Information processing device, information processing method, and program

Abstract

The goal is to provide information that addresses any lack of understanding or acceptance of the patient's treatment. [Solution] The information processing device of the embodiment includes a question acquisition unit, a prompt generation unit, an answer candidate acquisition unit, a filtering unit, and an answer information generation unit. The question acquisition unit acquires questions related to the patient's medical information. The prompt generation unit generates prompts by adding predetermined information to the questions acquired by the question acquisition unit. The answer candidate acquisition unit uses the prompts generated by the prompt generation unit to acquire answer candidates for the questions from a large-scale medical information processing model. The filtering unit filters the answer candidates acquired by the answer candidate acquisition unit based on information related to the patient's medical treatment. The answer information generation unit generates answer information based on the filtering results of the filtering unit.

Description

Technical Field

[0005]

[0001] Embodiments of the present invention relate to an information processing apparatus, an information processing method, and a program.

Background Art

[0002] In conventional examination and follow-up consultations, the final treatment policy is determined while the doctor and the patient interact based on the diagnosis results and the like. In recent years, in order to improve the efficiency of diagnosis and the like, technologies have been known that use AI (Artificial Intelligence) models or language models to generate inquiry information, convert questions from patients with little medical knowledge into medical questions and ask the doctor, or provide information according to the mental state of the patient such as a sense of uneasiness.

[0003] However, in actual examinations and follow-up consultations, since the time and timing for communication between the doctor and the patient are limited, there is a possibility that the doctor may not be able to ask sufficient questions, resulting in insufficient understanding and acceptance by the patient.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Patent Document 3

Summary of the Invention

Problems to be Solved by the Invention

[0005] One of the problems that the present invention aims to solve is to provide information that will resolve the lack of understanding or acceptance of patients' treatment. However, the problems that the embodiments disclosed herein and in the drawings aim to solve are not limited to the above problem, and problems corresponding to the effects of each configuration shown in the embodiments described later can also be positioned as other problems. [Means for solving the problem]

[0006] The information processing device of this embodiment includes a question acquisition unit, a prompt generation unit, an answer candidate acquisition unit, a filtering unit, and an answer information generation unit. The question acquisition unit acquires questions related to the patient's medical information. The prompt generation unit generates prompts by adding predetermined information to the questions acquired by the question acquisition unit. The answer candidate acquisition unit uses the prompts generated by the prompt generation unit to acquire answer candidates for the questions from a large-scale medical information processing model. The filtering unit filters the answer candidates acquired by the answer candidate acquisition unit based on information related to the patient's medical treatment. The answer information generation unit generates answer information based on the filtering results of the filtering unit. [Brief explanation of the drawing]

[0007] [Figure 1] A diagram showing an example of the configuration of an information processing system 1 including an information processing device of an embodiment. [Figure 2] A diagram showing an example of the functional configuration of terminal device 100. [Figure 3] A diagram showing an example of the functional configuration of the information processing device 200. [Figure 4] A diagram showing an example of the functional configuration of the large-scale medical information processing server 300. [Figure 5] A diagram illustrating the process for determining a treatment plan using the information processing device 200 of this embodiment. [Figure 6] A diagram illustrating the generation of response information. [Figure 7] A flowchart showing a series of processes performed by the processing circuit 240 of the embodiment. [Figure 8] A diagram showing a modified example of the information processing device of the embodiment. [Modes for carrying out the invention]

[0008] The information processing apparatus, information processing method, and program of the embodiment will be described below with reference to the drawings.

[0009] Figure 1 shows an example of the configuration of an information processing system 1 including an information processing device of an embodiment. The information processing system 1 includes, for example, an inspection device 10, a medical information database (DB) 20, a terminal device 100, an information processing device 200, and a large-scale medical information processing server 300. The inspection device 10, the medical information database 20, the terminal device 100, the information processing device 200, and the large-scale medical information processing server 300 are connected to each other via, for example, a network NW. The network NW refers to an information communication network in general that utilizes telecommunications technology. The network NW includes, for example, wireless / wired LANs (Local Area Networks), WANs (Wide Area Networks), the Internet, telephone communication lines, Wi-Fi networks, Bluetooth®, optical fiber communication networks, cable communication networks, satellite communication networks, etc. Note that the information processing system 1 may have multiple instances of at least one of the inspection device 10, medical information database 20, inspection device 10, terminal device 100, information processing device 200, and large-scale medical information processing server 300. Furthermore, the information processing system 1 may be configured without the inspection device 10.

[0010] The examination device 10 is a device that examines various parts (examination sites) of a user, such as a patient. The examination device 10 includes, for example, a medical image generation device. Medical image generation devices include, for example, X-ray computed tomography scanners, X-ray diagnostic devices, magnetic resonance imaging devices, ultrasound diagnostic devices, nuclear medicine diagnostic devices, etc. The examination device 10 generates examination results (e.g., medical images, etc.) related to the patient based on the operation of an operator, such as a physician. The examination device 10 transmits the examination results to the medical information DB 20 via a network NW. The examination device 10 may also include various other examination devices such as blood test devices and blood pressure monitors.

[0011] The medical information DB 20 is a database that manages medical information for each patient, for example, by acquiring and storing test results from the testing device 10, and storing patient diagnosis results (examination results) registered by doctors, etc. The medical information DB 20 may also include hospital equipment information, etc. The medical information DB 20 may be, for example, a Hospital Information System (HIS), a Picture Archiving and Communication System (PACS), a Radiology Information System (RIS), or a system that integrates at least a part of these. For example, the medical information DB 20 has a function to extract medical information (for example, test results or doctor's diagnosis results) corresponding to patient information (for example, patient identification information) from the stored data in response to a query for medical information from an external device, and transmit it to the external device that made the query.

[0012] The terminal device 100 is, for example, a personal computer (PC), a tablet device, or a smartphone. The terminal device 100 is used, for example, by the patient. However, the terminal device 100 may also be used by the patient's family or other supporters, or by healthcare professionals. Healthcare professionals include, for example, doctors, but may also include nurses or other individuals involved in medical treatment. In the following explanation, the terminal device 100 will be described as being used by the patient.

[0013] The information processing device 200 may be, for example, a PC or server device, a workstation, a web server, or a cloud server. The information processing device 200 generates response information corresponding to the input information from the patient entered by the terminal device 100, and outputs the generated response information to the terminal device 100. The input information includes, for example, question information regarding the patient's medical information and additional question information regarding the response information. For example, the information processing device 200 generates a prompt that includes instructions to obtain information regarding the response information to the question information (for example, candidate answers, proposed answers) from a Large Language Model (LLM) managed by the Large-Scale Medical Information Processing Server 300, sends it to the Large-Scale Medical Information Processing Server 300 via the network NW to query it, and obtains information regarding the response information from the Large-Scale Medical Information Processing Server 300. The response information output to the terminal device 100 by the information processing device 200 may be sent to the medical information DB 20 and managed in the medical information DB 20.

[0014] The large-scale medical information processing server 300 may be, for example, a PC or a server device, or may be a workstation, a web server, or a cloud server. The large-scale medical information processing server 300 manages a large-scale medical information processing model, receives a prompt transmitted from the information processing device 200 via the network NW, generates an answer candidate (answer) corresponding to the received prompt, and transmits it to the information processing device 200. The large-scale medical information processing model is mainly a large language model (LLM) specialized for outputting medical and medical-related information. The large language model is, for example, a model that autonomously generates data such as text, images, and voices for input data. The large language model may be, for example, an interactive generation AI model specialized for enabling natural conversations with users by natural language processing (NLP; Natural Language Processing), etc., or may be a multimodal AI (Multimodal Artificial Intelligence) that handles different types of information together, or may be a combination of these. The large-scale medical information processing model may be a model learned in advance by teacher data, etc., or may be a model that is appropriately updated using the diagnosis results of actual doctors and the determined treatment policies.

[0015] Next, the functional configurations of the terminal device 100, the information processing device 200, and the large-scale medical information processing server 300 will be specifically described.

[0016] [Terminal Device] FIG. 2 is a diagram showing an example of the functional configuration of the terminal device 100. The terminal device 100 includes, for example, a communication unit 110, a touch panel 120, a microphone 130, a speaker 140, a processing unit 150, and a memory (storage circuit) 160. The communication unit 110 communicates with external devices such as the information processing device 200 via the network NW, for example. The communication unit 110 includes, for example, a communication interface such as a NIC (Network Interface Card).

[0017] The touch panel 120 functions as, for example, an input interface for a user (such as a patient) to input information and an output interface for displaying information such as the information input by the user and the processing result of the processing unit 150. The touch panel 120 displays, for example, a GUI (Graphical User Interface) for the user to perform various input operations. The touch panel 120 outputs information corresponding to the input operation by the user to the processing unit 150.

[0018] The microphone 130 is a microphone for voice input. The microphone 130 generates input voice information corresponding to the user's utterance and outputs the generated voice information to the processing unit 150. The speaker 140 is a speaker for sound output. The speaker 140 outputs a voice or the like corresponding to the output voice information generated by the processing unit 150.

[0019] The processing unit 150 is realized, for example, by a hardware processor such as a CPU (Central Processing Unit) executing a program (software) such as a dedicated application for performing questions and the like in the present embodiment. For example, the processing unit 150 acquires a question item from the user corresponding to the user's operation input by the touch panel 120 or the microphone 130 and transmits it to the information processing apparatus 200 via the communication unit 110. The processing unit 150 causes an image or the like included in the answer information transmitted by the information processing apparatus 200 and received by the communication unit 110 to be displayed on the touch panel 120 or outputs a voice to the speaker 140. Note that, in addition to the dedicated application, the processing unit 150 may access a Web page provided by the information processing apparatus 200 to input a question and acquire answer information from the information processing apparatus 200.

[0020] Memory 160 can be implemented by, for example, semiconductor memory elements such as RAM (Random Access Memory) or flash memory, a hard disk, or an optical disc. These non-transient storage media may also be implemented by other storage devices connected via a network NW, such as a NAS (Network Attached Storage) or an external storage server device. Alternatively, these non-transient storage media may also be implemented by storage devices such as ROM (Read Only Memory) or registers. Memory 160 stores various types of information, such as information input via the touch panel 120 or microphone 130, information acquired via the network NW, programs, and other information.

[0021] [Information Processing Device] Figure 3 shows an example of the functional configuration of the information processing device 200. The information processing device 200 includes, for example, a communication interface 210, an input interface 220, a display 230, a processing circuit 240, and a memory 250.

[0022] The communication interface 210 includes, for example, a communication interface such as a NIC or an antenna for wireless communication. The communication interface 210 communicates with terminal devices 100, a large-scale medical information processing server 300, a medical information database 20, and other external devices via the network NW, and outputs the acquired information to the processing circuit 240, memory 250, etc. The communication interface 210 also transmits information to terminal devices 100, a medical information database 20, and other external devices via the network NW under the control of the processing circuit 240.

[0023] The input interface 220 receives various input operations from the user of the information processing device 200 (in this case, an administrator, etc.), converts the received input operations into electrical signals, and outputs them to the processing circuit 240. For example, when an input operation is performed by a user, the input interface 220 generates information corresponding to the input operation. The input interface 220 outputs the generated information corresponding to the input operation to the processing circuit 240.

[0024] The input interface 220 includes, for example, a mouse, keyboard, trackball, switch, button, joystick, touch panel, etc. The input interface 220 may also be a user interface that accepts audio input, such as a microphone. If the input interface 220 is a touch panel, it may also incorporate the display function of the display 230.

[0025] In this specification, the input interface 220 is not limited to those equipped with physical operating components such as a mouse or keyboard. For example, an electrical signal processing circuit that receives an electrical signal corresponding to an input operation from an external input device provided separately from the device and outputs this electrical signal to a control circuit is also included as an example of the input interface 220.

[0026] The display 230 displays various types of information. For example, the display 230 displays images generated by the processing circuit 240, or a GUI for receiving various input operations from the user. For example, the display 230 may be an LCD (Liquid Crystal Display), a CRT (Cathode Ray Tube) display, or an organic EL (Electro Luminescence) display. The display 230 is an example of a "display unit".

[0027] The processing circuit 240 includes, for example, a question acquisition function 241, a prompt generation function 242, an inquiry function 243, a filtering function 244, an answer information generation function 245, and an output control function 246. The processing circuit 240 implements these functions, for example, by a hardware processor (computer) executing a program stored in the memory (storage circuit) 250.

[0028] Hardware processors refer to circuits such as CPUs (Central Processing Units), GPUs (Graphics Processing Units), Application Specific Integrated Circuits (ASICs), and programmable logic devices (e.g., Simple Programmable Logic Devices (SPLDs), Complex Programmable Logic Devices (CPLDs), and Field Programmable Gate Arrays (FPGAs)).

[0029] Instead of storing the program in memory 250, the system may be configured to directly incorporate the program into the hardware processor's circuitry. In this case, the hardware processor performs its functions by reading and executing the program incorporated into the circuitry. The program may be stored in a storage device beforehand, or it may be stored in a non-temporary storage medium such as a DVD or CD-ROM, and installed from the non-temporary storage medium to the storage device when the non-temporary storage medium is mounted in the drive device (not shown) of the information processing device 200.

[0030] A hardware processor is not limited to being a single circuit; it may also be composed of multiple independent circuits combined to perform various functions. Alternatively, multiple components may be integrated into a single hardware processor to perform various functions.

[0031] The memory 250 is implemented by, for example, semiconductor memory elements such as RAM and flash memory, hard disks, optical discs, etc. These non-transient storage media may be implemented by other storage devices connected via a network NW, such as NAS or external storage server devices. Alternatively, these non-transient storage media may be implemented by storage devices such as ROM and registers. The memory 250 stores, for example, patient information 251 acquired from the terminal device 100 and the medical information DB 20, processing data 252 showing the results processed by the processing circuit 240, programs, and various other information. Patient information 251 is, for example, information that associates patient identification information with the patient's personal information, information for communicating with the patient's terminal device 100, and information such as questions from the patient. Processing data 252 is, for example, data (information) that associates, for each patient, with answer information to questions, information acquired from the medical information DB 20, and information on candidate answers acquired from the large-scale medical information processing server 300. The patient information 251 and processing data 252 are updated or deleted as appropriate.

[0032] The question acquisition function 241 acquires input information from the terminal device 100 via the communication interface 210. This input information includes, for example, questions regarding the patient's medical information and additional questions regarding the answers. The question acquisition function 241 also acquires information on candidate answers to the questions from the information processing device 200 via the communication interface 210.

[0033] The prompt generation function 242 generates a prompt by adding predetermined information (for example, instruction information that instructs the output of answer candidates) to the input information obtained by the question acquisition function 241.

[0034] The inquiry function 243 uses the prompt generated by the prompt generation function 242 to query the large-scale medical information processing server 300 and obtain candidate answers (suggested answers) corresponding to the prompt (question). The inquiry function 243 is an example of a "candidate answer acquisition function".

[0035] The filtering function 244 narrows down the candidate answers obtained by the query function 243 based on information related to the patient's medical treatment. Information related to the patient's medical treatment includes, for example, at least one of the following: information on the patient's diagnosis, information on the doctor's treatment plan (draft treatment plan), and hospital equipment information. Information related to the patient's medical treatment can be obtained, for example, by querying the medical information DB 20 using patient identification information. A detailed explanation of the filtering function 244 will be given later.

[0036] The response information generation function 245 generates response information to be provided to the patient using the response candidates narrowed down by the filtering function 244. The response information may be text information, image information (still images or videos), audio information, or a combination of these. The content of the response information may be determined according to the type of response, or it may be set according to instructions from the terminal device 100 or the type of terminal device 100. For example, if the terminal device 100 is a smartphone, the output control function 246 generates an image showing the response information to be displayed on the touch panel 120, or outputs audio from the speaker 140.

[0037] The output control function 246 generates output information that includes the response information generated by the response information generation function 245. The output information may include, for example, information regarding the number of responses and information regarding the output format at the destination. The output control function 246 transmits the generated output information to the terminal device 100 via the communication interface 210 or displays it on the display 230.

[0038] [Large-scale medical information processing server] Figure 4 shows an example of the functional configuration of the large-scale medical information processing server 300. The large-scale medical information processing server 300 includes, for example, a communication interface 310, a processing circuit 320, and a memory 330.

[0039] The communication interface 310 includes, for example, a communication interface such as a NIC or an antenna for wireless communication. The communication interface 310 communicates with the information processing device 200, the medical information DB 20, and other external devices via the network NW, and outputs the acquired information to the processing circuit 320. The communication interface 310 also transmits information to the information processing device 200, the medical information DB 20, and other external devices via the network NW under the control of the processing circuit 320.

[0040] The processing circuit 320 includes, for example, an acquisition function 321, a response candidate generation function 322, and a provision function 323. The processing circuit 320 realizes these functions, for example, by having a hardware processor (computer) execute a program stored in memory (storage circuit) 250. Alternatively, instead of storing the program in memory 250, the program may be directly incorporated into the hardware processor's circuit. In this case, the hardware processor realizes the functions by reading and executing the program incorporated into the circuit. The above program may be stored in a storage device in advance, or it may be stored in a non-temporary storage medium such as a DVD or CD-ROM, and the non-temporary storage medium may be installed from the non-temporary storage medium to the storage device when it is mounted in a drive device (not shown) of the large-scale medical information processing server 300.

[0041] The memory 330 is implemented by, for example, semiconductor memory elements such as RAM and flash memory, hard disks, optical discs, etc. These non-transient storage media may also be implemented by other storage devices connected via a network NW, such as NAS or external storage server devices. Alternatively, these non-transient storage media may also be implemented by storage devices such as ROM and registers. The memory 330 stores, for example, processing data, programs, and various other information that show the results processed by the large-scale medical information processing model 331 and the processing circuit 320. The large-scale medical information processing model 331 outputs information on candidate answers (suggested answers) in response to prompt input.

[0042] The acquisition function 321 acquires prompts, which are patient questions and predefined information, transmitted from the information processing device 200 via the communication interface 310.

[0043] The answer candidate generation function 322 inputs the prompt acquired by the acquisition function 321 into the large-scale medical information processing model 331 to obtain answers to the questions and other items included in the prompt, and generates answer candidates (draft answers) based on the acquired answers. There may be multiple answer candidates, in which case the prompt may be instructed in advance to obtain multiple answers.

[0044] The provisioning function 323 generates provisioning information to be provided to the information processing device 200 based on the answer candidates generated by the answer candidate generation function 322. The provisioning information may be text information, image information, audio information, or a combination thereof, corresponding to the answer candidates. The provisioning function 323 transmits the generated provisioning information to the information processing device 200 via the communication interface 310.

[0045] In addition to the configuration shown in Figure 4, the large-scale medical information processing server 300 may also be equipped with input interfaces and displays, similar to the information processing device 200.

[0046] [Examples] Next, we will describe an example of using the information processing system 1 described above to provide information that helps resolve any lack of understanding or acceptance of the patient's treatment, thereby supporting communication between doctors and patients. Figure 5 is a diagram illustrating the flow of determining a treatment plan using the information processing device 200 of the embodiment.

[0047] In the example shown in Figure 5, user (patient) U experiences shoulder pain during sports and goes to a hospital, where they visit a sports medicine clinic. There, medical professionals such as physician D conduct a brief medical interview with the patient and, based on the interview results, perform examinations using the examination device 10 (such as a CT (Computed Tomography) scan). The examination device 10 transmits the examination results to the medical information database 20. At this time, the examination device 10 may also transmit other information about the patient (such as the interview results) to the medical information database 20. The medical information database 20 manages the transmitted information by associating it with the patient identification information.

[0048] Next, physician D obtains test results and other information from the medical information DB 20 using, for example, a terminal device 100D for physician D, diagnoses user U's symptoms based on the obtained test results, consultation results, interview results, etc., and registers the diagnosis result in the medical information DB 20. The diagnosis result includes information such as the name of the disease (e.g., shoulder labral tear), the content of the disease, proposed treatment plan (treatment plan options), and prognosis for each treatment plan. The diagnosis result may also include information about hospital facilities included in the information related to the patient's medical treatment. Proposed treatment plans may include, for example, drug treatment or surgical treatment, and may also include treatment with different drugs. User U may obtain the above-mentioned diagnosis result in person with physician D, or may obtain it from the medical information DB 20 using a terminal device 100U for user U.

[0049] Next, user U accesses the information processing device 200 using terminal device 100U and asks questions to the information processing device 200 regarding the diagnostic results. The information processing device 200 generates prompts based on the questions obtained from user U's terminal device 100 (questions and uncertainties regarding the content of the disease name, treatment options, prognosis for each treatment option, etc.) and queries the large-scale medical information processing server 300.

[0050] The large-scale medical information processing server 300 generates answer candidates using the large-scale medical information processing model 331 based on prompts received from the information processing device 200. At this time, the large-scale medical information processing server 300 may refer to the medical information DB 20 to obtain information such as the user U's test results and the doctor's diagnosis, and generate answer candidates based on the obtained information. The large-scale medical information processing server 300 sends the provided information, including the generated answer candidates, to the information processing device 200.

[0051] The information processing device 200 narrows down the answer candidates based on the information (answer candidates) provided by the large-scale medical information processing server 300, using the diagnosis results of physician D obtained from the medical information DB 20, generates answer information, and transmits it to the terminal device 100U. User U can obtain answers to questions that they could not ask physician D directly. Here, user U may ask further questions regarding the answers obtained from the information processing device 200. This allows user U to more accurately understand the content of the diagnosis results and select a more detailed treatment plan.

[0052] Therefore, in a discussion between user U and physician D, as shown in Figure 5, to determine the treatment plan, user U can communicate a more specific treatment plan to physician D, for example, stating that they would like to undergo conservative treatment first, and then surgery if their symptoms do not improve. This allows for an efficient determination of the treatment plan desired by user U, leading to more appropriate treatment.

[0053] [Generating response information] Next, we will specifically explain the generation of response information through the question-and-answer session between User U and Information Processing System 1 as described above. Figure 6 is a diagram illustrating the generation of response information. The Information Processing Device 200 shown in Figure 5 answers questions from Doctor D regarding the diagnosis results to User U (for example, a patient about to undergo a medical examination to discuss treatment plans, etc.).

[0054] The information processing device 200 receives input from the user (patient) U via the terminal device 100, consisting of at least one question related to medical information (examination / diagnosis information, etc.), and generates a prompt by adding predetermined text to at least one of the received questions.

[0055] In this embodiment, the response information is intended to respect and support the judgment of physician D. Therefore, the response information is not a second opinion, but rather complements and supplements the communication with physician D. Accordingly, the response information includes, for example, information that supplements the advantages and disadvantages of the options presented by physician D (e.g., surgical treatment or drug treatment), information showing past statistical data (e.g., 5-year survival rate), information that supplements the basis for physician D's judgment, and information that explains the test items, test results, and how to interpret them, and is constrained so as not to include any content that contradicts physician D's judgment. Accordingly, the prompt generation function 242 generates a prompt with constraints (an example of predetermined information) added to the question input by user U so that the output is as described above. In addition, the prompt may also have conditions such as Who, When, Where, What, and How added to it, and furthermore, various conditions such as How (+many, much, long, far, often) may be added.

[0056] Next, the information processing device 200 sends the generated prompt to a large-scale medical information processing server 300, which is different from the information processing device 200, to inquire about candidate answers (draft answers). The large-scale medical information processing server 300 inputs the prompt into a large-scale medical information processing model 331 built internally (for example, a model using a multi-center integrated DB on the cloud side) and obtains candidate answers (draft answers) to the questions from this large-scale medical information processing model 331. In this case, the large-scale medical information processing server 300 obtains information on physician D's diagnosis results (diagnosis content before treatment policy decision) from the medical information DB 20 and generates candidate answers based on the obtained information, taking into account physician D's diagnosis results. For example, if the diagnosis results state that physician D is considering two different treatment policies, the server generates information on the advantages and disadvantages of each as a draft answer.

[0057] Next, the filtering function 244 of the information processing device 200 narrows down the candidate answers to the inquiry based on information related to the patient's medical treatment. For example, if a CT scan has been performed using the examination device 10 but an MRI scan has not, the filtering function 244 will provide an answer such as "Your CT image showed the finding ○○, therefore it was determined to be △△," but will exclude answer candidates that cannot be obtained without an MRI scan. Furthermore, the filtering function 244 may restrict the response to suggest an MRI scan if the hospital does not have the equipment to perform an MRI scan, based on information related to the patient's medical treatment obtained from the medical information DB 20. This makes it possible to exclude information that the doctor has not conveyed to the patient as part of the diagnosis from being included in the answer information, thereby preventing a lack of understanding or acceptance of the patient's treatment.

[0058] Furthermore, when the filtering function 244 narrows down the answer candidates based on Dr. D's treatment plan, it may also filter the answers in a way that suppresses the output of answers that differ from the treatment plan. For example, if it is obtained from information related to the patient's medical treatment that Dr. D is considering two treatment options for the patient—medication and surgery—the filtering function 244 will filter the answer candidates in a way that suppresses the output of treatment plans other than medication and surgery. This can prevent the patient from being provided with information other than Dr. D's diagnosis, which could lead to a further lack of understanding or acceptance of the patient's treatment.

[0059] Furthermore, the filtering function 244 narrows down the response options so that, if Doctor D recommends drug treatment over surgery, the output will at least include responses that do not reject drug treatment. This ensures that responses are provided that align with Doctor D's treatment plan, making it easier to resolve any lack of understanding or acceptance of the patient's treatment.

[0060] Furthermore, the filtering function 244 may narrow down the response candidates according to the priority levels that physician D has pre-registered for multiple treatment plans. For example, if physician D has a higher priority for drug treatment than surgery (i.e., recommends drug treatment), the filtering function 244 may narrow down the response information to output more information about drug treatment, or to suppress the output of information about surgical treatment.

[0061] Furthermore, the filtering function 244 may, instead of (or in addition to) the priorities registered in advance by Doctor D, set the priority of treatment plans from the diagnostic results information included in the patient's medical treatment information, and narrow down the answer candidates according to the set priority. In this case, the filtering function 244 may, for example, set the priority of the first treatment plan to appear among the multiple treatment plans included in Doctor D's diagnostic results text to be higher than the other priorities, or it may be set so that the more often a treatment plan appears in the text, the higher its priority. This allows Doctor D to obtain information about priorities without having to set them, and to provide more appropriate information to the user (patient, etc.) based on the priority.

[0062] Furthermore, the filtering function 244 may narrow down the answer candidates based on the type of disease the patient has or the site of examination. For example, the filtering function 244 may narrow down the answer candidates so that answers other than the type of disease diagnosed by physician D are not output, or so that answers related to sites other than those examined by the patient are not output. This makes it easier to provide information directly related to the physician's diagnosis and the site of examination, thereby making it easier to resolve any lack of understanding or acceptance of the patient's treatment.

[0063] Furthermore, the filtering function 244 may narrow down the answer candidates based on the patient's past medical information or test results. In this case, the filtering function 244 retrieves the medical information or test results corresponding to the patient from the medical information DB 20 and outputs answer candidates related to the retrieved information. This makes it possible to output answers that also take into account the patient's past illnesses and injuries, making it easier to resolve any lack of understanding or acceptance of the patient's treatment. The filtering function 244 may also narrow down the answer candidates based on the timing (freshness) of the patient's medical information or test results. Even if the patient has undergone multiple medical examinations or tests in the past, if the information is old, there is a possibility that answers that the patient does not need will be output. Therefore, the filtering function 244 filters using the content of medical results or test results within a predetermined period from the present (for example, up to about 3 to 5 years ago), and does not include the content of medical results or test results older than the predetermined time in the filtering conditions. This makes it possible to provide answer information that is closer to the patient's recent condition.

[0064] The response information generation function 245 of the information processing device 200 generates response information using the narrowed-down response candidates, and the output control function 246 outputs the generated response information to the terminal device 100U. If there is no answer (cannot create) to a question, the response information generation function 245 generates information indicating that there is no answer (or cannot create) or information prompting the user to change the way the question is asked or to ask Doctor D directly, and outputs the generated information to the terminal device 100U.

[0065] Furthermore, if the information processing device 200 obtains additional questions after acquiring the answer information, it generates prompts for the acquired additional questions, queries the large-scale medical information processing server 300, acquires answer candidates, narrows them down, etc., as described above, to generate answer information, and outputs the generated answer information to the terminal device 100. If user U requests an answer other than the previous answer for the additional questions, the narrowing function 244 may narrow down the answer candidates from those excluded in the previous narrowing. This makes it possible to provide an answer different from the previous answer. However, even in this case, the narrowing function 244 narrows down the answers so as not to output answers that strongly contradict the diagnosis of physician D.

[0066] Furthermore, the output control function 246 may store the pair of questions from user U and the answer information generated by the answer information generation function 245 in the processing data 252, or it may send it to the medical information DB 20 for registration. This allows past questions and answer information to be referenced.

[0067] Furthermore, physician D may use terminal device 100D to refer to information processing device 200 and medical information DB 20 to obtain questions and answers exchanged between the patient and the information processing device 200. This allows physician D to supplement the content of the answers during subsequent face-to-face meetings with the patient, and to provide explanations to resolve any lack of understanding or acceptance from the patient, more efficiently and effectively. Thus, communication between physician D and the patient can be supported more effectively.

[0068] [Processing flow] Next, the processing flow of the processing circuit 240 in the embodiment will be described. Figure 7 is a flowchart showing a series of processes performed by the processing circuit 240 in the embodiment. In the example in Figure 7, the question acquisition function 241 acquires a question (step S100). Next, the prompt generation function 242 generates a prompt based on the question (step S110). Next, the inquiry function 243 queries the large-scale medical processing model for an answer to the generated prompt (step S120) and obtains answer candidates from the large-scale medical processing model (step S130).

[0069] Next, the filtering function 244 narrows down the answer candidates based on information related to the patient's medical treatment (step S140). Next, the answer information generation function 245 generates answer information using the information of the narrowed-down answer candidates (step S150). Next, the output control function 246 outputs the generated answer information to the terminal device 100 and provides it to the patient, etc. (step S160).

[0070] Next, the question acquisition function 241 determines whether or not additional questions have been acquired (step S170). If it is determined that additional questions have been acquired, the process returns to step S110, and the prompt generation function 242 generates a prompt for those questions and proceeds with the subsequent processing. If it is determined that no additional questions have been acquired, the process in this flowchart ends.

[0071] [Differentiation] In this embodiment, the inspection device 10, the medical information database 20, the terminal device 100, the information processing device 200, and the large-scale medical information processing server 300 may be configured together with other components. For example, in this embodiment, some of the components of the large-scale medical information processing server 300 may be provided in the information processing device 200.

[0072] Figure 8 shows a modified example of the information processing device of the embodiment. The information processing device 200A shown in Figure 8 includes, for example, a communication interface 210, an input interface 220, a display 230, a processing circuit 240A, and a memory 250A. Compared to the information processing device 200 shown in Figure 3, the information processing device 200A includes a processing circuit 240A and a memory 250A instead of the processing circuit 240 and memory 250. Therefore, the following description will mainly focus on the processing circuit 240A and the memory 250A, and the other components have the same functions as the information processing device 200, so their description will be omitted here.

[0073] The processing circuit 240A includes, for example, a question acquisition function 241, a prompt generation function 242, a filtering function 244, an answer information generation function 245, an output control function 246, and an answer candidate acquisition function 247. Compared to the processing circuit 240, the processing circuit 240A includes the answer candidate acquisition function 247 instead of the inquiry function 243. The memory 250A stores patient information 251, processing data 252, and a large-scale medical information processing model 253. The large-scale medical information processing model 253 is a model similar to the large-scale medical information processing model 331 described above, and is acquired from an external device via a network NW, for example, and updated as needed.

[0074] In the example shown in Figure 8, the answer candidate acquisition function 247 inputs the prompt generated by the prompt generation function 242 into the large-scale medical information processing model 253 stored in memory 250A, and acquires answer candidates corresponding to the prompt. The filtering function 244 filters the answer candidates acquired by the answer candidate acquisition function 247 based on information related to the patient's medical treatment. In this way, the information processing device 200A can acquire answer candidates within its own device without querying an external device (large-scale medical information processing server) for answer candidates, and can perform filtering and generate answer information.

[0075] In addition, in this embodiment, the terminal device 100 may be equipped with a prompt generation function, and the terminal device 100 may generate a prompt and then provide it to the information processing device 200.

[0076] In the embodiments described above, the question acquisition function 241 is an example of a "question acquisition unit," the prompt generation function 242 is an example of a "prompt generation unit," the inquiry function 243 or answer candidate acquisition function 247 is an example of an "answer candidate acquisition unit," the filtering function 244 is an example of a "filtering unit," the answer information generation function 245 is an example of an "answer information generation unit," and the provision function 323 is an example of a "provision unit."

[0077] According to at least one embodiment described above, the information processing device of the embodiment includes: a question acquisition unit that acquires questions relating to a patient's medical information; a prompt generation unit that generates prompts by adding predetermined information to the questions acquired by the question acquisition unit; an answer candidate acquisition unit that acquires answer candidates to the questions from a large-scale medical information processing model using the prompts generated by the prompt generation unit; a narrowing unit that narrows down the answer candidates acquired by the answer candidate acquisition unit based on information relating to the patient's medical treatment; and an answer information generation unit that generates answer information based on the narrowing results of the narrowing unit. By having these components, the information processing device of the embodiment can provide information that resolves a patient's lack of understanding or acceptance of their treatment.

[0078] Specifically, according to the embodiment, even if there is a lack of communication between the patient and the doctor during examinations or rounds, and the patient does not fully understand or agree with the examination or treatment plan, the patient can resolve this lack of understanding and agreement through dialogue with the information processing device 200 (200A). Therefore, according to the embodiment, an interactive system that supports both doctors and patients can be provided.

[0079] The embodiments described above can be expressed as follows. Equipped with processing circuitry, The aforementioned processing circuit is We obtain questions regarding the patient's medical information. A prompt is generated by adding predetermined information to the acquired questions. Using the generated prompt, candidate answers to the questions are obtained from the large-scale medical information processing model. The obtained candidate answers are narrowed down based on the information related to the patient's medical treatment. Generate response information based on the filtering results. Information processing device.

[0080] While several embodiments have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These embodiments can be carried out in a variety of other forms, and various reductions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of Symbols]

[0081] 1…Information processing system, 10…Inspection device, 20…Medical information DB, 100…Terminal device, 110…Communication unit, 120…Touch panel, 130…Microphone, 140…Speaker, 150…Processing unit, 160, 250, 250A, 330…Memory, 200…Information processing device, 210…Communication interface, 220…Input interface, 230…Display, 240, 240A, 320…Processing circuit, 241…Question acquisition function, 242…Prompt generation function, 243…Inquiry function, 244…Narrowing function, 245…Answer information generation function, 246…Output control function, 247…Answer candidate acquisition function, 321…Acquisition function, 322…Answer candidate generation function, 323…Provision function

Claims

1. A question acquisition unit that acquires questions regarding the patient's medical information, A prompt generation unit generates a prompt by adding predetermined information to the question items acquired by the question item acquisition unit, A response candidate acquisition unit obtains response candidate answers to the questions from a large-scale medical information processing model using the prompts generated by the prompt generation unit, The answer candidates obtained by the answer candidate acquisition unit are narrowed down by the narrowing down unit based on the information related to the patient's medical treatment, A response information generation unit generates response information based on the filtering results of the aforementioned filtering unit, An information processing device equipped with the following features.

2. The information relating to the medical treatment of the aforementioned patient includes at least one of the following: information regarding the patient's diagnosis, the physician's treatment plan, and information regarding the hospital's facilities. The information processing apparatus according to claim 1.

3. The filtering unit narrows down the candidate answers according to the priority levels pre-registered for multiple treatment plans by physicians. The information processing apparatus according to claim 1.

4. The filtering unit sets a priority for treatment plans based on the patient's diagnosis information registered by the physician, and narrows down the response candidates according to the set priority. The information processing apparatus according to claim 1.

5. The filtering unit narrows down the candidate answers based on the type of disease or examination site of the patient. The information processing apparatus according to claim 1.

6. The filtering unit narrows down the candidate answers based on the patient's medical information or the timing of the test results. The information processing apparatus according to claim 1.

7. The filtering unit, when narrowing down the candidate answers based on the physician's treatment plan, performs the filtering in such a way that outputs answers that differ from the treatment plan are suppressed. The information processing apparatus according to claim 1.

8. The filtering unit, when narrowing down the candidate answers based on the physician's treatment plan, narrows down the answers so that they do not reject the treatment plan. The information processing apparatus according to claim 1.

9. Computers We obtain questions regarding the patient's medical information. A prompt is generated by adding predetermined information to the acquired questions. Using the generated prompt, candidate answers to the questions are obtained from the large-scale medical information processing model. The obtained candidate answers are narrowed down based on the information related to the patient's medical treatment. Generate response information based on the filtering results. Information processing methods.

10. On the computer, Obtain questions regarding the patient's medical information, The system generates prompts by adding predetermined information to the acquired questions. Using the generated prompt, candidate answers to the questions are obtained from the large-scale medical information processing model. The obtained candidate answers are narrowed down based on information related to the patient's medical treatment. Generate response information based on the narrowed-down results. program.

Images