Program, information processing method, and information processing device

Abstract

This program causes a computer to execute processing for: acquiring input information representing conversational content between a medical worker and a patient, and summary information summarizing history related to treatment of the patient; and outputting record information on the treatment of the patient by inputting the input information and the summary information to a language model. Preferably, the record information is outputted by the language model in the SOAP format including subjective information, objective information, an assessment, and a plan.

Description

Program, information processing method and information processing device

[0001] The present disclosure relates to a program, an information processing method, and an information processing device.

[0002] Patent document 1 discloses a system that uses a patient's medical records, patient profile, surgery records, examination records, etc. to generate a medical summary of a patient's past medical treatment and a medical summary such as a patient's prognosis prediction.

[0003] JP 2024-034528 A

[0004] However, the invention of Patent Document 1 had a problem in that it was not possible to output recorded information regarding a patient's treatment based on input information indicating the content of the conversation between a medical professional and a patient, and summary information summarizing the patient's treatment history.

[0005] In one aspect, an object of the present invention is to provide a program or the like that can output record information related to the treatment of a patient.

[0006] A program according to one aspect causes a computer to execute a process of acquiring input information indicating the content of a conversation between a medical professional and a patient and summary information summarizing the patient's treatment history, and outputting record information regarding the patient's treatment by providing the input information and the summary information to a language model.

[0007] In one aspect, recorded information regarding the patient's treatment can be output.

[0008] FIG. 1 is an explanatory diagram showing an overview of a medical record system. FIG. 1 is a block diagram showing an example of the configuration of a server. FIG. 2 is an explanatory diagram showing an example of the record layout of a summary information DB. FIG. 3 is an explanatory diagram showing an example of the record layout of a record information DB. FIG. 4 is a block diagram showing an example of the configuration of a computer. FIG. 5 is a flowchart showing an example of a processing procedure of a medical record system. FIG. 6 is a flowchart showing an example of a processing procedure of a medical record system. FIG. 7 is an explanatory diagram showing an example of an initial screen. FIG. 8 is an explanatory diagram showing an example of a setting screen. FIG. 9 is an explanatory diagram showing an example of a first prompt. FIG. 10 is an explanatory diagram showing an example of a setting screen. FIG. 11 is an explanatory diagram showing an example of a second prompt. FIG. 12 is an explanatory diagram showing an example of a setting screen. FIG. 13 is a flowchart showing an example of a processing procedure of a medical record system according to embodiment 2. FIG. 14 is an explanatory diagram showing an example of a setting screen according to embodiment 2. FIG. 15 is an explanatory diagram showing an example of an additional prompt. FIG. 16 is an explanatory diagram showing an example of a setting screen according to embodiment 2. FIG. 17 is a flowchart showing an example of a processing procedure of a medical record system according to embodiment 3. FIG. 18 is an explanatory diagram showing an example of a setting screen according to embodiment 3. FIG. 19 is an explanatory diagram showing an example of a correction prompt. FIG. 19 is an explanatory diagram showing an example of a setting screen according to embodiment 3. FIG. 19 is an explanatory diagram showing an example of a third prompt. FIG. 19 is an explanatory diagram showing an example of a setting screen according to embodiment 3. FIG. 19 is an explanatory diagram showing an example of a configuration of a second language model. FIG. 20 is a flowchart showing an example of a processing procedure of a medical record system according to embodiment 4. FIG. 10 is an explanatory diagram showing an example of a setting screen according to embodiment 4. FIG. 10 is an explanatory diagram showing an example of the configuration of a medical record system according to embodiment 5. FIG. 10 is a flowchart showing an example of a processing procedure of the medical record system according to embodiment 5. FIG. 10 is an explanatory diagram showing an example of a screen. FIG. 10 is a flowchart showing an example of a processing procedure of the medical record system according to embodiment 6. FIG. 10 is an explanatory diagram showing an example of a file corresponding to the first type. FIG. 10 is an explanatory diagram showing an example of a file corresponding to the second type. FIG. 10 is an explanatory diagram showing an example of a fourth prompt. FIG. 10 is a flowchart showing an example of a processing procedure of the medical record system according to embodiment 7. FIG. 10 is an explanatory diagram showing an example of a fifth prompt. FIG. 10 is an explanatory diagram showing an example of an explanation for a patient. FIG. 10 is a flowchart showing an example of a procedure for outputting advice for medical professionals. FIG. 10 is an explanatory diagram showing an example of advice for medical professionals.10 is a flowchart showing an example of a procedure for outputting a report for medical staff. FIG. 11 is an explanatory diagram showing an example of a report for medical staff. FIG.

[0009] (Embodiment 1) Fig. 1 is an explanatory diagram showing an overview of a medical record system. The medical record system outputs record information related to a patient's treatment based on input information indicating the content of a conversation between a medical professional and a patient and summary information summarizing the patient's treatment history. The medical record system includes an information processing device 10 and an information processing device 20. The medical record system transmits and receives information via a network N.

[0010] The information processing device 10 processes, stores, and transmits / receives various types of information. The information processing device 10 is, for example, a server device, a personal computer, or a general-purpose tablet PC (personal computer). The information processing device 10 may also be a cloud server device that provides functions included in the information processing device 10 as cloud services. In this embodiment, the information processing device 10 will be described as the server 10.

[0011] The information processing device 20 is, for example, a server device, a smartphone, a tablet, a personal computer (hereinafter referred to as a computer), a general-purpose tablet PC, etc. In this embodiment, the information processing device 20 will be described as a computer 20.

[0012] 2 is a block diagram showing an example of the configuration of the server 10. The server 10 includes a control unit 11, a storage unit 12, a communication unit 13, a mass storage unit 14, and a reading unit 15. The above-mentioned units are connected to each other via a bus. The control unit 11 is configured using one or more processors, such as a central processing unit (CPU), a micro-processing unit (MPU), a graphics processing unit (GPU), a general-purpose computing on graphics processing unit (GPGPU), a tensor processing unit (TPU), or an AI chip (semiconductor for AI). The control unit 11 appropriately executes a control program 12P stored in the storage unit 12 to perform various information processing and control processing related to the server 10.

[0013] The storage unit 12 includes a RAM (Random Access Memory) or a ROM (Read Only Memory), etc. The storage unit 12 stores various data and the like necessary for a control program 12P (program product) executed by the control unit 11. The storage unit 12 temporarily stores data and the like generated when the control unit 11 executes the control program 12P.

[0014] Furthermore, the storage unit 12 stores a language model M1. The language model M1 is, for example, a general-purpose large-scale language model (LLM) constructed by performing unsupervised pre-learning using a large set of sentences and images. The language model M1 is, for example, a Generative Pre-trained Transformer (GPT)-3, GPT-3.5, GPT-4, Receptance Weighted Key Value (RWKV), PaLM2, or Large Language Model Meta AI (LLaMa). When data such as input information and summary information is input, the language model M1 performs a calculation to generate record information according to the data and outputs the generated record information. When data such as record information is input, the language model M1 performs a calculation to generate summary information according to the data and outputs the generated summary information. The input information includes, for example, the content of a conversation between a medical professional and a patient, the medical professional's findings on the patient, or treatment information for the patient. The summary information is, for example, a history of the patient's treatment. The summary information may include peripheral information about the patient in addition to the patient's treatment history. The peripheral information about the patient includes the patient's wishes regarding treatment policy, the patient's functional assessment, and the patient's living situation. The record information includes information about the patient's treatment. The record information and summary information output by the language model M1 are examples and are not limited to these. The language model M1 can output multiple types of information tailored to the intended use based on the input information, summary information, etc.

[0015] The language model M1 may be configured by combining a plurality of algorithms. Instead of storing the language model M1 in the storage unit 12, the control unit 11 may input the input information and the summary information into the language model M1 by accessing a language processing server that stores the language model M1.

[0016] The communication unit 13 is a communication module that transmits and receives information to and from the computer 20 via the network N. The mass storage unit 14 includes a RAM, a ROM, etc. The mass storage unit 14 stores a summary information DB 14A and a record information DB 14B. The summary information DB 14A and the record information DB 14B will be described later.

[0017] In this embodiment, the storage unit 12 and the mass storage unit 14 may be configured as an integrated storage device. The mass storage unit 14 may be configured by a plurality of storage devices. The mass storage unit 14 may be an external storage device connected to the server 10.

[0018] The reading unit 15 reads information stored in the portable storage medium 1a. The portable storage medium 1a may be, for example, a CD (Compact Disc), a DVD (Digital Versatile Disc), a USB (Universal Serial Bus) memory, or an SD (Secure Digital). The reading unit 15 reads the control program 12P from the portable storage medium 1a. The control unit 11 stores the read control program 12P in the storage unit 12. The control unit 11 may download the control program 12P from another computer via the network N. In this case, the control unit 11 stores the downloaded control program 12P in the storage unit 12. The control unit 11 may store the read control program 12P in the mass storage unit 14.

[0019] In this embodiment, the server 10 may be configured with multiple servers. The server 10 may be a virtual machine virtually constructed by software within a single device. The server 10 may be a local server installed within the facility where the server 10 is located. The server 10 may be a cloud server connected for communication via a network N. Furthermore, the control program 12P may be executed on a single server, or may be distributed and executed across multiple servers connected to each other via the network N.

[0020] FIG. 3 is an explanatory diagram showing an example of a record layout of the summary information DB. The summary information DB 14A stores summary information to be input into the language model M1. The summary information DB 14A includes a name column, a history column, a patient's intention column, a functional assessment column, and a living situation column. The name column stores the patient's name. The history column stores the patient's treatment history. The patient's treatment history is stored in chronological order for each item. The patient's intention column stores the patient's intention regarding treatment. The functional assessment column stores information regarding the patient's motor function, daily living function, or cognitive function. If there is no information regarding the patient's motor function, daily living function, or cognitive function, "No information currently listed" is stored. The living situation column stores information regarding the patient's daily life, such as the patient's family and level of care required. The summary information DB 14A may store a patient ID, password, etc. in association with the patient's name.

[0021] In the example of Figure 3, the record containing the name "Yamada Taro" stores the history "treatment history for smoking cessation," "follow-up observation of high LDL cholesterol," and "follow-up observation of depressive state," as well as the patient's intentions "expectations and wishes (wish to be cared for at home)," functional assessment "no information currently listed," and living situation "key person (daughter living with the patient)" and "care certification (care level 3)."

[0022] FIG. 4 is an explanatory diagram showing an example of a record layout of the record information DB. The record information DB 14B stores existing record information output by the language model M1 for each patient. The record information DB 14B includes a name column and a record information column. The name column stores the patient's name, as in the summary information DB 14A. The record information column stores the patient's existing record information. The existing record information is stored in a SOAP format including subjective information (S: Subjective data), objective information (O: Objective data), assessment (A: Assessment), and plan (P: Plan). The record information DB 14B may store a patient ID, password, etc. in association with the patient's name. In the example of FIG. 4, in a record in which the name "Yamada Taro" is stored, the existing record information for the name "Yamada Taro" is stored in a SOAP format.

[0023] FIG. 5 is a block diagram showing an example configuration of a computer. The computer 20 includes a control unit 21, a storage unit 22, a communication unit 23, an input unit 24, and a display unit 25. The above-mentioned units are interconnected via a bus. The control unit 21 is configured using one or more processors, such as a CPU, an MPU, or a GPU. The control unit 21 executes various information processing and control processes related to the computer 20 by appropriately executing a control program 22P and an application for implementing processing of the medical record system (hereinafter referred to as a medical record application AP) stored in the storage unit 22. The storage unit 22 includes RAM or ROM. The storage unit 22 stores various data, etc., required for the control program 22P (program product) executed by the control unit 21. The storage unit 22 also stores the medical record application AP. The storage unit 22 temporarily stores data, etc., generated when the control program 22P and the medical record application AP are executed.

[0024] The communication unit 23 is a communication module that transmits and receives information to and from the server 10 via the network N. The input unit 24 accepts operational inputs from medical personnel. The input unit 24 includes, for example, a keyboard, a mouse, and a microphone for voice input. The display unit 25 is a liquid crystal display or an organic EL display. The display unit 25 outputs various information in accordance with instructions from the control unit 21. The input unit 24 and the display unit 25 may be a touch panel configured as an integrated unit.

[0025] The process of outputting record information and summary information by the medical record system of this embodiment will be described below. Figures 6 and 7 are flowcharts showing an example of the processing procedure of the medical record system.

[0026] A medical professional instructs the computer 20 to start the medical record application AP. When the control unit 21 receives the instruction to start the medical record application AP via the input unit 24, the control unit 21 starts the medical record application AP and executes the following processing.

[0027] When the medical record app AP is launched, the control unit 21 displays an initial screen for inputting the patient's name on the display unit 25 (step S101). FIG. 8 is an explanatory diagram showing an example of the initial screen. The initial screen in FIG. 8 has a name input field I1 and a start button B0. The patient's name is input in the name input field I1. Instead of the patient's name, the patient's ID or the like may be input in the name input field I1. The start button B0 is, for example, a button for transmitting the patient's name to the server 10. The patient's name, "Yamada Taro," is displayed in the name input field I1 in FIG. 8.

[0028] The control unit 21 acquires the patient's name entered in the name input field I1 via the input unit 24 (step S102). The control unit 21 determines whether the start button B0 has been operated (step S103). If the control unit 21 determines that the start button B0 has not been operated (step S103: NO), the control unit 21 returns to the processing of step S102 and continues accepting the patient's name in the name input field I1. If the control unit 21 determines that the start button B0 has been operated (step S103: YES), the communication unit 23 transmits the patient's name to the server 10 (step S104). The control unit 21 may read a two-dimensional code, barcode, etc. associated with the patient using a code reader (not shown) or the like connected to the computer 20. In this case, the communication unit 23 transmits the read two-dimensional code, barcode, etc. of the patient to the server 10, rather than the patient's name.

[0029] The communication unit 13 receives the patient's name transmitted from the computer 20 (step S201). The control unit 11 reads summary information corresponding to the patient's name from the summary information DB 14A (step S202). The communication unit 13 transmits the read summary information to the computer 20 (step S203).

[0030] The communication unit 23 receives the summary information transmitted from the server 10 (step S105). The control unit 21 displays a setting screen for displaying the input information, the recorded information, and the summary information on the display unit 25 (step S106).

[0031] FIG. 9 is an explanatory diagram showing an example of a setting screen. The setting screen shown in FIG. 9 has an input field R0, a first interface R1, a second interface R2, and a third interface R3. The first interface R1, the second interface R2, and the third interface R3 are displayed in parallel from right to left. The input field R0 is displayed below the first interface R1, the second interface R2, and the third interface R3. The input field R0 accepts input information in one or more of the following ways: natural language (text data), voice, and document data (e.g., PDF documents, etc.). The input field R0 has a name field I2, a natural language field I3, a voice button B1, a document button B2, and a send button B3. The name field I2 displays the patient's name entered in the name input field I1 on the initial screen. The name field I2 in FIG. 9 displays the patient's name, "Yamada Taro."

[0032] The natural language field I3 accepts text input of input information via the input unit 24. The voice button B1 accepts voice input of input information. When the voice button B1 is selected, the control unit 21 collects sound using the microphone of the input unit 24 and records the conversation between the medical professional and the patient. The control unit 21 generates text data of the conversation by performing voice recognition processing on the recorded conversation. The control unit 21 may also generate text data of the conversation by performing similar processing on previously recorded conversation. The control unit 21 acquires the generated text data as input information. The document button B2 accepts input of document data. When the document button B2 is selected, the control unit 21 reads document data via the input unit 24. The control unit 21 generates text data of the document data by performing OCR (Optical Character Recognition) processing on the read document data. The control unit 21 acquires the generated text data as input information. The send button B3 is a button for sending the acquired input information to the server 10.

[0033] The first interface R1 displays input information acquired via the input field R0. The input information displayed in the first interface R1 of FIG. 9 includes the content of the conversation between the medical professional and the patient, as well as the medical professional's findings on the patient and treatment information for the patient. Furthermore, icons are displayed to distinguish the input information displayed in the first interface R1 from the method of acquisition: natural language, speech, or text data. The first interface R1 of FIG. 9 displays a set of a voice button B1 icon and speech-input information, and a set of a manual input icon (pencil icon) and manual input information. The second interface R2 displays record information output by the language model M1. Since FIG. 9 shows a state in which input information and summary information have not been provided to the language model M1, nothing is displayed in the second interface R2. When displaying record information in the second interface R2, the control unit 21 may also display the date and time when the record information was generated. The third interface R3 displays summary information read from the summary information DB 14A. The third interface R3 in FIG. 9 displays summary information corresponding to the patient's name "Yamada Taro."

[0034] 9 displays only the patient's treatment history from the summary information stored in the summary information DB 14A, but is not limited to this. In addition to the patient's treatment history, the third interface R3 may also display the patient's intention regarding the treatment plan, the patient's functional evaluation, and the patient's living situation.

[0035] The arrangement of the input field R0, the first interface R1, the second interface R2, and the third interface R3 is not limited to this. For example, the arrangement of the first interface R1 and the third interface R3 may be interchanged. Furthermore, the input field R0, the first interface R1, the second interface R2, and the third interface R3 may be arranged in order from top to bottom. Furthermore, the input field R0 may be arranged above the first interface R1, the second interface R2, and the third interface R3.

[0036] In the present embodiment, an example has been described in which the speech recognition processing and OCR processing are performed by the computer 20, but this is not limiting. The speech recognition processing and OCR processing may also be performed by the server 10. The processing in this case will be described below. The communication unit 23 transmits the conversation content (speech data) and document data received via the input field R0 to the server 10. The communication unit 13 receives the conversation content and document data transmitted from the computer 20. The control unit 11 generates text data of the conversation content by performing speech recognition processing on the conversation content. The control unit 11 also generates text data of the document data by performing OCR processing on the read document data. The communication unit 13 transmits the generated text data to the computer 20. The communication unit 23 receives the text data transmitted from the server 10. The display unit 25 displays the received text data as input information on the first interface R1.

[0037] The control unit 21 determines whether the input field R0 on the setting screen has received input information by one or more of text data, audio data, and document data (step S107). If the control unit 21 determines that no input information has been received (step S107: NO), it waits until it receives input information. If the control unit 21 determines that the input information has been received (step S107: YES), it displays the received input information on the first interface R1 (step S108). The control unit 21 determines whether the send button B3 has been operated (step S109). If the control unit 21 determines that the send button B3 has not been operated (step S109: NO), it returns to the processing of step S107 and continues receiving input information for the input field R0. If the control unit 21 determines that the send button B3 has been operated (step S109: YES), the communication unit 23 transmits the input information to the server 10 (step S110).

[0038] The communication unit 13 receives the input information transmitted from the computer 20 (step S204). The control unit 11 reads existing record information corresponding to the patient's name from the record information DB 14B (step S205). The control unit 11 determines a first prompt including the summary information read in step S202, the input information, an instruction for causing the language model M1 to output the record information, and the existing record information read in step S205 (step S206).

[0039] In the case of the first time for each subject, since existing record information is not stored in the record information DB 14B, the control unit 11 does not need to perform the process of step S205. In this case, the control unit 11 determines a first prompt including the summary information read in step S202, the input information, and an instruction for causing the language model M1 to output the record information.

[0040] FIG. 10 is an explanatory diagram showing an example of a first prompt. The first prompt 100 includes an existing summary information field 101, an input information field 102, a first instruction field 103, and an existing record information field 104. The existing summary information field 101 lists existing summary information for each subject read from the summary information DB 14A. The existing summary information field 101 in FIG. 10 lists summary information displayed in the third interface R3 in FIG. 9. The existing summary information field 101 in FIG. 10 lists, among the summary information, a patient's treatment history, but is not limited to this. In addition to the patient's treatment history, the existing summary information field 101 may also list peripheral information about the patient. The input information field 102 lists input information acquired via the input field R0 in the setting screen. The input information field 102 in FIG. 10 lists input information displayed in the first interface R1 in FIG. 9. The first instruction field 103 lists instructions for causing the language model M1 to output record information. The first instruction field 103 in Fig. 10 states, "Based on the above summary information and the input information, modify the existing record information and organize it in SOAP format." The existing record information field 104 contains existing record information read from the record information DB 14B. The existing record information field 104 in Fig. 10 contains existing record information corresponding to the patient's name "Yamada Taro." For the first visit for each subject, the existing record information field 104 is left blank.

[0041] The control unit 11 acquires the recorded information by inputting the determined first prompt into the language model M1 (step S207). The control unit 11 updates the existing recorded information stored in the recorded information DB 14B based on the recorded information acquired in step S207 (step S208). The communication unit 13 transmits the acquired recorded information to the computer 20 (step S209).

[0042] The communication unit 23 receives the record information transmitted from the server 10 (step S111). The display unit 25 displays the received record information on the second interface R2 (step S112).

[0043] FIG. 11 is an explanatory diagram showing an example of a setting screen. The display format of FIG. 11 is the same as that of FIG. 9. The second interface R2 of FIG. 11 displays the record information output by the language model M1 and an Apply button B4. The record information is displayed in SOAP format including four items: subjective information (S), objective information (O), evaluation (A), and plan (P). The Apply button B4 is displayed when record information is received from the server 10. The Apply button B4 is a button for instructing the server 10 to generate summary information.

[0044] The control unit 21 determines whether the Apply button B4 has been operated (step S113). If the control unit 21 determines that the Apply button B4 has not been operated (step S113: NO), the process returns to step S112 and the setting screen continues to be displayed. If the control unit 21 determines that the Apply button B4 has been operated (step S113: YES), the communication unit 23 transmits an instruction to generate summary information (hereinafter referred to as a generation instruction) to the server 10 (step S114).

[0045] The communication unit 13 receives the generation instruction sent from the computer 20 (step S210). The control unit 11 determines a second prompt including the record information acquired in step S207, an instruction for causing the language model M1 to output summary information, and the existing summary information read in step S202 (step S211).

[0046] FIG. 12 is an explanatory diagram showing an example of a second prompt. The second prompt 200 includes a record information field 201, a second instruction field 202, and an existing summary information field 203. The record information field 201 contains record information output by the language model M1. The record information field 201 in FIG. 12 contains record information displayed in the second interface R2 in FIG. 11. The second instruction field 202 contains instructions for causing the language model M1 to output summary information. The second instruction field 202 in FIG. 12 contains the message, "Modify and organize the existing summary information by summarizing the record information and the existing summary information." The existing summary information field 203 contains existing summary information read from the summary information DB 14A. The existing summary information field 203 in FIG. 12 contains existing summary information corresponding to the patient's name, "Yamada Taro."

[0047] The control unit 11 inputs the determined second prompt into the language model M1 to acquire summary information (step S212). The control unit 11 updates the existing summary information stored in the record information DB 14B based on the summary information acquired in step S212 (step S213). The communication unit 13 transmits the summary information acquired in step S212 to the computer 20 (step S214).

[0048] The communication unit 23 receives the summary information transmitted from the server 10 (step S115). The display unit 25 displays the received summary information on the third interface R3 (step S116).

[0049] FIG. 13 is an explanatory diagram showing an example of a setting screen. The display format of FIG. 13 is the same as that of FIG. 11. The third interface R3 of FIG. 13 displays summary information output by the language model M1. Differences between the summary information before the update ( FIG. 11 ) and the summary information after the update ( FIG. 13 ) are underlined. Differences between the summary information before the update and the summary information after the update may be highlighted (e.g., in bold or with hatching). Furthermore, even if there are differences between the summary information before the update and the summary information after the update, underlining or the like may not be used.

[0050] According to the first embodiment, the medical record system can acquire input information and summary information, and output record information based on the acquired input information and summary information.

[0051] According to the first embodiment, the medical record system can output record information in SOAP format.

[0052] According to the first embodiment, the medical record system can update the summary information based on the output record information.

[0053] According to the first embodiment, the medical record system can output the first interface, the second interface, and the third interface in association with each other.

[0054] (Embodiment 2) In embodiment 2, a method is described in which a medical record system receives additional information for input information and re-outputs record information based on the received additional information, input information, and summary information. The additional information is, for example, a supplement to the input information. A description of the same processing as in embodiment 1 is omitted.

[0055] Fig. 14 is a flowchart showing an example of the processing procedure of the medical record system according to embodiment 2. In the flowchart shown in Fig. 14, step S113 and the following steps in the processing shown in Fig. 6 to Fig. 7 are replaced with steps S301 to S306 and steps S401 to S405. Explanations of steps similar to those in Fig. 6 to Fig. 7 will be omitted.

[0056] The display unit 25 displays the received recorded information on the second interface R2 (step S112). The control unit 21 determines whether the Apply button B4 has been operated (step S113). If the control unit 21 determines that the Apply button B4 has been operated (step S113: YES), the process proceeds to step S114. If the control unit 21 determines that the Apply button B4 has not been operated (step S113: NO), the control unit 21 determines whether additional information has been accepted via the input field R0 (step S301). If the control unit 21 determines that additional information has not been accepted (step S301: NO), the process returns to step S112. If the control unit 21 determines that additional information has been accepted (step S301: YES), the display unit 25 displays the accepted additional information and the input information on the first interface R1 (step S302).

[0057] 15 is an explanatory diagram showing an example of a setting screen according to the second embodiment. In the first interface R1, additional information input via the input field R0 is displayed together with the input information. In the example of FIG. 15, the additional information "Runny nose +, cough ++" and "Appears to be gaining confidence in quitting smoking" are displayed with underlining. The additional information may also be displayed without underlining.

[0058] The control unit 21 determines whether the send button B3 has been operated (step S303). If the control unit 21 determines that the send button B3 has not been operated (step S303: NO), the process returns to step S302. If the control unit 21 determines that the send button B3 has been operated (step S303: YES), the communication unit 23 transmits the additional information to the server 10 (step S304).

[0059] The communication unit 13 receives the additional information transmitted from the computer 20 (step S401). The control unit 11 determines an additional prompt including the summary information read in step S202, the input information received in step S204, the additional information, an instruction to cause the language model M1 to re-output the recorded information, and the existing recorded information read in step S205 (step S402).

[0060] FIG. 16 is an explanatory diagram showing an example of an addition prompt. The addition prompt 110 includes an existing summary information field 111, an input information field 112, an additional information field 113, an addition instruction field 114, and an existing record information field 115. The existing summary information field 111 lists summary information for each subject read from the summary information DB 14A. The existing summary information field 111 of FIG. 16 lists summary information displayed in the third interface R3 of FIG. 15. The input information field 112 lists input information acquired via the input field R0 in the setting screen. The input information field 112 of FIG. 16 lists input information displayed in the first interface R1 of FIG. 15. The additional information field 113 lists additional information acquired via the input field R0 in the setting screen. The additional information field 113 of FIG. 16 lists additional information displayed in the first interface R1 of FIG. 15. The additional instruction field 114 lists instructions for causing the language model M1 to re-output the record information. The additional instruction field 114 in Figure 16 states, "Please revise the record information again based on the above summary information, input information, and additional information, and organize it in SOAP format." The existing record information field 115 states existing record information read from the record information DB 14B. The existing record information field 115 in Figure 16 states existing record information corresponding to the patient's name "Yamada Taro."

[0061] The configuration of the addition prompt 110 is not limited to that shown in Fig. 16. The addition prompt 110 may include a record information column instead of the existing summary information column 111, the input information column 112, and the existing record information column 115. In this case, the addition prompt 110 includes the updated record information, the additional information, and an instruction to cause the language model M1 to re-output the record information, and the addition instruction column 114 states, "Based on the above record information and the additional information, please revise the record information again and organize it in SOAP format."

[0062] The control unit 11 inputs the determined additional prompt into the language model M1 to reacquire the record information (step S403). The control unit 11 updates the record information stored in the record information DB 14B based on the record information reacquired in step S403 (step S404). The communication unit 13 transmits the reacquired record information to the computer 20 (step S405).

[0063] The communication unit 23 receives the record information transmitted from the server 10 (step S305). The display unit 25 displays the received record information on the second interface R2 (step S306).

[0064] Fig. 17 is an explanatory diagram showing an example of a setting screen according to the second embodiment. The second interface R2 in Fig. 17 displays record information re-output by the language model M1 and an Apply button B4. Portions of the re-output record information that differ from the previous record information (Fig. 15) are underlined. Portions that differ from the previous record information may also be displayed without underlining.

[0065] According to the second embodiment, the medical record system can accept additional information and re-output record information based on the accepted additional information, input information, and summary information.

[0066] According to the second embodiment, the medical record system can display the record information taking into consideration the additional information by re-outputting the record information.

[0067] (Embodiment 3) Embodiment 3 describes a method in which a medical record system acquires correction information for record information and re-outputs the record information based on the acquired correction information and record information. The correction information is, for example, a statement for correcting the record information. Descriptions of processes similar to those in Embodiments 1 and 2 will be omitted.

[0068] Figures 18 and 19 are flowcharts showing an example of the processing procedure of the medical record system according to embodiment 3. In the flowcharts shown in Figures 18 and 19, step S113 and the following steps in the processing shown in Figures 6 and 7 are replaced with steps S501 to S511 and steps S601 to S610. Explanations of steps similar to those in Figures 6 and 7 will be omitted.

[0069] The display unit 25 displays the received record information on the second interface R2 (step S112). The control unit 21 determines whether the Apply button B4 has been operated (step S113). If the control unit 21 determines that the Apply button B4 has been operated (step S113: YES), the process proceeds to step S114. If the control unit 21 determines that the Apply button B4 has not been operated (step S113: NO), the control unit 21 acquires correction information via the input unit 24 (step S501). The display unit 25 displays the acquired correction information on the second interface R2 (step S502).

[0070] FIG. 20 is an explanatory diagram showing an example of a setting screen according to the third embodiment. The second interface R2 in FIG. 20 includes a save button B5 and a delete button B6. The save button B5 is a button for accepting the acquired correction information. The delete button B6 is a button for deleting the acquired correction information. The second interface R2 in FIG. 20 also displays the correction information acquired via the input unit 24 together with the record information. In the example of FIG. 20 , the additional information "Please write about stress" is underlined and displayed below the evaluation (A) of the record information. The additional information "Please list all medications" is underlined and displayed in the middle of the plan (P) of the record information. The additional information may be displayed without underlining.

[0071] The control unit 21 determines whether the Delete button B6 has been operated (step S503). If the control unit 21 determines that the Delete button B6 has been operated (step S503: YES), it deletes the correction information acquired in step S501 and returns to the process of step S112. If the control unit 21 determines that the Delete button B6 has not been operated (step S503: NO), it determines whether the Save button B5 has been operated (step S504). If the control unit 21 determines that the Save button B5 has not been operated (step S504: NO), it returns to the process of step S502. If the control unit 21 determines that the Save button B5 has been operated (step S504: YES), the communication unit 23 transmits the correction information to the server 10 (step S505).

[0072] The communication unit 13 receives the correction information transmitted from the computer 20 (step S601). The control unit 11 determines a correction prompt including the record information updated in step S208, an instruction for causing the language model M1 to correct the record information, and the correction information (step S602).

[0073] FIG. 21 is an explanatory diagram showing an example of a correction prompt. The correction prompt 120 includes a record information field 121, a correction instruction field 122, and a correction information field 123. The record information field 121 describes the record information output by the language model M1. The record information field 121 in FIG. 21 describes the record information displayed in the second interface R2 of FIG. 17. The correction instruction field 122 describes instructions for causing the language model M1 to re-output the record information. The correction instruction field 122 in FIG. 21 states, "Based on the correction information below, correct the record information and organize it in SOAP format." The correction information field 123 describes the correction information acquired via the input unit 24. The correction information field 123 in FIG. 21 describes the correction information displayed in the second interface R2 of FIG. 20, "Please write about stress. Please list all medications."

[0074] The control unit 11 inputs the determined correction prompt into the language model M1 to reacquire the record information (step S603). The control unit 11 updates the record information stored in the record information DB 14B based on the record information reacquired in step S603 (step S604). The communication unit 13 transmits the reacquired record information to the computer 20 (step S605).

[0075] The communication unit 23 receives the record information transmitted from the server 10 (step S506). The display unit 25 displays the received record information on the second interface R2 (step S507).

[0076] Fig. 22 is an explanatory diagram showing an example of a setting screen according to the third embodiment. The display format of Fig. 22 is the same as that of Fig. 20. The second interface R2 of Fig. 22 displays the record information re-output by the language model M1, an Apply button B4, a Save button B5, and an Erase button B6. Portions of the re-output record information that differ from the previous record information (Fig. 20) are underlined. Portions that differ from the previous record information may also be displayed without underlining.

[0077] The control unit 21 determines whether the Apply button B4 has been operated (step S508). If the control unit 21 determines that the Apply button B4 has not been operated (step S508: NO), the process returns to step S507 and the setting screen continues to be displayed. If the control unit 21 determines that the Apply button B4 has been operated (step S508: YES), the communication unit 23 transmits an instruction to regenerate the summary information (hereinafter, a regeneration instruction) to the server 10 (step S509).

[0078] The communication unit 13 receives the regeneration instruction sent from the computer 20 (step S606). The control unit 11 determines a third prompt including the summary information updated in step S213, an instruction to cause the language model M1 to regenerate the summary information, and the record information acquired again in step S603 (step S607).

[0079] FIG. 23 is an explanatory diagram showing an example of a third prompt. The third prompt 300 includes a record information field 301, a third instruction field 302, and a summary information field 303. The record information field 301 contains the record information re-output by the language model M1. The record information field 301 in FIG. 23 contains the record information displayed in the second interface R2 in FIG. 22. The third instruction field 302 contains an instruction for causing the language model M1 to re-output summary information. The third instruction field 302 in FIG. 23 contains the message, "Re-revise and organize the summary information by summarizing the record information and summary information." The summary information field 303 contains summary information read from the summary information DB 14A. The summary information field 303 in FIG. 23 contains summary information corresponding to the patient's name, "Yamada Taro."

[0080] The control unit 11 inputs the determined third prompt into the language model M1 to reacquire the summary information (step S608). The control unit 11 updates the summary information stored in the summary information DB 14A based on the summary information reacquired in step S608 (step S609). The communication unit 13 transmits the reacquired summary information to the computer 20 (step S610).

[0081] The communication unit 23 receives the summary information transmitted from the server 10 (step S510). The display unit 25 displays the received summary information on the third interface R3 (step S511).

[0082] Fig. 24 is an explanatory diagram showing an example of a setting screen according to the third embodiment. The third interface R3 in Fig. 24 displays summary information re-output by the language model M1. Differences between the summary information before the update (Fig. 22) and the summary information after the update (Fig. 24) are underlined. Differences between the summary information before the update and the summary information after the update do not have to be underlined.

[0083] In the present embodiment, an example has been described in which the input information, record information, and summary information displayed in each interface are switched for each situation (e.g., when correction information is acquired), but this is not limited thereto. Previously displayed input information, record information, and summary information (e.g., the information in FIGS. 20 and 22) may be formatted so that they can be viewed in chronological order. In this case, a scroll bar (not shown) is provided at the right end of the first interface R1 or the left end of the third interface R2. The user can view the history of input information from the past to the present by moving the scroll bar up and down via the input unit 24.

[0084] According to the third embodiment, the medical record system can acquire correction information and re-output record information based on the acquired correction information.

[0085] According to the third embodiment, the medical record system can display record information that takes into account correction information by re-outputting the record information.

[0086] According to the third embodiment, the medical record system can re-output summary information based on the re-output record information.

[0087] Fourth Embodiment In a fourth embodiment, an example will be described in which the input information further includes image data showing the affected area of ​​the patient. Description of the same processes as those in the first, second, and third embodiments will be omitted.

[0088] FIG. 25 is an explanatory diagram showing an example of the configuration of a second language model. The storage unit 12 of the fourth embodiment stores a second language model M10 instead of the language model M1. The second language model M10 receives input information including image data, summary information, an instruction to generate record information, and existing record information as input data, performs calculations to generate record information according to the content of the input data, and outputs the generated record information. The image data is, for example, image data showing the affected area of ​​a patient or image data showing the patient's facial expression. The instruction to generate record information is, for example, "Using the summary information and the input information, organize the existing record information into SOAP format."

[0089] The second language model M10 has a feature extraction layer M2 and a language model M1. The language model M1 is the same as in the first embodiment. The feature extraction layer M2 includes convolutional layers and pooling layers corresponding to multiple types of input information (e.g., text data and image data). The feature extraction layer M2 extracts features from the text data (input information) by using a convolutional layer and pooling layer corresponding to the text data. The feature extraction layer M2 extracts features from the image data (input information) by using a convolutional layer and pooling layer corresponding to the image data.

[0090] The second language model M10 combines the extracted features to synthesize an input sequence involving the input information, summary information, instructions for generating record information, and existing record information. The second language model M10 inputs the synthesized input sequence to the language model M1. The language model M1 performs a calculation to generate record information from the synthesized input sequence, and outputs the generated record information as output data of the second language model M10.

[0091] The second language model M10 is not limited to the configuration shown in Fig. 25. For example, the second language model M10 may not include the feature extraction layer M2, and may instead be configured such that the input information, summary information, instructions for generating record information, and existing record information are directly input to the language model M1.

[0092] Fig. 26 is a flowchart showing an example of the processing procedure of the medical record system according to the fourth embodiment. In the flowchart shown in Fig. 26, steps S701 to S705 and steps S801 to S807 are added after step S105 in the processing shown in Fig. 6 to Fig. 7. Explanations of steps similar to those in Fig. 6 to Fig. 7 will be omitted. The control unit 21 displays a setting screen for displaying input information, record information, and summary information on the display unit 25 (step S701).

[0093] Fig. 27 is an explanatory diagram showing an example of a setting screen according to the fourth embodiment. The input field R0 in Fig. 27 accepts image data in addition to natural language (text data), voice, or document data. The input field R0 has a camera button B7. The camera button B7 accepts image data acquired from an external imaging device (not shown). The control unit 21 acquires the accepted image data as input information.

[0094] The first interface R1 in Fig. 27 displays a voice button icon, voice input information, a manual input icon, manual input information, as well as a camera button icon and image data input information. The image data input information in Fig. 27 is image data showing the affected area (throat) of the patient.

[0095] The control unit 21 determines whether the input field R0 in the setting screen has received input information by any one or more of the following means: natural language, voice, text data, and image data (step S702). If the control unit 21 determines that no input information has been received (step S702: NO), it waits until it receives input information. If the control unit 21 determines that the input information has been received (step S702: YES), the display unit 25 displays the received input information on the first interface R1 (step S703). The control unit 21 determines whether the send button B3 has been operated (step S704: NO). If the control unit 21 determines that the send button B3 has not been operated (step S704: NO), it returns to the processing of step S702 and continues receiving input information for the input field R0. If the control unit 21 determines that the send button B3 has been operated (step S704: YES), the communication unit 23 transmits the input information to the server 10 (step S705).

[0096] The communication unit 13 receives input information transmitted from the computer 20 (step S801). The control unit 11 reads existing record information corresponding to the patient's name from the record information DB 14B (step S802). The control unit 11 inputs the summary information, input information, record information generation instruction, and existing record information read in step S202 to the second language model M10 (step S803). The second language model M10 extracts features for each of the summary information, input information, record information generation instruction, and existing record information using the feature extraction layer M2 (step S804). The second language model M10 combines the extracted features to synthesize an input sequence relating to the input information, summary information, record information generation instruction, and existing record information (step S805). The second language model M10 inputs the synthesized input sequence to the language model M1 to acquire the record information output by the language model M1 (step S806). The control unit 11 updates the existing record information stored in the record information DB 14B based on the record information acquired in step S806 (step S807).The control unit 11 advances the process to step S209.

[0097] In the fourth embodiment, the control unit 11 outputs the record information using the second language model M10. However, the present invention is not limited to this. The control unit 11 may output the record information using a multimodal LLM such as GPT-4. In this case, the multimodal LLM includes multiple input layers and intermediate layers for extracting features from various input data such as image data and text data. The multimodal LLM outputs the record information by combining the features extracted from the multiple intermediate layers in a fully connected layer before the output layer.

[0098] The control unit 11 may output the record information using an image diagnosis AI and a language model M1. The image diagnosis AI is a learning model that uses, for example, an image of the patient's affected area (e.g., an image of the patient's throat) as input data and information indicating the patient's condition (e.g., normal, clear tonsils, or herpes simplex stomatitis) as output data. The control unit 11 acquires information indicating the patient's condition by inputting the acquired image data (input information) into the image diagnosis AI. The control unit 11 acquires the record information by inputting the information indicating the patient's condition and text data (input information) into the language model M1.

[0099] According to the fourth embodiment, the medical record system can acquire input information including image data and summary information, and output record information based on the acquired input information and summary information.

[0100] (Embodiment 5) Embodiment 5 describes a method in which a medical record system outputs record information regarding a patient's treatment by providing input information indicating the content of a dialogue entered through a smartphone and patient summary information corresponding to identification information for identifying the patient to a language model.

[0101] 28 is an explanatory diagram showing an example of the configuration of a medical record system according to embodiment 5. The medical record system includes a server 10, a computer 20, and an information processing terminal 30. The configurations of the server 10 and the computer 20 are the same as those in embodiment 1. The medical record system transmits and receives information via a network N. The information processing terminal 30 is, for example, a server device, a smartphone, a tablet, a computer, or a general-purpose tablet PC. In this embodiment, the information processing terminal 30 will be described as a smartphone 30.

[0102] The smartphone 30 includes a control unit 31, a storage unit 32, a communication unit 33, an input unit 34, and a display unit 35. The control unit 31 is configured using one or more processors, such as a CPU, an MPU, or a GPU. The control unit 31 executes various information processing and control processes related to the smartphone 30 by appropriately executing a control program 32P and a medical record app AP stored in the storage unit 32. The storage unit 32 includes RAM, ROM, etc. The storage unit 32 stores various data, etc. necessary for the control program 32P (program product) executed by the control unit 31. The storage unit 32 also stores the medical record app AP. The storage unit 32 temporarily stores data, etc. generated when the control program 32P and the medical record app AP are executed.

[0103] The communication unit 33 is a communication module that transmits and receives information between the server 10 and the computer 20 via the network N. The input unit 34 accepts operational inputs from medical professionals. The input unit 34 includes, for example, a keyboard, a mouse, and a microphone for voice input. The display unit 35 is a liquid crystal display or an organic EL display. The display unit 35 outputs various information in accordance with instructions from the control unit 31. The input unit 34 and the display unit 35 may be a touch panel configured as an integrated unit.

[0104] The processing of embodiment 5 will be described. Fig. 29 is a flowchart showing an example of the processing procedure of the medical record system according to embodiment 5. A medical professional instructs the smartphone 30 to start the medical record app AP. When the control unit 31 receives the instruction to start the medical record app AP via the input unit 34, the control unit 31 starts the medical record app AP and executes the following processing.

[0105] When the medical record app AP is launched, the display unit 35 displays a screen for accepting input of input information indicating the content of the conversation and identification information for identifying the patient (step S901). The input information indicating the content of the conversation includes audio data recording the content of the conversation between the patient, medical staff, etc. (hereinafter referred to as audio data of the conversation content), text data describing the content of the conversation between the patient, medical staff, etc., and image (video) data recording the content of the conversation between the patient, medical staff, etc. The identification information for identifying the patient includes the patient's name, patient ID, and date of birth.

[0106] FIG. 30 is an explanatory diagram showing an example of a screen. The screen d1 shown in FIG. 30 includes, from top to bottom, an identification information area E1, an input information area E2, and a menu area E3. The menu area E3 accepts selection of multiple menus by a medical professional or the like. The multiple menus include a recording menu, a WF (workflow) execution menu, and a setting menu. When the selection of the recording menu is accepted, a screen (not shown) for acquiring voice data (input information) of the dialogue content in real time is displayed. When the selection of the setting menu is accepted, a screen (not shown) for changing the settings of the medical record application AP is displayed. When the selection of the WF (workflow) execution menu is accepted, a screen for transmitting input information indicating the dialogue content and identification information for identifying the patient to the server 10 is displayed. The screen d1 shown in FIG. 30 is an example in which the "WF execution menu" is selected by a medical professional.

[0107] The identification information area E1 accepts input of patient identification information by a medical professional or the like. The identification information area E1 may be a pull-down menu button. In this case, the identification information area E1 displays patient identification information stored in advance in the storage unit 32. The input information area E2 displays input information indicating the content of the dialogue, including voice data indicating the content of the dialogue, text data indicating the content of the dialogue, and image (video) data indicating the content of the dialogue. The input information indicating the content of the dialogue displayed in the input information area E2 is stored in advance in the storage unit 32.

[0108] The control unit 31 accepts input or selection of input information indicating the patient's identification information and the content of the dialogue via the input unit 34. The control unit 11 may accept the input of the identification information and the input information by reading a one-dimensional code (barcode) or a two-dimensional code (QR Code (registered trademark)) using a camera (not shown) provided in the smartphone 30. In the identification information area E1 of the screen d1 shown in FIG. 30 , the patient's name "Yamada Taro" is entered as an example of identification information for identifying the patient, and in the input information area E2, audio data of the dialogue content "20241221_174805.m4a" is entered or selected as an example of input information indicating the content of the dialogue. The execute button B8 is a button for transmitting the input information indicating the content of the dialogue and the identification information for identifying the patient entered via the smartphone to the server 10.

[0109] In the fifth embodiment, an example has been shown in which the identification information for identifying the patient and the input information indicating the dialogue content are input or selected on one screen (screen d1) without screen transitions, etc., but this is not limiting. The identification information for identifying the patient and the input information indicating the dialogue content may be input or selected on multiple screens (not shown) by screen transitions. In this case, the control unit 31 accepts the input or selection of the identification information for identifying the patient and the input information indicating the dialogue content by sequentially displaying multiple screens.

[0110] The control unit 31 acquires input information indicating the dialogue content input via the screen d1 and identification information for identifying the patient (step S902). When the execute button B8 is selected, the communication unit 33 transmits the acquired input information and identification information to the server 10 (step S903).

[0111] The communication unit 13 receives input information indicating the content of the conversation entered via the smartphone and identification information for identifying the patient (step S1001). If the received input information is voice data, the control unit 11 converts the voice data into text data. The control unit 11 reads summary information of the patient corresponding to the received identification information from the summary information DB 14A (step S1002). Specifically, if the received identification information is "Yamada Taro," the control unit 11 refers to the summary information DB 14A and reads the summary information of the patient from a record in which "Yamada Taro" is stored in the name column.

[0112] The control unit 11 reads existing record information corresponding to the received identification information from the record information DB 14B (step S1003). The control unit 11 determines a first prompt including the received input information, the patient summary information read in step S1002, the existing record information read in step S1003, and an instruction to cause the language model M1 to output the record information (step S1004). The configuration of the first prompt is the same as in the first embodiment. The instruction to cause the language model M1 to output the record information is stored in advance in the storage unit 12. If the existing record information is not stored in the record information DB 14B, the existing record information does not need to be included in the first prompt.

[0113] The control unit 11 acquires the recorded information by inputting the determined first prompt into the language model M1 (step S1005). The control unit 11 replaces the existing recorded information with the recorded information acquired in step S1005, thereby updating the existing recorded information stored in the recorded information DB 14B (step S1006). The communication unit 13 transmits the acquired recorded information to the computer 20 (step S1007).

[0114] The communication unit 23 receives the record information transmitted from the server 10 (step S1101). The display unit 25 displays the received record information on the display unit 25 (step S1102). Specifically, the display unit 25 displays the received record information on the second interface R2 of the setting screen.

[0115] In embodiment 5, the medical record system receives input information indicating the content of the dialogue entered via a smartphone and identification information for identifying the patient, and outputs record information regarding the patient's treatment by providing the received input information and summary information of the patient corresponding to the identification information to a language model.

[0116] (Embodiment 6) Embodiment 6 describes how a medical record system outputs record information about a patient's treatment and updates patient summary information according to the type of file received.

[0117] 31 is a flowchart showing an example of a processing procedure of the medical record system according to embodiment 6. The control unit 21 determines whether or not a file transmitted from a facsimile machine, an external device, a smartphone 30, or the like has been received (step S1201). The types of the received file include, for example, a first type and a second type.

[0118] FIG. 32 is an explanatory diagram showing an example of a first-type file. FIG. 32 shows a home nursing report, which is an example of a first-type file. A home nursing report is a report created, for example, when a medical professional provides home nursing care to a patient's attending physician to report the details of the care provided. The home nursing report includes input information and patient identification information. The home nursing report of FIG. 32 includes multiple items and corresponding entries for each item. The multiple items include, for example, the user's name, date of birth, status of nursing care certification, address, date of visit, progress of illness, content of nursing care, status of nursing care at home, amount and usage status of hygiene products, etc., amount and usage status of hygiene products, etc., changes in type and amount of hygiene products, etc., matters to be noted, and the creator.

[0119] The first type of file may also include a vital sign report, which is a report that records vital signs (e.g., pulse, respiration, body temperature, blood pressure, etc.) measured to understand the patient's physical condition and details related to the vital signs, and includes input information, patient identification information, etc.

[0120] FIG. 33 is an explanatory diagram showing an example of a file that falls under the second type. FIG. 33 shows a medical information report, which is an example of a file that falls under the second type. The medical information report is, for example, a document created by a current medical professional when a patient receives treatment at another medical institution, and includes input information, patient identification information, etc. The medical information report of FIG. 33 includes multiple items and corresponding entries for each item. The multiple items include, for example, the patient's name, patient address, date of birth, details, progress during hospitalization, current prescription, and post-discharge precautions.

[0121] The second type of file may also include a home care service plan, which is a document that includes, for example, the name of the patient who will be receiving home care services such as visiting nursing care, the wishes of the patient and their family regarding their daily lives, and a comprehensive assistance policy, and includes input information and patient identification information.

[0122] If the control unit 21 determines that a file has not been received from a facsimile machine, an external device, the smartphone 30, or the like (step S1201: NO), it waits until the file is received. If the control unit 21 determines that a file has been received from a facsimile machine, an external device, the smartphone 30, or the like (step S1201: YES), it identifies identification information for identifying a patient corresponding to the file (step S1202). Specifically, the control unit 21 identifies the identification information corresponding to the received file by displaying a screen for inputting identification information for identifying a patient, or by reading a one-dimensional code or a two-dimensional code using a camera, a barcode reader, or the like. The communication unit 23 transmits the received file and the identification information corresponding to the file to the server 10 (step S1203).

[0123] The communication unit 13 receives a file transmitted from the computer 20 and identification information corresponding to the file (step S1301). The control unit 11 identifies the type of the received file by comparing the subject of the received file with the names of files corresponding to the first or second type pre-stored in the storage unit 12 (step S1302). The subject of the received file is, for example, a name indicating the type of the file, written in a predetermined position within the file (e.g., at the top in Figures 32 and 33). The control unit 11 may identify the type of the received file using the file name instead of the subject of the received file. The file name is, for example, a name associated with the received file for identifying the file.

[0124] When the files corresponding to the first type and the files corresponding to the second type are created by different creators, the control unit 11 may identify the type of the received file by comparing the creators of each file pre-stored in the storage unit 12 with the creator of the file transmitted from the computer 20. When the files corresponding to the first type and the files corresponding to the second type are created on different creation dates (for example, every Friday or the first Monday of every month), the control unit 11 may identify the type of the received file by comparing the creation dates of each file pre-stored in the storage unit 12 with the creation date of the file transmitted from the computer 20.

[0125] When the first type file and the second type file are created in different file formats (e.g., PDF, Word, Excel, etc.), the control unit 11 may identify the type of the received file by comparing the file formats of each file pre-stored in the storage unit 12 with the format of the file transmitted from the computer 20. When the control unit 11 receives a received file, the control unit 11 may identify the type of the received file by using a learning model that outputs the type of the received file. In this case, the control unit 11 generates the learning model using a large amount of training data in which multiple files used in medical settings are associated with the file types.

[0126] The control unit 11 may identify the type of the received file by using the language model M1. The process in this case will be described below. The control unit 11 reads the description of the received file through OCR processing. The control unit 11 determines a prompt including the read description and an instruction for identifying the type of the received file. The instruction for identifying the type of the received file is, for example, an instruction to the language model M1 to identify the type of the received file from the description read through OCR processing, such as "Based on the description of the received file, identify the type of the received file" or "Based on the description of the received file, answer whether the type of the received file is a first type or a second type." The instruction for identifying the type of the received file is stored in advance in the storage unit 12 or the mass storage unit 14. The control unit 11 identifies the type of the received file by inputting the prompt including the read description and an instruction for identifying the type of the received file into the language model M1.

[0127] The control unit 11 determines whether the identified type is a first type (e.g., a home nursing report or a vital sign report) (step S1303). If the control unit 11 determines that the identified type is the first type (step S1303: YES), it extracts input information from the received file (step S1304). In the example of FIG. 32, the control unit 11 uses OCR processing to read multiple items and corresponding descriptions from the home nursing report (a file corresponding to the first type). From the multiple items and corresponding descriptions read, the control unit 11 extracts, as input information, descriptions corresponding to the specified items: "Progression of the illness," "Contents of nursing care," "Nursing status at home," "Amount and usage status of sanitary materials, etc.", and "Changes in the type and amount of sanitary materials, etc."

[0128] When a home nursing report is input, the control unit 11 may extract input information from the home nursing report by using a learning model that outputs, as input information, entries corresponding to predetermined items in the home nursing report (e.g., entries corresponding to "progression of illness"). In this case, the control unit 11 generates the learning model using a large amount of training data that associates multiple home nursing reports with text data indicating entries corresponding to predetermined items in the home nursing report.

[0129] The control unit 11 reads summary information of the patient corresponding to the received identification information from the summary information DB 14A (step S1305). The control unit 11 reads existing record information corresponding to the received identification information from the record information DB 14B (step S1306). The control unit 11 determines a first prompt including the input information extracted in step S1304, the summary information of the patient read in step S1305, the existing record information read in step S1306, and an instruction to cause the language model M1 to output the record information (step S1307). The configuration of the first prompt is the same as in the first embodiment. The instruction to cause the language model M1 to output the record information is stored in advance in the storage unit 12. If the existing record information is not stored in the record information DB 14B, the existing record information does not need to be included in the first prompt.

[0130] The control unit 11 acquires the recorded information by inputting the determined first prompt into the language model M1 (step S1308). The control unit 11 replaces the existing recorded information with the recorded information acquired in step S1308, thereby updating the existing recorded information stored in the recorded information DB 14B (step S1309). The communication unit 13 transmits the updated recorded information to the computer 20 (step S1310).

[0131] The communication unit 23 receives the recording information transmitted from the server 10 (step S1204). The control unit 21 displays the received recording information on the display unit 25 (step S1205). Specifically, the display unit 25 displays the received recording information on the second interface R2 of the setting screen. The communication unit 13 may transmit updated recording information to the smartphone 30. In this case, the communication unit 33 receives the recording information transmitted from the server 10. The control unit 31 displays the received storage information on the display unit 35.

[0132] When the control unit 11 determines that the identified type is not the first type (it is the second type (medical information report or home service plan)) (step S1303: NO), it extracts input information from the received file (step S1311). In the example of FIG. 33, the control unit 11 reads, by OCR processing, multiple items and corresponding descriptions written in the medical information report (file corresponding to the second type). From the multiple items and corresponding descriptions read, the control unit 11 extracts, as input information, descriptions corresponding to the predetermined items "Content," "Progress during hospitalization," "Current prescribed medication," and "Notes after discharge."

[0133] When a medical information report is input, the control unit 11 may extract input information from the medical information report by using a learning model that outputs, as input information, entries corresponding to predetermined items in the medical information report (for example, entries corresponding to "contents"). In this case, the control unit 11 generates the learning model using a large amount of training data in which multiple medical information reports are associated with text data indicating entries corresponding to predetermined items in the medical information report.

[0134] The control unit 11 reads summary information of the patient corresponding to the received identification information from the summary information DB 14A (step S1312). The control unit 11 reads existing record information corresponding to the received identification information from the record information DB 14B (step S1313). The control unit 11 determines a first prompt including the input information extracted in step S1311, the summary information of the patient read in step S1312, the existing record information read in step S1313, and an instruction to cause the language model M1 to output the record information (step S1314). The configuration of the first prompt is the same as in the first embodiment. The instruction to cause the language model M1 to output the record information is stored in advance in the storage unit 12. If the existing record information is not stored in the record information DB 14B, the existing record information does not need to be included in the first prompt.

[0135] The control unit 11 inputs the determined first prompt into the language model M1 to acquire record information (step S1315). The control unit 11 replaces the existing record information with the record information acquired in step S1315 to update the existing record information stored in the record information DB 14B (step S1316). The control unit 11 determines a fourth prompt including the input information extracted in step S1311, the patient summary information read in step S1312, and an instruction to the language model M1 to update the summary information (step S1317).

[0136] 34 is an explanatory diagram showing an example of the fourth prompt. The fourth prompt 400 includes an input information field 401, an existing summary information field 402, and a fourth instruction field 403. The input information field 401 contains input information extracted from the received file. The existing summary information field 402 contains summary information of the patient corresponding to the received identification information. The fourth instruction field 403 contains an instruction for causing the language model M1 to update the summary information (e.g., "Update the summary information based on the above input information and the summary information")

[0137] The control unit 11 acquires summary information by inputting the determined fourth prompt into the language model M1 (step S1318). The control unit 11 updates the patient's summary information stored in the summary information DB 14A by replacing the existing summary information with the summary information acquired in step S1318 (step S1319). The communication unit 13 transmits the record information acquired in step S1315 and the summary information updated in step S1319 to the smartphone 30 (step S1320).

[0138] The communication unit 23 receives the record information and summary information transmitted from the server 10 (step S1206). The display unit 25 displays the received record information and summary information on the display unit 25 (step S1207). Specifically, the display unit 25 displays the received record information on the second interface R2 of the setting screen and displays the updated summary information on the third interface R3 of the setting screen. The control unit 21 may display only the received record information on the display unit 25. In that case, the display unit 25 displays the received record information on the second interface R2 of the setting screen. The communication unit 13 may transmit the record information and summary information to the smartphone 30. In that case, the communication unit 33 receives the record information and summary information transmitted from the server 10. The control unit 31 displays the received storage information and summary information on the display unit 35.

[0139] (Modification) In the sixth embodiment, an example was shown in which the computer 20 specifies the identification information for identifying the patient corresponding to the file received by the medical record system, but this is not limited to this. The medical record system may specify the identification information for identifying the patient by the server 10. The processing in this case will be described below. When a file is received from a facsimile machine, an external device, a smartphone 30, or the like, the communication unit 23 transmits the received file to the server 10.

[0140] The communication unit 13 receives a file transmitted from the computer 20. The control unit 11 reads the contents of the received file using OCR processing. The control unit 11 determines a prompt including the read contents and instructions for identifying identification information from the received file. The instructions for identifying the identification type from the received file are, for example, instructions to the language model M1 for identifying the identification information contained in the received file from the contents read by OCR processing, such as "Based on the contents of the received file, identify the identification information (name or date of birth) contained in the received file." The instructions for identifying the identification information from the received file are stored in advance in the storage unit 12 or the mass storage unit 14.

[0141] The control unit 11 inputs the read description and a prompt including an instruction for identifying identification information from the received file into the language model M1, thereby identifying the identification information corresponding to the received file. The control unit 11 identifies the type of the received file, for example, by comparing the subject of the received file with the name of a file corresponding to the first or second type pre-stored in the storage unit 12. The subsequent processing is the same as in the sixth embodiment, and therefore description thereof will be omitted.

[0142] In embodiment 6, the medical record system identifies the type of the received file and the identification information for identifying the patient contained in the file, and if the identified type is the first type or the second type, it can output record information regarding the treatment of the patient by providing the input information extracted from the received file and the patient summary information corresponding to the identification information to a language model.

[0143] According to embodiment 6, when the identified type is the second type, the medical record system can update the patient summary information by providing the input information extracted from the received file and the patient summary information corresponding to the identification information to the language model.

[0144] (Embodiment 7) Embodiment 7 describes a method in which a medical record system outputs a patient explanation by providing input information, summary information, and an instruction to create a patient explanation to a language model.

[0145] Fig. 35 is a flowchart showing an example of the processing procedure of the medical record system according to the seventh embodiment. In the flowchart shown in Fig. 35, the part after step S205 in the processing shown in Fig. 6 to Fig. 7 is replaced with steps S1401 to S1403 and steps S1501 to S1502. Explanations of steps similar to those in Fig. 6 to Fig. 7 will be omitted. The control unit 11 determines a fifth prompt including the input information received in step S204, the record information read in step S205, and a command to create an explanation for the patient for the language model M1 (step S1401).

[0146] 36 is an explanatory diagram showing an example of the fifth prompt. The fifth prompt 500 includes an input information field 501, an existing summary information field 502, and a fifth instruction field 503. The input information field 501 contains input information acquired via the input field R0 in the setting screen. The existing summary information field 502 contains existing summary information read from the summary information DB 14A. The fifth instruction field 503 contains a command for creating an explanation for the patient for the language model M1.

[0147] The command to create an explanation for the patient includes requests regarding the content, expression, format, etc. of the explanation for the patient, as well as the items that make up the explanation for the patient. The command to create an explanation for the patient is stored in advance in the memory unit 12 or the mass storage unit 14. The fifth instruction field 503 in Figure 36 states, "Based on the above input information and summary information, please summarize the medical results, instructions, medication and treatment precautions in simple language so that the patient can easily understand. Please avoid medical terminology as much as possible and include specific and practical advice. #Contents of today's treatment / procedure #Precautions and advice for daily life #Next schedule (examination, treatment, etc.)"

[0148] The control unit 11 acquires an explanation for the patient by inputting the determined fifth prompt into the language model M1 (step S1402). The communication unit 13 transmits the acquired explanation for the patient to the smartphone 30 (step S1403). The communication unit 33 receives the explanation for the patient transmitted from the server 10 (step S1501). The control unit 31 displays the received explanation for the patient on the display unit 35 (step S1502).

[0149] FIG. 37 is an explanatory diagram showing an example of an explanation for patients. Screen d2 shown in FIG. 37 displays explanations for patients by item. In FIG. 37, an example of an explanation for patients is displayed as follows: "# Details of today's treatment / procedure Today, you said that your cough has continued, so we have added a nasal spray (fluticasone) in consideration of the possibility of allergic rhinitis. You will continue to take the cholesterol-lowering medication (pravastatin) as before. # Precautions and advice for daily life It's great that you have continued to quit smoking! I'm relieved to hear that the environment at your new workplace, where there are few smokers, seems to be helping you continue not smoking. We're rooting for you to continue staying abstaining. I'm relieved that your cough has continued, but it doesn't seem serious, so for now I'm relieved. We'll try using the nasal spray prescribed this time and see if your symptoms improve. # Next appointment (test, treatment, etc.) Next month, we'll take your blood next month to check your cholesterol levels, etc."

[0150] In the seventh embodiment, the communication unit 13 transmits the explanation for the patient acquired by the language model M1 to the smartphone 30, but this is not limiting. The communication unit 13 may transmit the explanation for the patient to the computer 20. In this case, the communication unit 23 receives the explanation for the patient transmitted from the server 10. The control unit 21 displays the received explanation for the patient on the display unit 25.

[0151] In the seventh embodiment, the control unit 11 outputs an explanation for the patient based on input information acquired via the setting screen (first embodiment), summary information corresponding to the patient's name, and a command to create an explanation for the patient. However, this is not limiting. The control unit 11 may output an explanation for the patient based on input information indicating the content of the dialogue input via the smartphone 30, summary information for the patient corresponding to identification information for identifying the patient, and a command to create an explanation for the patient. In this case, the input information field 501 of the fifth prompt 500 contains input information indicating the content of the dialogue input via the smartphone 30, and the existing summary information field 502 contains summary information for the patient corresponding to the identification information.

[0152] The control unit 11 may also output an explanation for the patient based on the input information extracted from the received file, summary information for the patient corresponding to the identification information for identifying the patient, and a command to create an explanation for the patient. In this case, the input information extracted from the received file is entered in the input information field 501 of the fifth prompt 500, and the summary information for the patient corresponding to the identification information is entered in the existing summary information field 502.

[0153] According to the seventh embodiment, the medical record system can output an explanation for the patient by providing the input information, summary information, and an instruction to create an explanation for the patient to the language model.

[0154] (Embodiment 8) Embodiment 8 describes a method in which a medical record system outputs advice or a report for a medical professional by providing input information, summary information, and instructions for creating advice or a report for a medical professional to a language model.

[0155] The advice for medical professionals is, for example, advice indicating future treatment plans, etc., for medical professionals (e.g., the patient's attending physician and nurse, etc.) who are treating the patient. The report for medical professionals is, for example, a report for other medical institutions, etc. to confirm the details of treatments, etc., previously performed on the patient, or a report prepared for medical professionals, etc. who will be newly treating the patient due to personnel changes, etc.

[0156] First, the process for outputting advice for medical professionals will be described. Fig. 38 is a flowchart showing an example of the procedure for outputting advice for medical professionals. In the flowchart shown in Fig. 38, the process after step S205 in the processes shown in Figs. 6 to 7 is replaced with steps S1601 to S1603 and steps S1701 to S1702. Descriptions of steps similar to those in Figs. 6 to 7 will be omitted.

[0157] The control unit 11 determines a sixth prompt including the input information received in step S204, the record information read in step S205, and a command to create advice for medical professionals for the language model M1 (step S1601). The command to create advice for medical professionals is stored in advance in the storage unit 12. The sixth prompt includes an input information field, an existing summary information field, and a sixth instruction field. The configurations of the input information field and the existing summary information field are the same as those of the fifth prompt. The command to create advice for medical professionals is written in the sixth instruction field.

[0158] The command to create advice for medical professionals includes requests regarding the content, expression, format, etc. of the advice for medical professionals, as well as items that make up the advice for medical professionals. The command to create advice for medical professionals is stored in advance in the memory unit 12 or the mass storage unit 14. For example, the command to create advice for medical professionals is, "Based on the above input information and summary information, please briefly summarize any changes in the patient's condition, test items, medication adjustments, symptoms that require particular attention, and patient complaints that require particular attention at the next consultation. #Special notes (points requiring attention) #Points to note at the next consultation."

[0159] The control unit 11 acquires advice for medical professionals by inputting the determined sixth prompt into the language model M1 (step S1602). The communication unit 13 transmits the acquired advice for medical professionals to the smartphone 30 (step S1603). The communication unit 33 receives the advice for medical professionals transmitted from the server 10 (step S1701). The control unit 31 displays the received advice for medical professionals on the display unit 35 (step S1702).

[0160] Fig. 39 is an explanatory diagram showing an example of advice for medical professionals. On a screen d3 shown in Fig. 39, advice for medical professionals is displayed for each item. Figure 39 shows an example of advice for medical professionals, which reads, "#Special notes (points requiring attention) * As the cough symptoms continue, check the effectiveness of the treatment for allergic rhinitis. * Check and provide support for continued smoking cessation. * Regarding the stress following the workplace transfer, it seems that the patient has managed to overcome it successfully this time, but continue to monitor closely. #Points to note at the next appointment * **Check the progress of cough symptoms: ** Check the effectiveness of the nasal spray (fluticasone) for allergic rhinitis, and any changes in the frequency, severity, and nature of the cough. If there is no improvement, consider additional tests or treatment. * **Check and provide support for continued smoking cessation: ** The patient appears to be gaining confidence in quitting smoking, but continue to provide support to continue. * **Monitor hyper LDL cholesterol level: ** Check the blood test results next month and evaluate the effectiveness of treatment with pravastatin. * **Watch for signs of recurrence of depression: ** Check how the patient is coping with stress due to changes in the workplace environment, and monitor closely for signs of recurrence of depression. * **Uncomfortable throat: ** Check whether this is related to the cough or whether there are any other causes. * **Check symptoms of runny nose and cough: **Objectively evaluate the severity of runny nose and cough and determine the effectiveness of treatment.

[0161] Next, the process for outputting a report for medical professionals will be described. Fig. 40 is a flowchart showing an example of the procedure for outputting a report for medical professionals. In the flowchart shown in Fig. 40, the process after step S205 in the process shown in Figs. 6 to 7 is replaced with steps S1801 to S1803 and steps S1901 to S1902. Descriptions of steps similar to those in Figs. 6 to 7 will be omitted.

[0162] The control unit 11 determines a seventh prompt including the input information received in step S204, the record information read in step S205, and an instruction to create a report for medical professionals for the language model M1 (step S1801). The instruction to create a report for medical professionals is stored in advance in the storage unit 12. The seventh prompt includes an input information field, an existing summary information field, and a seventh instruction field. The configuration of the input information field and the existing summary information field is the same as that of the fifth prompt. The instruction to create a report for medical professionals is written in the seventh instruction field.

[0163] The instruction to create a report for medical professionals includes requests regarding the content, expression, and format of the report for medical professionals, as well as the items that make up the report for medical professionals. The instruction to create a report for medical professionals is stored in advance in the memory unit 12 or the mass storage unit 14. For example, the instruction to create a report for medical professionals may be as follows: "Based on the above input information and summary information, please clearly and specifically summarize the situation and history of the medical treatment, the treatment details, and future treatment plans, focusing on the key points, so that the patient's condition and medical progress can be clearly conveyed to team members who are not involved in the current medical treatment. Please clearly indicate any information that requires particular attention or that must be shared. #Summary of the condition and medical findings #Details of the tests and treatment performed #Future treatment policy and follow-up plan #Notes that should be shared within the team."

[0164] The control unit 11 acquires a report for medical professionals by inputting the determined seventh prompt into the language model M1 (step S1802). The communication unit 13 transmits the acquired report for medical professionals to the smartphone 30 (step S1803). The communication unit 33 receives the report for medical professionals transmitted from the server 10 (step S1901). The control unit 31 displays the received report for medical professionals on the display unit 35 (step S1902).

[0165] FIG. 41 is an explanatory diagram showing an example of a report for medical professionals. Screen d4 shown in FIG. 41 displays the report for medical professionals by item. In FIG. 41, an example of a report for medical professionals is shown as follows: "# Summary of symptoms and clinical findings The patient visited the clinic with a cough that has continued since last month and mild discomfort in the throat. He reportedly experiences cold-like symptoms at this time every year. He continues to smoke, but the urge to smoke has decreased due to changes in the work environment. Stress due to the transfer does not appear to be an issue this time. The depression has improved and no mention was made this time. The patient remains highly motivated to quit smoking and to follow a dietary regimen. Blood pressure is 126 / 78 mmHg. There is a runny nose and cough. There is no sign of serious illness. # Tests and treatments performed Based on the findings from the examination, allergic rhinitis was suspected, so fluticasone nasal spray was additionally prescribed. Pravastatin 5 mg / day will be continued. # Future treatment policy and follow-up plan Blood will be taken next month to monitor the progress of hyper-LDL cholesterol. We will continue to monitor the progress of the cough and support the patient in continuing to quit smoking. # Points to note to share within the team The patient has a strong desire to continue quitting smoking, and progress is going well. It is important that the changes in the work environment have contributed to the patient's continued abstinence, so we will continue to encourage and support him. Allergic rhinitis was suspected as the cause of the cough, so nasal sprays were started, but we will monitor his progress closely. In addition, to prevent a recurrence of past depression, it is necessary to continue regular follow-up and pay attention to any changes in the patient's mental state.

[0166] In the eighth embodiment, an example has been described in which the communication unit 13 transmits advice for medical professionals or a report for medical professionals acquired using the language model M1 to the smartphone 30, but this is not limiting. The communication unit 13 may transmit the advice for medical professionals or the report for medical professionals to the computer 20. In this case, the communication unit 23 receives the advice for medical professionals or the report for medical professionals transmitted from the server 10. The control unit 21 displays the received advice for medical professionals or the report for medical professionals on the display unit 25.

[0167] In the eighth embodiment, an example has been described in which the control unit 11 outputs advice for a healthcare professional or a report for a healthcare professional based on input information acquired via the setting screen (first embodiment), summary information corresponding to the patient's name, and an instruction to create advice for a healthcare professional or a report for a healthcare professional. However, this is not limiting. The control unit 11 may output advice for a healthcare professional or a report for a healthcare professional based on input information indicating the content of the dialogue input via the smartphone 30, patient summary information corresponding to identification information for identifying the patient, and an instruction to create advice for a healthcare professional or a report for a healthcare professional. In this case, the input information indicating the content of the dialogue input via the smartphone 30 is entered in the input information field of the sixth or seventh prompt, and the patient summary information corresponding to the identification information is entered in the existing summary information field.

[0168] The control unit 11 may also output advice for medical professionals or a report for medical professionals based on the input information extracted from the received file, summary information for the patient corresponding to the identification information for identifying the patient, and a command to create advice for medical professionals or a command to create a report for medical professionals. In this case, the input information field of the sixth or seventh prompt contains the input information extracted from the received file, and the summary information for the patient corresponding to the identification information is contained in the existing summary information field.

[0169] The matters described in each of the above embodiments can be combined with each other. Furthermore, the independent claims and dependent claims described in the claims can be combined with each other in any and all combinations, regardless of the reference format. Furthermore, the claims use a format in which a claim references two or more other claims (multiple claim format), but this is not limited to this. A multiple claim (multi-multi claim) that references at least one other multiple claim may also be used.

[0170] The embodiments disclosed herein are illustrative in all respects and should not be considered limiting. The scope of the present invention is defined by the claims, not by the above meaning, and is intended to include all modifications within the meaning and scope of the claims.

[0171] DESCRIPTION OF SYMBOLS 10 Information processing device (server) 11 Control unit 12 Memory unit 12P Control program 13 Communication unit 14 Large-capacity memory unit 14A Summary information DB 14B Record information DB 15 Reading unit 1a Portable storage medium AP Medical record application M1 Language model M2 Feature extraction layer M10 Second language model 20 Information processing device (computer) 21 Control unit 22 Memory unit 22P Control program 23 Communication unit 24 Input unit 25 Display unit 30 Information processing terminal (smartphone) 31 Control unit 32 Memory unit 32P Control program 33 Communication unit 34 Input unit 35 Display unit 100 First prompt 200 Second prompt 110 Addition prompt 120 Correction prompt 300 Third prompt 400 Fourth prompt 500 Fifth prompt

Claims

1. A program that causes a computer to execute the following process: acquire input information indicating the content of a conversation between a medical professional and a patient, and summary information summarizing the patient's treatment history; and output record information regarding the patient's treatment by providing the input information and summary information to a language model.

2. The program according to claim 1, wherein the recorded information is output in a SOAP format including subjective information, objective information, evaluation, and plan using the language model.

3. The program according to claim 1 or 2, wherein the input information includes, in addition to the content of the dialogue, the medical professional's findings regarding the patient or treatment information for the patient.

4. The program according to claim 2, further comprising: receiving additional information for the input information; and re-outputting the recorded information by providing the received additional information, the input information, and the summary information to the language model.

5. The program according to claim 2, further comprising: acquiring correction information for the recorded information; and re-outputting the recorded information by providing the acquired correction information and the recorded information to the language model.

6. The program according to claim 1 or 2, wherein the summary information is output by providing the output record information to the language model.

7. The program according to claim 4 or 5, wherein the summary information is output by providing the re-output recorded information to the language model.

8. The program according to claim 1 or 2, wherein a first interface for displaying the input information, a second interface for displaying the recorded information, and a third interface for displaying the summary information are output in association with each other.

9. The program according to claim 1 or 2, wherein the input information further includes image data showing the affected area of ​​the patient.

10. The program according to claim 1, which receives input information indicating the content of a conversation entered via a smartphone and identification information for identifying a patient, and outputs record information regarding the treatment of the patient by providing the received input information and summary information of the patient corresponding to the identification information to a language model.

11. The program according to claim 1, which identifies the type of a received file and identification information for identifying the patient corresponding to the file, and if the identified type is a first type or a second type, outputs record information regarding the treatment of the patient by providing input information extracted from the file and summary information of the patient corresponding to the identification information to a language model, and if the identified type is a second type, updates the summary information of the patient by providing input information extracted from the file and summary information of the patient corresponding to the identification information to a language model.

12. The program according to claim 1, which outputs an explanation for a patient by providing input information, summary information, and an instruction to create an explanation for a patient to a language model.

13. The program according to claim 1, which outputs advice for medical professionals or a report for medical professionals by providing input information, summary information, and instructions for creating advice for medical professionals or instructions for creating a report for medical professionals to a language model.

14. An information processing method that acquires input information indicating the content of a conversation between a medical professional and a patient and summary information summarizing the patient's treatment history, and outputs record information regarding the patient's treatment by providing the input information and the summary information to a language model.

15. An information processing device having a control unit, wherein the control unit acquires input information indicating the content of a conversation between a medical professional and a patient and summary information summarizing the patient's treatment history, and outputs record information regarding the patient's treatment by providing the input information and the summary information to a language model.

Images