further comprising means for storing the reported symptoms, medical image data and test results in a database.

[2115] (Claim 3)

[2116] 10. The system of claim 1, further comprising means for pre-processing the collected data.

[2117] "Example 2: Combining Emotion Engines"

[2118] (Claim 1)

[2119] A means for analyzing patient-reported symptoms, medical image data, and test results using a generative artificial intelligence model and presenting candidate diagnoses;

[2120] a means for displaying the analysis results to a medical professional;

[2121] means for recognizing and analyzing the emotional state o...

Claims

1. A means for analyzing patient-reported symptoms, medical image data, and test results using a generative artificial intelligence model and presenting candidate diagnoses; a means for displaying the analysis results to a medical professional; a means for collecting the final diagnosis results and feedback from medical professionals and using them as training data for a generative artificial intelligence model; A system including:

2. The system according to claim 1 , further comprising means for storing the reported symptoms, medical image data and test results in a database.

3. The system of claim 1 further comprising means for pre-processing the collected data.

Images