Medical support system, information processing apparatus, and inference apparatus

Abstract

To provide a medical support system using an existing large-scale language model to easily improve efficiency of recording of medical practices performed by medical workers.SOLUTION: A medical support system 1 includes: recording means for recording a conversation between a patient and a medical worker to generate conversation recording information; transcription means for generating conversation text information from the conversation recording information; and summary generation means for generating summary information of the conversation text information on the basis of conversation processing information including the conversation text information.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present invention relates to a medical support system, an information processing device, and an inference device. [Background technology]

[0002] Medical professionals, including doctors and pharmacists, must record the medical procedures they perform on patients in accordance with a prescribed format established by each medical institution. For example, pharmacists record the details of dispensing medication and medication instructions as a medication history management record (hereinafter referred to as a "medication history"). Techniques for improving the work efficiency of pharmacists in creating medication histories have been known (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2018-197926 Summary of the Invention [Problem to be solved by the invention]

[0004] However, in the conventional technology described in Patent Document 1, it is necessary to rely on rule-based medication history creation, and it is not easy to build a medication history creation support device.

[0005] In view of the above-mentioned problems, the present invention aims to provide a medical support system, an information processing device, and an inference device that make it easier to record medical procedures. [Means for solving the problem]

[0006] In order to achieve the above object, a medical support system according to one aspect of the present invention comprises: A medical support system that supports medical procedures performed by medical professionals, a recording means for recording a conversation between a patient and the medical staff to generate recorded conversation information; a transcription means for generating conversation text information from the conversation recording information; and a summary generating means for generating summary information of the conversation text information based on conversation processing information including the conversation text information. [Effects of the Invention]

[0007] According to a medical support system according to one aspect of the present invention, it is possible to easily build a system that improves the efficiency of medical practitioners' recording of medical procedures and the like.

[0008] Problems, configurations, and effects other than those described above will become apparent from the detailed description of the invention that follows. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is an overall configuration diagram showing an example of a medical support system 1 according to an embodiment. [Figure 2] FIG. 2 is a block diagram showing an example of a user terminal device 2. [Figure 3] FIG. 2 is a block diagram showing an example of an information processing device 3. [Figure 4] 2 is an example of a user interface of the medical support system 1 according to the embodiment. [Figure 5] 2 is an example of a database configuration of the medical support system 1 according to the embodiment. [Figure 6] FIG. 9 is a hardware configuration diagram showing an example of a computer 900. [Figure 7] 4 is a flowchart showing an example of an information processing method of the medical support system 1 according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, an embodiment for carrying out the present invention will be described with reference to the drawings. The scope necessary for the explanation to achieve the object of the present invention will be schematically shown, and the scope necessary for explaining the relevant parts of the present invention will be mainly explained, and the parts that are omitted from the explanation will be based on publicly known techniques.

[0011] The following embodiment will be described assuming that the user of the present invention is a pharmacist. However, when the present invention is actually implemented, the target users are not limited to pharmacists, but include all medical professionals such as doctors, dentists, nurses, and caregivers.

[0012] FIG. 1 is an overall configuration diagram showing an example of a medical support system 1 according to an embodiment. The medical support system 1 includes, as its main components, a user terminal device 2 and an information processing device 3. The user terminal device 2 and the information processing device 3 are connected to a wired or wireless network 4 and configured to be able to transmit and receive various data to and from each other. Note that the number of user terminal devices 2 and information processing devices 3 and the connection configuration of the network 4 are not limited to the example in FIG. 1 and may be changed as appropriate.

[0013] The user terminal device 2 is a client-type computer, and is configured as a general-purpose or dedicated computer (see FIG. 6 described later), etc. The user terminal device 2 accepts various input operations via a display screen such as an app or browser, and outputs various information via the display screen or voice, in order to record conversations between the medical professional U and the patient and to check various information.

[0014] The information processing device 3 is a server-type computer or a cloud-type computer, and is configured by a general-purpose or dedicated computer (see FIG. 6 described later), etc. The information processing device 3 uses a large-scale language model 320 to generate summary information D53, etc., which will be described later.

[0015] Fig. 2 is a block diagram showing an example of the user terminal device 2, Fig. 3 is a block diagram showing an example of the information processing device 3, Fig. 4 is an example of a user interface of the medical support system 1 according to the embodiment, and Fig. 5 is an example of a database configuration of the medical support system 1 according to the embodiment. Using these figures, the configurations of the user terminal device 2 and the information processing device 3 will be explained below, along with the functions of the user interface as appropriate.

[0016] In FIG. 4, the user interface is displayed on a display 240 (described later), and includes at least a recording button 40, a conversation playback function 41, a conversation text information display screen 42, and a summary information display screen 43.

[0017] (Configuration of user terminal device 2) The user terminal device 2 includes a control unit 20, a storage unit 21, a communication unit 22, an input unit 23, and an output unit 24.

[0018] The communication unit 22 is connected to an external device (for example, the information processing device 3, etc.) via the network 4, and functions as a communication interface for transmitting and receiving various types of data.

[0019] The input unit 23 includes at least a microphone 230. The input unit may also have a function of accepting various input operations via other input devices (not shown).

[0020] The microphone 230 functions as a recording means. That is, it records the conversation between the medical worker U who is near the user terminal device 2 and the patient, and generates conversation recording information D50. Specifically, when the medical worker U presses the record button 40 in Fig. 4, recording of the conversation by the microphone 230 begins. Pressing the record button 40 again stops the recording.

[0021] The output unit 24 includes at least a display 240 and a speaker 241 .

[0022] The display 240 functions as a user interface by outputting various information via a display screen. Specifically, the display 240 displays the user interface shown in FIG.

[0023] The speaker 241 functions as a user interface by outputting various information via voice. Specifically, when the medical worker U presses the play button 410 in Fig. 4, the conversation playback function 41 is activated. Here, the conversation playback function 41 has the function of playing back the recorded conversation information D50 as voice from the speaker 241 by pressing the play button 410, and adjusting the volume by operating the volume button 411.

[0024] The microphone 230, the display 240, and the speaker 241 may be built into a general-purpose or dedicated computer, or may be externally connected to the general-purpose or dedicated computer via wire or wirelessly.

[0025] The storage unit 21 stores a database 210 and an information processing program 211. As shown in Fig. 5, the database 210 stores various types of information processed or generated by the user terminal device 2, such as conversation recording information D50 and conversation text information D520 (described later).

[0026] The control unit 20 executes an information processing program 211 recorded in the storage unit 21 to function as a transmission / reception control unit 200, a database management unit 201, an information processing unit 202, and an output format selection unit 203.

[0027] (Transmission and reception control unit 200) The transmission / reception control unit 200 transmits and receives various types of data to and from an external device (such as the information processing device 3). For example, the transmission / reception control unit 200 transmits display information to the information processing device 3 for outputting various display screens on the information processing device 3, and receives operation information from the information processing device 3 for accepting input operations performed on the display screen of the information processing device 3. The transmission / reception control unit 200 cooperates with the database management unit 201, the information processing unit 202, and the output format selection unit 203 to transmit the display information and receive the operation information.

[0028] (Database Management Department 201) The database management unit 201 links the conversation recording information D50 with conversation text information D520 (described later) and registers them in the database 210.

[0029] (Information processing unit 202) The information processing unit 202 functions as a transcription unit. That is, it transcribes the conversation recording information D50 to generate conversation text information D520. Specifically, it uses existing voice recognition software to convert the conversation recording information D50, which is voice data, into the conversation text information D520, which is text data. An example of voice recognition software is a voice recognition AI called Whisper provided by OpenAI.

[0030] Then, the database management unit 201 links the conversation recording information D50 with the conversation text information D520 generated by the information processing unit 202 based on the conversation recording information D50, and registers the linked information in the database 210. Specifically, the conversation text information D520 generated by the information processing unit 202 is displayed on the conversation text information display screen 42 in FIG.

[0031] (Output format selection unit 203) The output format selection unit 203 allows the medical worker U to select the output format of the summary information D53 generated by the information processing device 3. Details will be described later.

[0032] (Configuration of information processing device 3) The information processing device 3 includes a control unit 30, a data storage unit 31, a trained model storage unit 32, a communication unit 33, an input unit 34, and an output unit 35.

[0033] The communication unit 33 is connected to an external device (such as the user terminal device 2) via the network 4 and functions as a communication interface for transmitting and receiving various types of data. The input unit 34 accepts various input operations, and the output unit 35 functions as a user interface by outputting various types of information via a display screen or voice. Note that the input unit 34 and the output unit 35 may be omitted.

[0034] The trained model storage unit 32 stores a large-scale language model 320.

[0035] A large-scale language model is a language model built using extremely large datasets and deep learning technology. For example, open source large-scale language models such as OpenAI's GPT and Google's PaLM2 are well-known. In addition, ChatGPT provided by OpenAI and BARD by Google are well-known as interactive generative AI services that use these models. In recent years, technologies and services that allow these large-scale language models to be used on-premise have also become known.

[0036] The large-scale language model 320 stored in the trained model storage unit 32 may be provided to another device via the network 4, a recording medium, etc. Furthermore, the number of large-scale language models 320 stored in the trained model storage unit 32 is not limited to one, and multiple large-scale language models with different conditions, such as different types of open-source large-scale language models, may be stored and made available selectively or in parallel.

[0037] The data storage unit 31 stores a database 310 and an information processing program 311. As shown in Fig. 5, the database 310 stores prompt information D521 indicating language processing commands for conversation text information D520. Note that the database 310 may include multiple pieces of prompt information D521 corresponding to each piece of output format information described below.

[0038] Prompts are typically part of instructions or information given to a large-scale language model that dictate the content and response the model generates. They are typically given in the form of a sentence or question, and the model generates a response based on that. Furthermore, when this prompt is input into a large-scale language model along with specific data, the specific data is processed based on the prompt, allowing the model to produce a desired output that is tailored to the specific data.

[0039] In this embodiment, the prompt information D521 has the same functions and properties as the prompt information described above.

[0040] In the medical field, a summary format known as the SOAP format is commonly used when recording patient information. When summarizing conversation text information D520 in the SOAP format, the prompt information D521 can include, for example, an explanation of the SOAP format and an example of a summary of the text when using the SOAP format.

[0041] The prompt information D521 may be, for example, "You are an assistant whose job is to extract the chief complaint from the conversation between the pharmacist and the patient. Create a medication record from the given conversation. Please interpret the conversation between the pharmacist and the patient from the text and create a medication record. It will be recorded according to the SOAP method. In the SOAP method, the contents are categorized and recorded. [S] (Subject): Subjective information of the patient [O] (Object): Objective information [A] (Assessment): Analysis and opinion as a pharmacist [P] (Plan): Future plan for the problem The extraction conditions are as follows: ===== The information that should be recorded includes symptoms, tests, illness, medication status, remaining medication, changes in physical condition, concomitant medications (including prescription drugs, over-the-counter drugs, quasi-drugs, and health foods), medical history (past illnesses), visits to other departments (whether the patient has been treated at a hospital other than the one for the current illness), side effects, food and drink, intention to use generic drugs (whether to use them), whether the patient has a medication record, lifestyle habits, patient characteristics, patient constitution (including allergy history, side effect history, etc.), and key points of medication instructions. Please extract and summarize the conversation. [S] (Subject): Subjective data of the patient Subjective information of the patient (what the patient felt, reported, or what the family said), symptoms, examinations, diseases, medication status, remaining medications confirmation, physical condition changes, concomitant medications (including drugs requiring guidance from a doctor, general drugs, quasi-drugs, and health foods), past medical history (past illnesses), consultations at other departments (whether there is a hospital where the patient is being treated for a disease other than the current one), side effects, diet, intention to use generic drugs, presence of a medication diary, lifestyle habits, patient characteristics, patient constitution (including allergy history, side effect history, etc.) are concisely summarized and described item by item. <Example sentences of S> · Improvement in constipation. However, there are times when it becomes loose. · Linzess may be taken after meals sometimes. · Low back pain has improved. There is stomach pain. [O] (Object): Objective data Information obtained through examinations, measurements, etc., and symptoms observed during a medical examination. [[ID=2^]] Let's summarize the objective information in [O]. Record examination values such as blood pressure and Ccr. ※ Also record the medication guidance implemented and the information obtained therefrom here. Example) Blood sampling data, examination findings such as X-ray, MRI, CT, vital signs, the patient's expression, skin color, symptoms, etc. <When asking about the following items for examination values, please attach the following units after the number. Record as "Examination value item (pronunciation): unit". The pronunciation is not required when outputting> ● Red blood cell count (sekkyekkūsuu): 10,000 / μL ● Hb (hemoglobin, blood hemoglobin amount kesshikisoryō): g / dL ● HbA1c (hemoglobin A1c): % <Example sentences of O> · Linzess may be administered after meals sometimes, and the compliance was not good. · Villanoa was taken on an empty stomach, resulting in poor compliance. · This prescription has been changed to Dezalex, which is to be taken after meals. [A] (Assessment) [S] (Subjective data) and [O] (Objective data) are considered to conduct an analysis of possible problems. [A] describes the evaluation of [S] and [O] from the perspective of a pharmacist. Example: What is the current condition of the user / patient, and what possible problems may arise in the future? <Example of [A]> · Post - meal administration of Liness may cause diarrhea. · Arrowzen is considered to be able to reduce diarrhea by switching to a single - dose administration. · During this medication guidance, the usage was re - confirmed to improve adherence. · Understanding of usage and dosage is good. · Since the usage of oral medications has changed, there is a possibility of forgetting to take them. · Concurrent use of alcohol may enhance side effects. · Explain to take Gegen Decoction before meals, and it has a relatively quick effect among Chinese herbal medicines. · Explain that loxoprofen and levamisole should be taken with a gap of more than 4 hours during pain. [P] (Plan): Medication guidance plan [A] (Assessment) is used to formulate care and other measures to improve the condition of the user / patient considered. The purpose of the medication guidance plan is to practice better evidence - based care for the user, and the family is also included. Not only the user but also the family is the target. Record the specific solutions to the problems revealed so far. When there is any change or when doubts or anxieties are expressed from a casual conversation, record it in [P] for follow - up after the next time. <Example of [P]> · Observe whether the compliance with the administration method of Liness can improve fecal control. Check how many times you have taken Aloezen as an emergency dose. - Double check that you understand how to take the medicine. It may also be stated as follows.

[0042] 3, the data storage unit 31 and the trained model storage unit 32 are shown as two storage units, but they may be configured as a single storage unit or three or more storage units. Furthermore, at least one of the data storage unit 31 and the trained model storage unit 32 may be configured as a storage unit of an external computer (for example, a server-type computer or a cloud-type computer).

[0043] The control unit 30 executes an information processing program 311 recorded in the data storage unit 31, thereby functioning as a transmission / reception control unit 300, a database management unit 301, and an information generation unit 302.

[0044] (Transmission and reception control unit 300) The transmission / reception control unit 300 transmits and receives various types of data to and from an external device (such as the user terminal device 2). For example, the transmission / reception control unit 300 transmits display information to the user terminal device 2 for outputting various display screens on the user terminal device 2, and receives operation information from the user terminal device 2 for accepting input operations performed on the display screen of the user terminal device 2. The transmission / reception control unit 300 cooperates with the database management unit 301 and the information generation unit 302 to transmit display information and receive operation information.

[0045] (Database Management Department 301) The database management unit 301 registers in the database 310 the conversation processing information D52, summary information D53, and feedback information D54 generated by the information generation unit 302 (details of which will be described later).

[0046] The various pieces of information registered in the database 310 may be made accessible to the user terminal device 2 by the transmission / reception control unit 300. In this case, editing operations such as adding, deleting, and correcting each piece of information may be performed on the display screen of the user terminal device 2.

[0047] (Information generation unit 302) The information generation unit 302 includes a conversation processing information generation unit 302A, a summary information generation unit 302B, and a feedback information generation unit 302C.

[0048] (Conversation processing information generation unit 302A) The conversation processing information generation unit 302A acquires conversation text information D520 that the transmission / reception control unit 300 receives from the user terminal 2. Then, the conversation text information D520 is combined with prompt information D521 stored in the database 310 to generate conversation processing information D52. (Summary information generation unit 302B)

[0049] The summary information generating unit 302B functions as a summary information generating unit, that is, generates summary information D53 for the conversation text information D520.

[0050] Specifically, the summary information generation unit 302B generates summary information D53 by inputting the conversation processing information D52 generated by the conversation processing information generation unit 302A into the large-scale language model 320. When the transmission / reception control unit 300 transmits the summary information D53 to the user terminal device 2, the transmission / reception control unit 200 of the user terminal device 2 receives the summary information D53. The summary information D53 received by the user terminal device 2 is registered in the database 210 by the database management unit 201. Furthermore, the summary information D53 is displayed on the summary information display screen 43 in FIG. 4.

[0051] The summary information D53 displayed on the summary information display screen 43 may be further selected by the medical worker U. For example, when the summary information D53 is displayed for each predetermined item on the summary information display screen 43 in Fig. 4, the medical worker U may check the check boxes 430 for the items of the summary information D53 that he / she deems necessary and uncheck the check boxes 430 for the items of the summary information D53 that he / she deems unnecessary, thereby treating only the checked items of the summary information D53 as the summary information D53.

[0052] (Feedback information generation unit 302C) The feedback information generation unit analyzes the summary information D53 and generates feedback information D54 for the medical worker U. Details will be described later.

[0053] (Hardware configuration of each device) 6 is a hardware configuration diagram showing an example of a computer 900. The user terminal device 2 and the information processing device 3 in the medical support system 1 are configured by a general-purpose or dedicated computer 900.

[0054] 6, the computer 900 includes, as its main components, a bus 910, a processor 912, a memory 914, an input device 916, an output device 917, a display device 918, a storage device 920, a communication I / F (interface) unit 922, an external device I / F unit 924, an I / O (input / output) device I / F unit 926, and a media input / output unit 928. Note that the above components may be omitted as appropriate depending on the application of the computer 900.

[0055] The processor 912 is composed of one or more arithmetic processing devices (such as a central processing unit (CPU), a micro-processing unit (MPU), a digital signal processor (DSP), or a graphics processing unit (GPU)), and operates as a control unit that controls the entire computer 900. The memory 914 stores various data and programs 930, and is composed of, for example, a volatile memory (such as a DRAM or SRAM) that functions as a main memory, a non-volatile memory (ROM), a flash memory, etc.

[0056] The input device 916 is composed of, for example, a keyboard, a mouse, a numeric keypad, an electronic pen, etc., and functions as an input unit. The output device 917 is composed of, for example, a sound (audio) output device, a vibration device, etc., and functions as an output unit. The display device 918 is composed of, for example, a liquid crystal display, an organic EL display, electronic paper, a projector, etc., and functions as an output unit. The input device 916 and the display device 918 may be integrated into one device, such as a touch panel display. The storage device 920 is composed of, for example, a hard disk drive (HDD), a solid state drive (SSD), etc., and functions as a storage unit. The storage device 920 stores various data necessary for executing the operating system and the program 930.

[0057] The communication I / F unit 922 is connected to a network 940 (which may be the same as network 4 in FIG. 1 ) such as the Internet or an intranet via a wired or wireless connection, and functions as a communication unit that transmits and receives data to and from other computers in accordance with a predetermined communication protocol. The external device I / F unit 924 is connected to an external device 950 (such as a camera, printer, scanner, or reader / writer) via a wired or wireless connection, and functions as a communication unit that transmits and receives data to and from the external device 950 in accordance with a predetermined communication protocol. The I / O device I / F unit 926 is connected to an I / O device 960 (such as various sensors and actuators), and functions as a communication unit that transmits and receives various signals and data, such as detection signals from sensors and control signals to actuators, to and from the I / O device 960. The media input / output unit 928 is composed of a drive device (such as a DVD (Digital Versatile Disc) drive or a CD (Compact Disc) drive), a memory card slot, and a USB connector, and reads and writes data from and to media (non-transitory storage media) 970 (such as a DVD, CD, memory card, or USB memory).

[0058] In the computer 900 having the above configuration, the processor 912 loads a program 930 stored in the storage device 920 into the memory 914, executes the program, and controls each unit of the computer 900 via the bus 910. The program 930 may be stored in the memory 914 instead of the storage device 920. The program 930 may be recorded on the medium 970 in an installable file format or an executable file format and provided to the computer 900 via the media input / output unit 928. The program 930 may be provided to the computer 900 by being downloaded via the network 940 via the communication I / F unit 922. Furthermore, the computer 900 may implement various functions realized by the processor 912 executing the program 930 using hardware such as an FPGA or an ASIC.

[0059] The computer 900 is, for example, a desktop computer or a portable computer, and is an electronic device of any type. The computer 900 may be a client computer, a server computer, or a cloud computer. The computer 900 may also be applied to devices other than the devices 2 to 6. Furthermore, when the computer 900 is a portable computer, it is preferable that the computer 900 has a function of acquiring location information of its own location and is configured to use the location information for various data processing.

[0060] (Information processing method) Fig. 7 is a flowchart showing an example of an information processing method in the medical support system 1 according to the embodiment. Note that the series of information processing methods in the medical support system 1 shown in Fig. 7 assumes that a single large-scale language model 320 is stored in the trained model storage unit 32 of the information processing device 3.

[0061] First, in step S100 (conversation recording information generation process), the microphone 230 of the user terminal device 2 records the conversation between the medical staff U and the patient to generate conversation recording information D50. The database management unit 201 registers the conversation recording information D50 in the database 210.

[0062] In step S110 (conversation text information generation step), the information processing unit 202 of the user terminal device 2 generates conversation text information D520, which is text data, by performing a voice recognition process on the conversation recording information D50, which is the voice data generated in step S100. Then, the database management unit 201 links the conversation text information D520 to the conversation recording information D50 stored in the database 210 and registers it in the database 210.

[0063] In step S120 (conversation processing information generation step), the conversation processing information generation unit 302A of the information processing device 3 acquires the conversation text information D520 by receiving the conversation text information D520 generated in step S110 from the user terminal device 2. Then, the conversation text information D520 is combined with prompt information D521 stored in the database 310 to generate conversation processing information D52.

[0064] In step S130 (conversation processing information input step), the summary information generation unit 304A inputs the conversation processing information D52 generated in step S120 to the large-scale language model 320.

[0065] In step S131 (summary information generation process), the large-scale language model 320 performs language processing on the conversation text information D520 included in the conversation processing information D52 generated in step S120 in accordance with the prompt information D521 included in the conversation processing information D52 generated in step S120, and generates summary information D53.

[0066] In step S140 (summary information acquisition step), the transmission / reception control unit 300 of the information processing device 3 transmits the summary information D53 generated in step S131 to the user terminal device 2. As a result, the summary information D53 is displayed on the display 240 of the user terminal device 2 as a summary of the conversation text information D521.

[0067] Next, we will explain the functions of the output format selection unit 203 of the user terminal device 2 and the feedback information generation unit 302C of the information processing device 3. Note that the configurations, operations, and various data of the user terminal device 2 and the information processing device 3 other than these units are the same as those in the above embodiment, so detailed explanations will be omitted.

[0068] (Function of output format selection unit 203) The output format selection unit 203 allows the medical worker U to select an output format for the summary information D53. In the medical field, formats used to record patient information include the SOAP format, the itemized list format, and the categorized format, in addition to the SOAP format described above, and the format adopted varies depending on the medical institution. For example, the interface shown in FIG. 4 illustrates a summary in the SOAP format. To generate summary information D53 that conforms to the format determined by each medical institution, the output format selection unit 203 presents the medical worker U with output format candidate information D51, which is a list of two or more output format information candidates (e.g., output format information D510 in the SOAP format, output format information D511 in the itemized list format, output format information D512 in the itemized list format, and output format information D513 in the categorized format), and the medical worker U selects one of the output format candidate information D51.

[0069] Specifically, the output format selection unit 203 displays output format candidate information D51 on the display 240 of the user terminal device 2. When the medical professional U performs an input operation to select one of the output format information from the output format candidate information D51 via the input unit 23, the output format selection unit 203 transmits the output format information selected by the medical professional U to the information processing device 3 together with conversation text information D520. The database 310 of the information processing device 3 stores prompt information D521 for each of the output format information. When the information processing device 3 acquires the conversation text information D520 and the output format information, the conversation processing information generation unit 302A combines the prompt information D521 corresponding to the output format information with the conversation text information D520 to generate conversation processing information D52. The summary information generation unit 302B generates summary information D53 in accordance with the output format information selected by the medical professional U.

[0070] (Function of feedback information generation unit 302C) The feedback information generation unit 302C analyzes the summary information D53 and generates feedback information for the medical professional U. For example, the quality of information contained in a medication history varies greatly depending on the pharmacist's experience in providing medication instructions. Therefore, analyzing the medication history created by the pharmacist can be used to educate the pharmacist. In order to provide information that is effective for educating medical professionals, the feedback information generation unit 302C analyzes the summary information D53 generated by the summary information generation unit 302B and generates feedback information D54.

[0071] Specifically, the feedback information generation unit 302C calculates the similarity between past summary information created by a skilled medical professional (not shown) stored in the database 330 of the information processing device 3 and summary information D53 generated by the summary information generation unit 302B based on a conversation between a patient and a medical professional U, who is the subject of feedback and a user of the medical support system 1, by, for example, performing morphological analysis on both information and vectorizing them, and then calculating the cosine similarity between the vectors.The calculated cosine similarity is then evaluated using a predetermined reference value, thereby generating feedback information D54 that expresses an evaluation of the medical treatment, etc. performed by the medical professional U on a predetermined scale, etc.

[0072] (Other embodiments) The present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit of the present invention, all of which are included in the technical concept of the present invention.

[0073] In the above embodiment, the transcription means has been described as functioning in the information processing unit 202 of the user terminal device 2, but the transcription means may also function in the information processing device 3. For example, when the information processing device 3 receives conversation recording information D50 from the user terminal device 2, the conversation processing information generation unit 302A may transcribe the conversation recording information D50 to generate conversation text information D520.

[0074] In the above embodiment, the information processing device 3 has been described as being configured as a single device, but it may be configured as a plurality of devices. For example, the units 302A, 302B, and 302C of the information processing device 3 may be distributed across a plurality of devices, so that the information processing device 3 may be configured as an information processing device including a conversation processing information generation unit 302A and performing a conversation processing information generation step, an information processing device including a summary information generation unit 302B and performing a summary information generation step, and an information processing device including a feedback information generation unit 302C and performing a feedback information generation step. In this case, each unit (each step) of each of the above devices may be realized by a program (an information processing program or a machine learning program) executable by the computer 900.

[0075] Each of the above devices can be configured, for example, as follows: The configuration and operation of each part of each device and the various types of data handled by each device are the same as those in the above embodiment, and therefore detailed description thereof will be omitted.

[0076] When the information processing device performing the conversation processing information generation step acquires conversation text information D520 generated by the user terminal device 2, it generates conversation processing information D52 by combining the conversation text information D520 with prompt information D521 stored in the database 310 of the information processing device 3. Note that if multiple pieces of prompt information D521 are stored in the database 310, the conversation processing information D52 is generated by combining the conversation text information D520 with any one of the multiple pieces of prompt information D521.

[0077] The information processing device performing the summary information generation step inputs the conversation processing information D52 into the large-scale language model 320, thereby generating summary information D53 of the conversation text information D520 included in the conversation processing information D52.

[0078] The information processing device performing the feedback information generating step analyzes the summary information D53 and generates feedback information D54 for the medical worker U.

[0079] The transmission / reception control unit 300 and the database management unit 301 may be provided in each of the above devices. The database 310 may be configured to be accessible from each of the above devices, and may be stored in a storage unit of one of the devices, or in a storage unit of an external computer. Furthermore, some of the above devices may be realized by the user terminal device 2.

[0080] (Inference device, inference method and inference program) The present invention can be provided not only in the form of the information processing device 3 (information processing method or information processing program) according to the above embodiment, but also in the form of an inference device (inference method or inference program) used to infer summary information D53. In this case, the inference device (inference method or inference program) can include a memory and a processor, and the processor executes a series of processes. The series of processes includes an information acquisition process (information acquisition step) for acquiring conversation processing information D52, and an inference process for inferring summary information of the conversation text information D520 included in the conversation processing information D52 based on prompt information D521 included in the conversation processing information D52, once the conversation processing information D52 is acquired by the information acquisition process.

[0081] By providing the inference device (inference method or inference program), it becomes possible to apply the inference device (inference method or inference program) to various devices more easily than when it is implemented in an information processing device. It will be obvious to those skilled in the art that when the language processing program (the language processing program) infers summary information, the inference method implemented by the information generation unit may be applied using the large-scale language model according to the above embodiment.

[0082] In the above-described embodiment, it is assumed that the medical professional U who is the user of the user terminal 2 is a pharmacist, but the user is not limited to a pharmacist and may be any medical professional such as a doctor, dentist, nurse, caregiver, etc. For example, when a doctor writes the details of medical treatment in a medical record (chart), it is common to summarize the details in a predetermined format such as SOAP format, SO format, itemized format, or categorized format, so the present invention can be implemented with a doctor as the user. [Explanation of symbols]

[0083] 1... medical support system, 2... user terminal device, 3... information processing device, 20...control unit, 21...storage unit, 22...communication unit, 23...input unit, 24...output unit, 200...transmission / reception control unit, 201...database management unit, 202...information processing unit, 203...output format selection unit, 210...database, 211...information processing program, 230...microphone, 240...display, 241...speaker, 30...control unit, 31...data storage unit, 32...trained model storage unit, 33...communication unit, 34...input unit, 35...output unit, 300...transmission / reception control unit, 301...database management unit, 302A...conversation processing information generation unit, 302B...summary information generation unit, 302C...feedback information generation unit, 310...Database, 311...Information processing program, 320...Large-scale language model

Claims

1. A medical support system that supports medical procedures performed by medical professionals, a recording means for recording a conversation between a patient and the medical staff to generate recorded conversation information; a transcription means for generating conversation text information from the conversation recording information; a summary generating means for generating summary information of the conversation text information based on conversation processing information including the conversation text information, Medical support system.

2. the conversation processing information in the summary generating means further includes prompt information indicating a language processing command for the conversation text information; The medical support system according to claim 1 .

3. further comprising an output format selection means for selecting an output format of the summary information; the conversation processing information in the summary generating means further includes the prompt information corresponding to the output format selected by the output format selecting means; The medical support system according to claim 2 .

4. The system further comprises a feedback information generating means for analyzing the summary information and generating feedback information for the medical professional. The medical support system according to any one of claims 1 to 3.

5. the summary generation means generates the summary information from the conversation text information included in the conversation processing information by inputting the conversation processing information into a learning model; The learning model is a large-scale language model that learns a correlation between the conversation processing information and the summary information; The medical support system according to any one of claims 1 to 3.

6. a conversation processing information generation unit that acquires conversation processing information including conversation text information obtained by transcribing a conversation between a patient and a medical professional and prompt information that indicates a language processing command for the conversation text information; an information generation unit that generates summary information of the conversation text information from the conversation text information included in the conversation processing information by inputting the conversation processing information acquired by the conversation processing information generation unit into a learning model, The learning model is a large-scale language model that learns a correlation between the conversation processing information and the summary information; Information processing device.

7. An inference device comprising a memory and a processor, The processor: an information acquisition process for acquiring conversation processing information including conversation text information obtained by transcribing a conversation between a patient and a medical professional and prompt information indicating a language processing instruction for the conversation text information; When the conversation processing information is acquired by the information acquisition process, an inference process is executed to infer summary information of the conversation text information from the conversation text information included in the conversation processing information. Reasoning device.

Images