Medical record generation support device, medical record generation support method, and program

Abstract

To provide a medical record generation support device, a medical record generation support method, and a program capable of easily supporting generation of a medical record for a medical worker.SOLUTION: One aspect of a medical record generation support device of the present disclosure includes a medical action information acquisition unit that acquires medical action information that is information on a prescription or an examination related to a symptom of a patient, a voice recognition result acquisition unit that acquires a voice recognition result of a conversation related to the symptom performed between a medical worker and the patient, and a medical record generation unit that inputs the medical action information and the voice recognition result to a generation model, acquires subjective information that the patient complains about the symptom, objective information that objectively captures the symptom, evaluation information indicating an evaluation content for the symptom, and a treatment plan for the symptom from the generation model, and generates a medical record.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present disclosure relates to a medical record creation support device, a medical record creation support method, and a program. [Background technology]

[0002] BACKGROUND ART Conventionally, techniques for supporting medical personnel in creating medical records, such as pharmacists creating medication histories and doctors creating medical charts, have been known.

[0003] For example, Patent Document 1 discloses a technology that supports the creation of a medication history by using conversation information between a pharmacist and a patient during medication instructions and the patient's previously created medication history to generate multiple candidate sentences for creating a medication history and allow the pharmacist to select one. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2018-197926 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the above-mentioned techniques have room for improvement in terms of easily realizing support for medical professionals in creating medical records.

[0006] The present disclosure has been made in consideration of the above circumstances, and aims to provide a medical record creation support device, a medical record creation support method, and a program that can easily provide support to medical professionals in creating medical records. [Means for solving the problem]

[0007] One aspect of the medical record generation support device disclosed herein includes a medical treatment information acquisition unit that acquires medical treatment information, which is prescription or test information related to a patient's symptoms; a voice recognition result acquisition unit that acquires voice recognition results of a conversation related to the symptoms between a medical professional and the patient; and a medical record generation unit that inputs the medical treatment information and the voice recognition results into a generation model and acquires from the generation model subjective information reported by the patient regarding the symptoms, objective information that objectively captures the symptoms, evaluation information that indicates an evaluation of the symptoms, and a treatment plan for the symptoms, and generates a medical record.

[0008] One aspect of the medical record generation support method disclosed herein includes a medical treatment information acquisition step in which a medical treatment information acquisition unit acquires medical treatment information, which is prescription or test information related to a patient's symptoms; a voice recognition result acquisition step in which a voice recognition result acquisition unit acquires voice recognition results of a conversation related to the symptoms between a medical professional and the patient; and a medical record generation step in which a medical record generation unit inputs the medical treatment information and the voice recognition results into a generation model, and acquires from the generation model subjective information reported by the patient about the symptoms, objective information that objectively captures the symptoms, evaluation information that indicates an evaluation of the symptoms, and a treatment plan for the symptoms, and generates a medical record.

[0009] One aspect of the program of the present disclosure is to cause a computer to function as a medical treatment information acquisition unit that acquires medical treatment information, which is prescription or test information related to a patient's symptoms; a voice recognition result acquisition unit that acquires voice recognition results of a conversation regarding the symptoms between a medical professional and the patient; and a medical record generation unit that inputs the medical treatment information and the voice recognition results into a generation model and acquires from the generation model subjective information reported by the patient regarding the symptoms, objective information that objectively captures the symptoms, evaluation information that indicates an evaluation of the symptoms, and a treatment plan for the symptoms, and generates a medical record. [Effects of the Invention]

[0010] According to the present disclosure, it is possible to provide a medical record creation support device, a medical record creation support method, and a program that can easily realize support for medical professionals in creating medical records. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a block diagram showing an example of the configuration of a medical record creation support system according to this embodiment. [Figure 2] FIG. 2 is a block diagram showing an example of the hardware configuration of the medical record creation support device of this embodiment. [Figure 3] FIG. 3 is a block diagram showing an example of the functional configuration of the medical record generation support device of this embodiment. [Figure 4] FIG. 4 is a diagram showing an example of the medication history generation screen of this embodiment. [Figure 5] FIG. 5 is a diagram showing an example of the medication history generation screen of this embodiment. [Figure 6] FIG. 6 is a diagram showing an example of the medication history generation screen of this embodiment. [Figure 7] FIG. 7 is a diagram showing an example of the medication history generation screen of this embodiment. [Figure 8] FIG. 8 is a diagram showing an example of the medication history generation screen of this embodiment. [Figure 9] FIG. 9 is a diagram showing an example of the medication history generation screen of this embodiment. [Figure 10] FIG. 10 is a flowchart showing an example of the process of displaying estimated subjective information and estimated treatment plans performed by the medical record generation support device of this embodiment. [Figure 11] FIG. 11 is a flowchart showing an example of medication history generation processing performed by the medical record generation support device of this embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, an embodiment of the present disclosure (hereinafter simply referred to as "the present embodiment") will be described in detail with reference to the drawings. Note that the present disclosure is not limited to the following embodiment. Furthermore, the following embodiment and modified examples can be combined as appropriate.

[0013] First, the configuration of the medical record creation support system of this embodiment will be described.

[0014] Fig. 1 is a block diagram showing an example of the configuration of a medical record generation support system 1 according to this embodiment. As shown in Fig. 1, the medical record generation support system 1 includes a medical record generation support device 10 and a generative model 20. The medical record generation support device 10 and the generative model 20 are connected via a network 2. The network 2 can be realized by, for example, at least one of the Internet and a LAN (Local Area Network).

[0015] The medical record creation support device 10 supports medical personnel in creating medical records, and can be realized by, for example, a PC (Personal Computer) or a tablet terminal.

[0016] A medical professional may be anyone engaged in medical care, such as a person who performs medical procedures, a person who supports medical procedures, or a person involved in medical administration, including, but not limited to, a doctor, a pharmacist, a nurse, a medical technician, and a medical administrative staff. A medical record may be any record related to medical care, including, but not limited to, a medical chart (including both medical records and medical records), a medication history, and a nursing record.

[0017] In this embodiment, the medical record creation support device 10 (medical record creation support system 1) supports the creation of a medical record by a medical professional, and the creation of a medication history by a pharmacist will be described as an example, but the present invention is not limited to this. For example, the creation of a medical record by a medical professional may also be the creation of a medical chart by a doctor.

[0018] A medication history refers to a patient's medication history, and is a record of the pharmacist's dispensing and medication instructions for the patient, as well as the patient's basic information, constitution, and illnesses. Medication histories include, but are not limited to, information on prescribed medications, the names of medications the patient is already taking, and the patient's allergy history, side effect history, and medical history. Information on prescribed medications includes, for example, the name, dosage, usage, and administration period of the prescribed medication.

[0019] In this embodiment, the medical history will be described as a SOAP medical history recorded in SOAP format, but is not limited to this. The SOAP format is a document format used in fields such as medicine and psychology when evaluating and recording patients, and is a method of recording findings divided into subjective findings (S: Subjective), objective findings (O: Objective), assessment (A: Assessment), and plan (P: Plan).

[0020] Subjective findings (S: Subjective) are subjective information obtained from what the subject has said, and in the case of a SOAP medication history, this includes, but is not limited to, information that the patient complains about regarding the symptoms when the pharmacist provides medication instructions for the patient's symptoms.

[0021] Objective findings (O) are objective information that objectively captures a patient's symptoms, and include, but are not limited to, findings by medical professionals obtained through examinations and tests.

[0022] Assessment (A: Assessment) is evaluation information that indicates the evaluation of the patient's symptoms, and includes, but is not limited to, a diagnosis by a medical professional or a comprehensive evaluation that analyzes and interprets based on subjective and objective findings.

[0023] The plan (P) is a treatment plan for the patient's symptoms, and includes, but is not limited to, the treatment policy and content determined based on the assessment (A), as well as lifestyle guidance.

[0024] The generative model 20 is a natural language processing model known as generative AI (Generative Artificial Intelligence), and provides a service using natural language processing. In the first embodiment, the generative model 20 will be described taking as an example a case where an AI chat service is provided by fine-tuning a large-scale language model (LLM), but is not limited to this.

[0025] In the medical record generation support system 1 of this embodiment, for example, after a pharmacist has given medication instructions to a patient, during a time when there are few customers at the pharmacy, or after the pharmacy's dispensing reception hours have ended, the pharmacist generates a medication history using the medical record generation support device 10. Specifically, the medical record generation support device 10 generates prompts based on speech recognition results obtained by speech recognition of the prescription prescribed to the patient and the dialogue when the pharmacist gave medication instructions to the patient, and inputs these into the generation model 20, thereby obtaining the medication history details from the generation model 20 and generating the medication history.

[0026] Fig. 2 is a block diagram showing an example of the hardware configuration of the medical record generation support device 10 of this embodiment. As shown in Fig. 2, the medical record generation support device 10 includes a control device 11, a main memory device 12, an auxiliary memory device 13, an input device 14, a display device 15, a voice input device 16, an imaging device 17, a communication device 18, and various buses 19. The control device 11, the main memory device 12, the auxiliary memory device 13, the input device 14, the display device 15, the voice input device 16, the imaging device 17, and the communication device 18 are connected via the various buses 19. As such, the medical record generation support device 10 of this embodiment has a general hardware configuration of electronic devices such as PCs and tablet terminals.

[0027] The control device 11 controls the overall operation of the medical record generation support device 10. The control device 11 may be, for example, at least one of a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit), but is not limited to these. There may be any number of CPUs or GPUs as long as they are one or more, and they may be single-core or multi-core.

[0028] Examples of the main memory device 12 include, but are not limited to, a ROM (Read Only Memory) and a RAM (Random Access Memory). The ROM stores various programs, such as a program for controlling the medical record generation support device 10 and a program for supporting the generation of medical records (medication history). The RAM is used as a working area when the control device 11 performs various controls based on the programs stored in the ROM.

[0029] The auxiliary storage device 13 stores the various programs described above and various data, such as format data used to generate prompts. The various programs described above may be stored in at least one of the main storage device 12 and the auxiliary storage device 13. Examples of the auxiliary storage device 13 include, but are not limited to, existing storage devices capable of magnetic, electrical, or optical storage, such as a hard disk drive (HDD), a solid state drive (SSD), and a digital versatile disc (DVD). The auxiliary storage device 13 may be built into the medical record creation support device 10 or may be externally attached to the medical record creation support device 10 via an interface such as a universal serial bus (USB). The auxiliary storage device 13 may also be a network-attached storage (NAS) connected via a network, such as a LAN or a wide area network (WAN).

[0030] The input device 14 is used for various inputs, selections, and specifications used when generating a medical record (medication history), and serves as a user interface between the medical record generation support device 10 and the medical professional (pharmacist). Examples of the input device 14 include, but are not limited to, a keyboard, a mouse, and a touch panel. The input device 14 may be built into the medical record generation support device 10, or may be externally connected to the medical record generation support device 10 via an interface such as a USB.

[0031] The display device 15 displays various screens used when generating a medical record (medication history) and serves as a user interface between the device and a medical professional (pharmacist). Examples of the display device 15 include, but are not limited to, various displays such as a liquid crystal display, an organic electroluminescence (EL) display, and a touch panel display. The display device 15 may be a built-in display built into the medical record generation support device 10, or an external display connected to the medical record generation support device 10 via a display interface such as HDMI (registered trademark).

[0032] The voice input device 16 is used to input spoken voices such as those from a conversation between a medical professional (pharmacist) and a patient, and can be realized by a sound collecting device such as a microphone.

[0033] The photographing device 17 photographs a prescription given to a patient, and may be, for example, an image sensor (digital camera).

[0034] The communication device 18 is used to communicate with the generative model 20 and the like via the network 2. Examples of the communication device 18 include, but are not limited to, a communication device for a wired LAN and a wireless communication device for a wireless LAN.

[0035] In addition to the above configuration, the medical record generation support device 10 may further include hardwired circuits such as an IC (Integrated Circuit), ASIC (Application Specific Integrated Circuit), and FPGA (Field-Programmable Gate Array) that are specific to the medical record generation support device 10.

[0036] Fig. 3 is a block diagram showing an example of the functional configuration of the medical record generation support device 10 of this embodiment. As shown in Fig. 3, the medical record generation support device 10 includes a medical treatment information acquisition unit 201, an inferred information generation unit 203, a display control unit 205, a dialogue voice acquisition unit 207, a voice recognition unit 209, a voice recognition result acquisition unit 211, and a medical record generation unit 213. The medical record generation unit 213 includes a first medical record information acquisition unit 215 and a second medical record information acquisition unit 217.

[0037] The medical treatment information acquisition unit 201, the inferred information generation unit 203, the display control unit 205, the dialogue voice acquisition unit 207, the voice recognition unit 209, the voice recognition result acquisition unit 211, the medical record generation unit 213, the first medical record information acquisition unit 215, and the second medical record information acquisition unit 217 can be realized, for example, by the control device 11, the main memory device 12, and the communication device 18 described in FIG. 2 . For example, the control device 11 reads a program for supporting the generation of a medical record (medication history) stored in the main memory device 12 (ROM) or the auxiliary memory device 13 and loads it into the main memory device 12 (RAM). The control device 11 realizes each of the above-mentioned functional units by executing various processes in accordance with the loaded program. Here, the above-mentioned functional units are realized as software, but at least a portion of the above-mentioned functional units may be realized as hardware. In this case, the functional units realized as hardware may be realized, for example, by the above-mentioned hardwired circuit. Furthermore, any of the above-mentioned functional units may be realized by a combination of software and hardware.

[0038] The medical treatment information acquisition unit 201 acquires medical treatment information, which is information on prescriptions or tests related to the patient's symptoms. In this embodiment, the medical treatment information acquisition unit 201 acquires information on prescriptions prescribed for the patient's symptoms as the medical treatment information.

[0039] For example, a pharmacist photographs a prescription brought by a patient to a pharmacy using a photographing device 17 (see FIG. 2). The medical treatment information acquisition unit 201 acquires image data of the photographed prescription, performs character recognition (OCR: Optical Character Reader) processing, or causes the generative model 20 to perform character recognition processing and acquires the processing results, thereby acquiring prescription information (text information). Alternatively, for example, the pharmacist may input information (text information) of the prescription brought by the patient to a pharmacy or the like using an input device 14 (see FIG. 2), thereby causing the medical treatment information acquisition unit 201 to acquire prescription information (text information). Alternatively, for example, if the prescription has been issued as an electronic prescription, the medical treatment information acquisition unit 201 may acquire the prescription information (text information) by acquiring the electronic prescription from a management server (not shown) that manages electronic prescriptions.

[0040] The inferred information generating unit 203 generates inferred subjective information that infers subjective information and an inferred treatment plan that infers a treatment plan, using the medical treatment information acquired by the medical treatment information acquiring unit 201. Specifically, the inferred information generating unit 203 generates inferred subjective information and an inferred treatment plan that infer the contents of the subjective information (S in SOAP medication history) and the treatment plan (P in SOAP medication history) obtained from a dialogue between a pharmacist and a patient during medication instructions, which will be described later, using prescription information acquired by the medical treatment information acquiring unit 201.

[0041] The inferred information generation unit 203 generates inferred subjective information and a predicted treatment plan by performing inference processing on the acquired prescription information, such as processing using statistical methods or generating prompts based on the prescription information and inputting the prompts into the generative model 20. For example, if the prescription contains "Tamiflu" (registered trademark), a known anti-influenza virus drug, the inferred information generation unit 203 can generate (acquire) general subjective information and a treatment plan for a patient suffering from influenza by performing the above-mentioned inferential processing. The inferred information generation unit 203 regards the generated general subjective information and treatment plan as inferred subjective information and a predicted treatment plan, respectively.

[0042] The display control unit 205 displays the estimated subjective information and estimated treatment plan generated by the estimated information generating unit 203 on the display device 15 (see FIG. 2).

[0043] Note that, after receiving a prescription from a patient, a pharmacist usually dispenses medicine and provides medication instructions to the patient. Therefore, the inferred information generating unit 203 can generate the inferred subjective information and the inferred treatment plan, and the display control unit 205 can display them on the display device 15, before a dialogue regarding the patient's symptoms takes place between a medical professional and a patient, such as a dialogue when the pharmacist provides medication instructions to the patient.

[0044] This allows the pharmacist to check the estimated subjective information and estimated treatment plan displayed on the display device 15 before providing medication instructions to the patient, and to provide medication instructions to the patient after making prior preparations. Therefore, according to this embodiment, even pharmacists who have difficulty eliciting sufficient information from patients during medication instructions, due to their limited practical experience or insufficient pharmaceutical knowledge, can improve the quality of the information they obtain. As described above, this embodiment is expected to achieve a high level of uniformity in the content of information elicited from patients during medication instructions, without relying on the pharmacist's skill, and also to improve the accuracy of the medication history generated using the dialogue during medication instructions, as described below.

[0045] The following describes a UI (User Interface) used to generate and display the estimated subjective information and the estimated treatment plan. In this embodiment, an example is described in which the estimated subjective information and the estimated treatment plan are generated and displayed on a medication history generation screen (an example of a medical record generation screen used to generate a medical record) used to generate a medication history, which will be described later. However, the present invention is not limited to this, and the estimated subjective information and the estimated treatment plan may be generated and displayed on a screen different from the medication history generation screen.

[0046] Fig. 4 is a diagram showing an example of a medication history generation screen 301 used to generate a medication history in this embodiment. The medication history generation screen 301 shown in Fig. 4 generates and displays estimated subjective information and an estimated treatment plan, and is displayed on the display device 15 by the display control unit 205 when a pharmacist inputs a display instruction using the input device 14 after prescription information has been acquired by the medical treatment information acquisition unit 201 and before medication instructions are given to the patient. Note that in the following description of the UI, it is assumed that various operational inputs, such as instructions, selections, and specifications, from the pharmacist to the medication history generation screen 301 (medical record generation support device 10) are made using the input device 14, unless otherwise specified.

[0047] As shown in Figure 4, the medication history generation screen 301 includes a prescription selection display field 311, an estimated SP generation button 313, a dialogue information selection display field 315, an SP generation button 317, an OA generation button 319, a prescription content display field 321, an S (subjective information) display field 323, a P (treatment plan) display field 325, a dialogue content display field 331, an O (objective information) display field 333, and an A (evaluation information) display field 335.

[0048] 4 is in an initial state, no information is selected in the prescription selection display field 311 and the dialogue information selection display field 315. Also, the estimated SP generation button 313, SP generation button 317, and OA generation button 319 are controlled so that their operations are disabled. Also, no information is displayed in the prescription content display field 321, S (subjective information) display field 323, P (treatment plan) display field 325, dialogue content display field 331, O (objective information) display field 333, and A (evaluation information) display field 335.

[0049] The prescription selection display field 311 is for selecting and displaying a prescription to be used for generating estimated subjective information, an estimated treatment plan, and a medication history. The prescription content display field 321 is a display field for displaying information on the prescription (prescription content) selected in the prescription selection display field 311. The estimated SP generation button 313 is an operation button used for generating estimated subjective information and an estimated treatment plan.

[0050] When the pharmacist selects the pull-down button in the prescription selection display field 311, a list of prescriptions (for example, a list of prescription file names) for which information has been acquired by the medical treatment information acquisition unit 201 is displayed. When the pharmacist selects a prescription to be used for generating estimated subjective information, an estimated treatment plan, and a medication history from the list of prescriptions, the display control unit 205 displays the selected prescription (prescription file name) in the prescription selection display field 311. The display control unit 205 also displays information about the selected prescription (prescription text information acquired by the medical treatment information acquisition unit 201) in the prescription content display field 321. When a prescription is selected, the display control unit 205 also controls the estimated SP generation button 313 to change from an inoperative state to an operable state.

[0051] 5 is a diagram showing an example of a medication history generation screen 301 used to generate a medication history in this embodiment, and shows an example of the state after a prescription has been selected by a pharmacist. In the medication history generation screen 301 shown in FIG. 5, the prescription "Prescription 1" selected by the pharmacist is displayed in the prescription selection display field 311, and the prescription content, which is information about "Prescription 1," is displayed in the prescription content display field 321.

[0052] 5, prescription information has been input, and it is possible to generate estimated subjective information and estimated treatment plans, so that control is performed so that operation of the estimated SP generation button 313 is enabled. Specifically, the display control unit 205 controls the display of the estimated SP generation button 313 from a state in which operation is disabled (a state in which the estimated SP generation button 313 is shown shaded) to a state in which operation is enabled (a state in which the estimated SP generation button 313 is shown without shading). Note that, in a state in which operation of the estimated SP generation button 313 is disabled, the control device 11 does not accept an operation signal input by selecting the estimated SP generation button 313, and in a state in which operation of the estimated SP generation button 313 is enabled, the control device 11 accepts an operation signal input by selecting the estimated SP generation button 313.

[0053] The S (subjective information) display field 323 is a display field for displaying estimated subjective information and subjective information. The P (treatment plan) display field 325 is a display field for displaying estimated treatment plans and treatment plans. The display of subjective information and treatment plans will be described later.

[0054] When the pharmacist selects the estimated SP generation button 313 on the medication history generation screen 301 shown in FIG. 5, the estimated information generation unit 203 generates a prompt for acquiring estimated subjective information and an estimated treatment plan using the prescription details (prescription information) displayed in the prescription details display field 321. Examples of prompts include, but are not limited to, prompts that instruct the pharmacist to generate S (subjective information) and P (treatment plan) information estimated from the prescription information, with reference to the contents of a standard format that describes the SOAP medication history description and description example. For example, if the patient is filling out a questionnaire, the prompt may instruct the pharmacist to generate S (subjective information) and P (treatment plan) information estimated from the prescription information, with further reference to the contents of the questionnaire. Examples of the explanation of the SOAP medication history include the explanation of each of the SOAP contents described above. Examples of the description example of the SOAP medication history include, for example, the description example of each of the SOAP descriptions.

[0055] Note that a standard format describing the SOAP medication history and an example description may be stored in, for example, the auxiliary storage device 13 (see FIG. 2), and the inferred information generating unit 203 may generate a prompt by adding prescription information to the standard format. The prompt may further include a role and an output format specification. The role is a role given to the generative model 20, and it is expected that the generative model 20 will generate an answer based on the knowledge of the specified role. In this embodiment, an example of the role is a pharmacist. An example of the output format is to have S and P information answered separately in JSON format.

[0056] The estimated information generation unit 203 inputs the prompt generated in this manner into the generative model 20, and acquires the S and P information generated by the generative model 20 as estimated subjective information and estimated treatment plan, respectively. When the estimated subjective information and estimated treatment plan are acquired by the estimated information generation unit 203, the display control unit 205 displays them in the S (subjective information) display field 323 and the P (treatment plan) display field 325, respectively.

[0057] 6 is a diagram showing an example of a medication history generation screen 301 used to generate a medication history in this embodiment, and shows an example of a state in which estimated subjective information is displayed in the S (subjective information) display field 323 and an estimated treatment plan is displayed in the P (treatment plan) display field 325. By checking the estimated subjective information and the estimated treatment plan on the medication history generation screen 301 shown in FIG. 6, a pharmacist can provide medication instructions to a patient after making advance preparations, as described above.

[0058] Returning to Figure 3, the functions used to generate a medication history will be explained.

[0059] The dialogue voice acquisition unit 207 acquires the voice of a dialogue between a medical professional and a patient regarding the patient's symptoms. For example, the dialogue voice acquisition unit 207 acquires the voice of a dialogue when a pharmacist gives medication instructions to a patient, which is collected (input) by the voice input device 16 (see FIG. 2).

[0060] The voice recognition unit 209 recognizes the voice acquired by the dialogue voice acquisition unit 207 , generates information (text information) indicating the content of the dialogue, and stores it in the auxiliary storage device 13 .

[0061] The voice recognition result acquisition unit 211 acquires the voice recognition result of a conversation between a medical professional and a patient regarding the patient's symptoms. For example, when a medication history is generated after a pharmacist has given medication instructions to a patient, during a time when there are few customers at the pharmacy, or after the pharmacy's prescription reception hours have ended, the voice recognition result acquisition unit 211 acquires information indicating the content of the conversation as a voice recognition result from the auxiliary storage device 13.

[0062] The medical record generation unit 213 inputs the medical treatment information acquired by the medical treatment information acquisition unit 201 and the voice recognition results acquired by the voice recognition result acquisition unit 211 into the generation model 20, acquires subjective information, objective information, evaluation information, and treatment plans from the generation model 20, and generates a medical record.

[0063] Specifically, the first medical record information acquisition unit 215 included in the medical record generation unit 213 inputs a first prompt using at least the acquired speech recognition result to the generation model 20, and acquires subjective information and a treatment plan from the generation model 20. The first prompt includes, for example, an instruction to extract the subjective information (S in SOAP medication history) and the treatment plan (P in SOAP medication history) from the acquired speech recognition result. The first prompt may also include, for example, at least one of an explanation of SOAP in the medical record (medical history) and a description example of a medical record (medical history) in SOAP format. The first prompt may also include medical procedure information (prescription information).

[0064] Furthermore, the second medical record information acquisition unit 217 included in the medical record generation unit 213 inputs a second prompt using the acquired subjective information, treatment plan, and medical procedure information into the generative model 20, and acquires objective information and evaluation information from the generative model 20. The second prompt includes, for example, an instruction to generate objective information (O of SOAP medical history) and evaluation information (A of SOAP medical history) based on the acquired subjective information (S of SOAP medical history), treatment plan (P of SOAP medical history), and medical procedure information (prescription information). Furthermore, the second prompt may further include at least one of an explanation of SOAP in medical records and a description example of a medical record in SOAP format.

[0065] The medical record generation unit 213 generates a medical record (medication history) using the acquired subjective information, objective information, evaluation information, and treatment plan as S, O, A, and P, respectively, and stores the medical record in the auxiliary storage device 13.

[0066] The inventors of this embodiment discovered that it is highly likely that S and P of the SOAP medication history are what primarily appear in the dialogue when a pharmacist provides medication instructions to a patient. For this reason, in this embodiment, S and P of the SOAP medication history are generated using the content of the dialogue when a pharmacist provides medication instructions to a patient, and then O and A of the SOAP medication history are generated using the generated S and P and prescription information.

[0067] Therefore, according to this embodiment, the accuracy of the contents of the SOAP medication history can be improved compared to when S, O, A, and P of the SOAP medication history are generated all at once using the conversations that a pharmacist has with patients when providing medication instructions and prescription information.

[0068] The following describes the UI used for generating and displaying subjective information, objective information, evaluation information, and treatment plans. The following describes an example in which a medication history is generated from the state of the medication history generation screen 301 shown in FIG. 6 described above. When a pharmacist generates a medication history after providing medication instructions to a patient, it is assumed that the medication history is generated from this state. However, this is not limited to this, and the medication history may be generated from the initial state of the medication history generation screen 301 shown in FIG. 4. When a medication history is generated during a time when there are few customers at the pharmacy or after the pharmacy's dispensing reception hours have ended, it is assumed that the medication history is generated from this state. Note that when a medication history is generated from the initial state of the medication history generation screen 301 shown in FIG. 4, it is necessary to select the prescription again.

[0069] The dialogue information selection and display field 315 is for selecting and displaying a dialogue to be used in generating a medication history. The dialogue content display field 331 is a display field for displaying the voice recognition result (text information indicating the content of the dialogue) of the dialogue selected in the dialogue information selection and display field 315. The SP generation button 317 is an operation button used to generate subjective information and a treatment plan.

[0070] When the pharmacist selects a pull-down button in the dialogue information selection display field 315, a list of dialogues for which voice recognition results have been acquired by the voice recognition result acquisition unit 211 (for example, a list of file names of the voice recognition results of the dialogues) is displayed. When the pharmacist selects a dialogue to be used for generating a medication history from the list of dialogues, the display control unit 205 displays the selected dialogue (file name of the voice recognition result) in the dialogue information selection display field 315. Furthermore, the display control unit 205 displays the voice recognition result of the selected dialogue (the voice recognition result acquired by the voice recognition result acquisition unit 211) in the dialogue content display field 331. Furthermore, when a dialogue is selected, the display control unit 205 controls the SP generation button 317 from a disabled state to an enabled state, and controls the estimated SP generation button 313 from an enabled state to a disabled state.

[0071] 7 is a diagram showing an example of a medication history generation screen 301 used to generate a medication history in this embodiment, and shows an example of the state after a dialogue has been selected by a pharmacist. On the medication history generation screen 301 shown in FIG. 7, the dialogue "Dialogue 1" selected by the pharmacist is displayed in the dialogue information selection display field 315, and the dialogue content, which is the speech recognition result of "Dialogue 1," is displayed in the dialogue content display field 331.

[0072] 7, the voice recognition results of the dialogue are input, and since it is possible to generate subjective information and a treatment plan, the operation of the SP generation button 317 is controlled to be enabled. Specifically, the display control unit 205 controls the display of the SP generation button 317 from a state in which the operation is disabled (a state in which the SP generation button 317 is shown shaded) to a state in which the operation is enabled (a state in which the SP generation button 317 is shown without shading). Note that, in the state in which the operation of the SP generation button 317 is disabled, the control device 11 does not accept an operation signal input by selecting the SP generation button 317, and in the state in which the operation of the SP generation button 317 is enabled, the control device 11 accepts an operation signal input by selecting the SP generation button 317.

[0073] 7, the voice recognition result of the dialogue has been input, and since it is not necessary to generate estimated subjective information and a estimated treatment plan, the display control unit 205 controls so as to disable operation on the estimated SP generation button 313. Furthermore, on the medication history generation screen 301 shown in FIG. 7, the estimated subjective information displayed in the S (subjective information) display field 323 and the estimated treatment plan displayed in the P (treatment plan) display field 325 have been initialized by the display control unit 205.

[0074] When the pharmacist selects the SP generation button 317 (an example of a first symbol) on the medication history generation screen 301 shown in FIG. 7 , the first medical record information acquisition unit 215 generates a first prompt for acquiring subjective information and a treatment plan using the dialogue content (the voice recognition result of the dialogue) displayed in the dialogue content display field 331. Examples of the first prompt include, but are not limited to, an instruction to extract S (subjective information) and P (treatment plan) from the voice recognition result of the dialogue, with reference to the content of the aforementioned standard format that describes the SOAP medication history explanation and description example. For example, the first prompt may instruct to extract S (subjective information) and P (treatment plan) from the voice recognition result of the dialogue, with further reference to the prescription content (prescription information) displayed in the prescription content display field 321. Furthermore, for example, if the patient has filled out a questionnaire or has a history of past visits, the prompt may instruct to generate S (subjective information) and P (treatment plan) information estimated from the prescription information, with further reference to the content. The explanation of the SOAP medication history, an example of a SOAP medication history description, the method for generating the first prompt, and the information to be included in the first prompt (specification of role and output format) are the same as those for the prompts for obtaining the estimated subjective information and estimated treatment plan described above, so they will not be explained here.

[0075] The first medical record information acquisition unit 215 inputs the first prompt generated in this manner into the generative model 20, and acquires the S and P information generated by the generative model 20 as subjective information and a treatment plan, respectively. When the subjective information and treatment plan are acquired by the first medical record information acquisition unit 215, the display control unit 205 displays them in the S (subjective information) display field 323 and the P (treatment plan) display field 325, respectively.

[0076] 8 is a diagram showing an example of a medication history generation screen 301 used to generate a medication history in this embodiment, and shows an example of a state in which subjective information is displayed in the S (subjective information) display field 323 and a treatment plan is displayed in the P (treatment plan) display field 325. The display control unit 205 displays the subjective information displayed in the S (subjective information) display field 323 and the treatment plan displayed in the P (treatment plan) display field 325 so that they can be modified in response to a modification operation by a pharmacist. Therefore, if it is necessary to modify the subjective information or the treatment plan, the pharmacist can perform the modification operation.

[0077] The OA generation button 319 is an operation button used to generate objective information and evaluation information. The O (objective information) display field 333 is a display field for displaying objective information. The A (evaluation information) display field 335 is a display field for displaying evaluation information.

[0078] 8, subjective information and a treatment plan have been acquired, and objective information and an evaluation plan can be generated. Therefore, the display control unit 205 controls the display of the OA generation button 319 so that operation of the OA generation button 319 is enabled. Specifically, the display control unit 205 controls the display of the OA generation button 319 from a state in which operation is disabled (a state in which the OA generation button 319 is shown shaded) to a state in which operation is enabled (a state in which the OA generation button 319 is shown without shading). Note that, in a state in which operation of the OA generation button 319 is disabled, the control device 11 does not accept an operation signal input by selecting the OA generation button 319, and in a state in which operation of the OA generation button 319 is enabled, the control device 11 accepts an operation signal input by selecting the OA generation button 319.

[0079] When a pharmacist selects the OA generation button 319 (an example of a second symbol) on the medication history generation screen 301 shown in FIG. 8 , the second medical record information acquisition unit 217 generates a second prompt for acquiring objective information and evaluation information using the prescription details (prescription information) displayed in the prescription details display field 321, the subjective information displayed in the S (subjective information) display field 323, and the treatment plan displayed in the P (treatment plan) display field 325. Examples of the second prompt include, but are not limited to, the aforementioned fixed format containing the description and description example of the SOAP medication history, and instructions to generate O (objective information) and A (evaluation information) information based on the prescription information, subjective information, and the contents of the treatment plan. Note that the description of the SOAP medication history, the description example of the SOAP medication history, the method for generating the second prompt, and the information included in the second prompt (such as the role and output format specification) are similar to those of the prompt for acquiring the estimated subjective information and estimated treatment plan described above, and therefore will not be described here.

[0080] The second medical record information acquisition unit 217 inputs the second prompt generated in this manner into the generative model 20, and acquires the information O and A generated by the generative model 20 as objective information and evaluation information, respectively. When the objective information and evaluation information are acquired by the second medical record information acquisition unit 217, the display control unit 205 displays them in the O (objective information) display field 333 and the A (evaluation information) display field 335, respectively.

[0081] 9 is a diagram showing an example of the medication history generation screen 301 used to generate a medication history in this embodiment, and shows an example of a state in which objective information is displayed in the O (objective information) display field 333 and evaluation information is displayed in the A (evaluation information) display field 335. The display control unit 205 displays the objective information displayed in the O (objective information) display field 333 and the evaluation information displayed in the A (evaluation information) display field 335 so that they can be modified in accordance with a modification operation by the pharmacist. Therefore, if the objective information or evaluation information needs to be modified, the pharmacist can modify it by a modification operation.

[0082] In this embodiment, an example has been described in which all of the subjective information, objective information, evaluation information, and treatment plan are displayed so that they can be modified in response to the pharmacist's operations, but this is not limited to this, and at least one of the subjective information, objective information, evaluation information, and treatment plan may be displayed so that they can be modified in response to the pharmacist's operations.

[0083] If there is no problem with the subjective information, objective information, evaluation information, and treatment plan displayed on the medication history generation screen 301, the pharmacist performs a registration operation using a registration button (not shown). As a result, the medical record generation unit 213 generates a medication history with the subjective information, objective information, evaluation information, and treatment plan as S, O, A, and P, respectively, and stores the medication history in the auxiliary storage device 13.

[0084] As described above, the medical record generation support device 10 of this embodiment supports the generation of a medication history by utilizing a generative model (generation AI), but in both cases of generating subjective information and a treatment plan and generating objective information and evaluation information, the generation of a prompt is automated by button operation (selection of the SP generation button 317 or the OA generation button 319). Therefore, according to this embodiment, even when a medication history is generated using a generation AI, there is no need to be aware of the prompt, so even pharmacists who are not familiar with generation AI can easily generate a medication history.

[0085] Next, the operation of the medical record creation support system of this embodiment will be described.

[0086] FIG. 10 is a flowchart showing an example of the process of displaying estimated subjective information and estimated treatment plans performed by the medical record generation support device 10 of this embodiment.

[0087] First, the medical treatment information acquisition unit 201 acquires, as medical treatment information, information on a prescription prescribed for a patient's symptoms (step S101).

[0088] Next, in response to a display instruction given by the pharmacist, the display control unit 205 displays the medication history generation screen 301 on the display device 15 (step S103).

[0089] Next, when the pharmacist selects a prescription to be used for generating the estimated subjective information and estimated treatment plan in the prescription selection display field 311 of the medication history generation screen 301 (step S105), the display control unit 205 displays information about the selected prescription (text information of the prescription acquired by the medical treatment information acquisition unit 201) in the prescription content display field 321 (step S107).

[0090] Next, when the pharmacist selects the estimated SP generation button 313 on the medication history generation screen 301 (step S109), the estimated information generation unit 203 generates a prompt to obtain estimated subjective information and an estimated treatment plan using the prescription information displayed in the prescription content display field 321 (step S111).

[0091] Next, the inferred information generating unit 203 inputs the generated prompt to the generative model 20, and acquires the information S and P generated by the generative model 20 as inferred subjective information and an inferred treatment plan, respectively (step S113).

[0092] Next, when the estimated subjective information and estimated treatment plan are acquired by the estimated information generating unit 203, the display control unit 205 displays them in the S (subjective information) display field 323 and the P (treatment plan) display field 325, respectively (step S115).

[0093] Fig. 11 is a flowchart showing an example of medication history generation processing performed in the medical record generation support device 10 of this embodiment. Note that the medication history generation processing shown in Fig. 11 is intended to be performed after a pharmacist has provided medication instructions to a patient, and is performed following the processing shown in Fig. 10.

[0094] First, the dialogue voice acquisition unit 207 acquires the voice of the dialogue when the pharmacist gives medication instructions to the patient, which is collected (input) by the voice input device 16 (see FIG. 2) (step S201).

[0095] Next, the voice recognition unit 209 recognizes the voice acquired by the dialogue voice acquisition unit 207, generates information (text information) indicating the content of the dialogue, and stores it in the auxiliary storage device 13 (step S203).

[0096] Next, when the pharmacist selects the dialogue to be used for generating the medication history in the dialogue information selection display field 315 of the medication history generation screen 301 (step S205), the display control unit 205 displays the voice recognition results of the selected dialogue in the dialogue content display field 331 (step S207).

[0097] Next, when the pharmacist selects the SP generation button 317 on the medication history generation screen 301 (step S209), the first medical record information acquisition unit 215 generates a first prompt for acquiring subjective information and a treatment plan using the dialogue content (voice recognition results of the dialogue) displayed in the dialogue content display field 331 (step S211).

[0098] Next, the first medical record information acquisition unit 215 inputs the generated first prompt into the generative model 20, and acquires the S and P information generated by the generative model 20 as subjective information and a treatment plan, respectively (step S213).

[0099] Next, when the subjective information and treatment plan are acquired by the first medical record information acquisition unit 215, the display control unit 205 displays them in the S (subjective information) display field 323 and the P (treatment plan) display field 325, respectively (step S215).

[0100] Next, when the pharmacist selects the OA generation button 319 on the medication history generation screen 301 (step S217), the second medical record information acquisition unit 217 generates a second prompt for acquiring objective information and evaluation information using the prescription content (prescription information) displayed in the prescription content display field 321, the subjective information displayed in the S (subjective information) display field 323, and the treatment plan displayed in the P (treatment plan) display field 325 (step S219).

[0101] Next, the second medical record information acquisition unit 217 inputs the generated second prompt into the generative model 20 and acquires the information O and A generated by the generative model 20 as objective information and evaluation information, respectively (step S221).

[0102] Next, when the second medical record information acquisition unit 217 acquires the objective information and the evaluation information, the display control unit 205 displays them in the O (objective information) display field 333 and the A (evaluation information) display field 335, respectively (step S223).

[0103] As described above, in this embodiment, the generation of a medication history is supported using a generative model (generative AI), so there is no need to develop a function for estimating the items to be recorded in the medication history, and the support for the generation of a medication history can be easily realized. This reduces development costs and the introduction costs of introducing a medication history generation support function.

[0104] Furthermore, in this embodiment, the estimated subjective information generation unit 203 can generate the estimated subjective information and the estimated treatment plan before the pharmacist engages in a dialogue during medication instruction to the patient, and the display control unit 205 can display them on the display device 15. This allows the pharmacist to check the estimated subjective information and the estimated treatment plan displayed on the display device 15 before providing medication instruction to the patient, allowing them to prepare in advance before providing medication instruction to the patient. Therefore, this embodiment allows even pharmacists who have difficulty eliciting sufficient information from patients during medication instruction, such as those with little practical experience or insufficient pharmaceutical knowledge, to improve the quality of the information they obtain. As described above, this embodiment is expected to achieve a high level of uniformity in the content of information elicited from patients during medication instruction, regardless of the pharmacist's skill, and also to improve the accuracy of the medication history generated using the dialogue during medication instruction, as described below.

[0105] Furthermore, in this embodiment, S and P of the SOAP medication history are generated using the content of the conversation between the pharmacist and the patient when providing medication instructions, and then O and A of the SOAP medication history are generated using the generated S and P and prescription information. Therefore, according to this embodiment, the accuracy of the contents written in the SOAP medication history can be improved compared to when S, O, A, and P of the SOAP medication history are generated all at once using the conversation between the pharmacist and the prescription information when providing medication instructions to the patient.

[0106] Furthermore, in this embodiment, a generative model (generative AI) is used to support the generation of a medication history, but in both cases of generating subjective information and a treatment plan, and generating objective information and evaluation information, the generation of a prompt is automated by button operation (selection of the SP generation button 317 or the OA generation button 319). Therefore, according to this embodiment, even when a medication history is generated using a generation AI, there is no need to be aware of the prompt, so even pharmacists who are not familiar with generation AI can easily generate a medication history.

[0107] (Variation 1) In the above embodiment, the creation of medical records by medical professionals is explained as an example in which a pharmacist creates a medication history, but a doctor may also create a medical record. In this case, medical procedure information may be test information, such as the results of examining the patient's symptoms, rather than prescription information. Furthermore, the dialogue between a medical professional and a patient regarding the patient's symptoms may be the dialogue between a doctor and a patient that takes place during an examination, rather than the dialogue between a pharmacist and a patient providing medication instructions.

[0108] (Variation 2) In the above embodiment, the generative model (generative AI) is an AI chat service in which large-scale language models (LLMs) are fine-tuned, but it may be realized by small-scale language models (SLMs), etc. When the generative model (generative AI) is realized by a small-scale language model, the medical record generation support device 10 may be provided with the generative model (generative AI).

[0109] (Variation 3) In the above embodiment, if it is possible to obtain past patient information (prescriptions, etc.), medication history, medication information, etc., these may be used to generate a prompt. The medication information may include contraindication information, etc. Furthermore, by using this information, it may be possible not only to generate a medication history, but also to point out errors in the prescription content, etc.

[0110] (program) The programs executed by the medical record generation support device 10 of the above embodiment and each of the above variants are provided as files in an installable or executable format stored on a computer-readable storage medium such as a CD-ROM, CD-R, memory card, DVD, or flexible disk (FD).

[0111] The programs executed by the medical record generation support device 10 of the above embodiment and each of the above modifications may be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. The programs executed by the medical record generation support device 10 of the above embodiment and each of the above modifications may be provided or distributed via a network such as the Internet. The programs executed by the medical record generation support device 10 of the above embodiment and each of the above modifications may be provided by being pre-installed in a ROM or the like.

[0112] The programs executed by the medical record generation support device 10 of the above embodiment and each of the above modifications have a modular configuration for realizing the above-mentioned units on a computer. In terms of actual hardware, for example, the CPU reads the learning program from the HDD onto the RAM and executes it, thereby realizing the above-mentioned units on a computer.

[0113] The above-described embodiment and each of the modifications merely illustrate examples of implementations of the present disclosure, and the technical scope of the present disclosure should not be construed as being limited by these. Therefore, the present disclosure can be implemented in various forms without departing from the spirit or main features thereof. For example, the above-described embodiment and each of the modifications may be appropriately combined in their respective constituent units. Furthermore, for example, some components may be deleted from all components in the above-described embodiment and each of the modifications.

[0114] The present disclosure includes the following aspects.

[0115] (1) a medical treatment information acquisition unit that acquires medical treatment information, which is information on prescriptions or tests related to a patient's symptoms; a speech recognition result acquisition unit that acquires a speech recognition result of a conversation regarding the symptom between a medical professional and the patient; a medical record generation unit that inputs the medical treatment information and the speech recognition result into a generation model, acquires from the generation model subjective information complained by the patient about the symptom, objective information that objectively captures the symptom, evaluation information that indicates the evaluation content of the symptom, and a treatment plan for the symptom, and generates a medical record; and A medical record creation support device comprising:

[0116] (2) The medical record generation unit a first medical record information acquisition unit that inputs a first prompt using at least the speech recognition result into the generative model and acquires the subjective information and the treatment plan from the generative model; a second medical record information acquisition unit that inputs the subjective information and the treatment plan acquired by the first medical record information acquisition unit and a second prompt using the medical treatment information into the generative model, and acquires the objective information and the evaluation information from the generative model; The medical record creation support device according to (1) above.

[0117] (3) a speculative information generating unit that generates speculative subjective information by using the medical treatment information and a speculative treatment plan by using the medical treatment information; a display control unit that displays the estimated subjective information and the estimated treatment plan. The medical record creation support device according to (1) or (2) above.

[0118] (4) The medical treatment information includes information on prescriptions prescribed for the symptoms, the inferred information generator generates the inferred subjective information and the inferred treatment plan before the interaction occurs; The display control unit is capable of displaying the estimated subjective information and the estimated treatment plan before the interaction occurs. The medical record creation support device according to (3) above.

[0119] (5) the first prompt includes an instruction to extract the subjective information and the treatment plan from the speech recognition result; The medical record creation support device according to (2) above.

[0120] (6) The medical records are recorded in SOAP format; The subjective information corresponds to S in the SOAP The treatment plan corresponds to P in the SOAP, the first prompt further includes at least one of a description of the SOAP in the medical record and an example description of the medical record in the SOAP format; The medical record creation support device according to (5) above.

[0121] (7) The second prompt is: instructions for generating the objective information and the evaluation information based on the subjective information, the treatment plan, and the medical action information; The medical record creation support device according to (5) or (6) above.

[0122] (8) The medical records are recorded in SOAP format; The subjective information corresponds to S in the SOAP The objective information corresponds to "O" in the SOAP The evaluation information corresponds to A of the SOAP, The treatment plan corresponds to P in the SOAP, the second prompt further includes at least one of a description of the SOAP in the medical record and an example description of the medical record in the SOAP format; The medical record creation support device according to (7) above.

[0123] (9) A display control unit controls the display of a medical record generation screen used to generate the medical record. the display control unit displays the voice recognition result and the medical treatment information in a dialogue content display field and a prescription content display field included in the medical record generation screen, respectively; the first medical record information acquisition unit generates the first prompt and inputs it into the generative model in response to an operation of selecting a first symbol included in the medical record generation screen; the display control unit displays the subjective information and the treatment plan in a subjective information display field and a treatment plan display field included in the medical record generation screen, respectively. The medical record creation support device according to (2) above.

[0124] (10) The second medical record information acquisition unit generates the second prompt and inputs it into the generative model in response to an operation of selecting a second symbol included in the medical record generation screen; the display control unit displays the objective information and the evaluation information in an objective information display field and an evaluation information display field included in the medical record generation screen, respectively. The medical record creation support device according to (9) above.

[0125] (11) The display control unit controls to disable an operation of selecting the second symbol before displaying the subjective information and the treatment plan, and to enable an operation of selecting the second symbol after displaying the subjective information and the treatment plan. The medical record creation support device according to (10) above.

[0126] (12) The display control unit displays at least one of the subjective information, the objective information, the evaluation information, and the treatment plan in a manner that allows modification in response to an operation by the medical professional. The medical record creation support device according to (10) above.

[0127] (13) a medical treatment information acquisition step in which a medical treatment information acquisition unit acquires medical treatment information, which is information on prescriptions or tests related to symptoms of a patient; a speech recognition result acquisition step in which a speech recognition result acquisition unit acquires a speech recognition result of a conversation regarding the symptom between a medical professional and the patient; a medical record generation step in which a medical record generation unit inputs the medical treatment information and the speech recognition result into a generation model, acquires from the generation model subjective information complained by the patient about the symptom, objective information that objectively captures the symptom, evaluation information that indicates the evaluation content of the symptom, and a treatment plan for the symptom, and generates a medical record; A medical record generation support method including:

[0128] (14) a medical treatment information acquisition unit that acquires medical treatment information, which is information on prescriptions or tests related to the patient's symptoms; a speech recognition result acquisition unit that acquires a speech recognition result of a conversation regarding the symptom between a medical professional and the patient; a medical record generation unit that inputs the medical treatment information and the speech recognition result into a generation model, acquires from the generation model subjective information complained by the patient about the symptom, objective information that objectively captures the symptom, evaluation information that indicates the evaluation content of the symptom, and a treatment plan for the symptom, and generates a medical record; and A program that allows a computer to function. [Explanation of symbols]

[0129] 1 Medical record generation support system 2 Network 10 Medical record creation support device 20 Generative Model 201 Medical Practice Information Acquisition Department 203 Inferred information generation unit 205 Display control unit 207 Dialogue voice acquisition unit 209 Speech Recognition Unit 211 Speech recognition result acquisition unit 213 Medical Record Generation Department 215 First Medical Record Information Acquisition Department 217 Second Medical Record Information Acquisition Department

Claims

1. a medical treatment information acquisition unit that acquires medical treatment information, which is information on prescriptions or tests related to the patient's symptoms; a speech recognition result acquisition unit that acquires a speech recognition result of a conversation regarding the symptom between a medical professional and the patient; a medical record generation unit that inputs the medical treatment information and the speech recognition result into a generation model, acquires from the generation model subjective information complained by the patient about the symptom, objective information that objectively captures the symptom, evaluation information that indicates the evaluation content of the symptom, and a treatment plan for the symptom, and generates a medical record; and A medical record creation support device comprising:

2. The medical record generation unit a first medical record information acquisition unit that inputs a first prompt using at least the speech recognition result into the generative model and acquires the subjective information and the treatment plan from the generative model; a second medical record information acquisition unit that inputs the subjective information and the treatment plan acquired by the first medical record information acquisition unit and a second prompt using the medical action information into the generation model, and acquires the objective information and the evaluation information from the generation model; The medical record creation support device according to claim 1 .

3. an inferred information generating unit that generates inferred subjective information by inferring the subjective information and an inferred treatment plan by inferring the treatment plan using the medical treatment information; a display control unit that displays the estimated subjective information and the estimated treatment plan.

3. The medical record creation support device according to claim 1.

4. The medical treatment information includes information on a prescription prescribed for the symptom, the inferred information generator generates the inferred subjective information and the inferred treatment plan before the interaction occurs; The display control unit is capable of displaying the estimated subjective information and the estimated treatment plan before the interaction occurs. The medical record creation support device according to claim 3 .

5. the first prompt includes an instruction to extract the subjective information and the treatment plan from the speech recognition result. The medical record creation support device according to claim 2 .

6. The medical record is recorded in SOAP format; The subjective information corresponds to S in SOAP, The treatment plan corresponds to P in the SOAP, the first prompt further includes at least one of a description of the SOAP in the medical record and an example description of the medical record in the SOAP format; The medical record creation support device according to claim 5 .

7. The second prompt: instructions for generating the objective information and the evaluation information based on the subjective information, the treatment plan, and the medical action information; 7. The medical record creation support device according to claim 5 or 6.

8. The medical record is recorded in SOAP format; The subjective information corresponds to S in SOAP, The objective information corresponds to O in the SOAP, The evaluation information corresponds to A of the SOAP The treatment plan corresponds to P in the SOAP, the second prompt further includes at least one of a description of the SOAP in the medical record and an example description of the medical record in the SOAP format; The medical record creation support device according to claim 7.

9. a display control unit that controls the display of a medical record generation screen used to generate the medical record; the display control unit displays the voice recognition result and the medical treatment information in a dialogue content display field and a prescription content display field included in the medical record generation screen, respectively; the first medical record information acquisition unit generates the first prompt and inputs the first prompt to the generative model in response to an operation of selecting a first symbol included in the medical record generation screen; the display control unit displays the subjective information and the treatment plan in a subjective information display field and a treatment plan display field included in the medical record generation screen, respectively. The medical record creation support device according to claim 2 .

10. the second medical record information acquisition unit generates the second prompt and inputs the second prompt to the generative model in response to an operation of selecting a second symbol included in the medical record generation screen; the display control unit displays the objective information and the evaluation information in an objective information display field and an evaluation information display field included in the medical record generation screen, respectively. The medical record creation support device according to claim 9.

11. the display control unit controls to disable an operation of selecting the second symbol before the subjective information and the treatment plan are displayed, and to enable an operation of selecting the second symbol after the subjective information and the treatment plan are displayed. The medical record creation support device according to claim 10.

12. the display control unit displays at least one of the subjective information, the objective information, the evaluation information, and the treatment plan in a manner that allows modification in response to an operation by the medical professional. The medical record creation support device according to claim 10.

13. a medical treatment information acquisition step in which a medical treatment information acquisition unit acquires medical treatment information, which is information on prescriptions or tests related to symptoms of a patient; a speech recognition result acquisition step in which a speech recognition result acquisition unit acquires a speech recognition result of a conversation regarding the symptom between a medical professional and the patient; a medical record generation step in which a medical record generation unit inputs the medical treatment information and the speech recognition result into a generation model, acquires from the generation model subjective information complained by the patient about the symptom, objective information that objectively captures the symptom, evaluation information that indicates the evaluation content of the symptom, and a treatment plan for the symptom, and generates a medical record; A medical record generation support method including:

14. a medical treatment information acquisition unit that acquires medical treatment information, which is information on prescriptions or tests related to the patient's symptoms; a speech recognition result acquisition unit that acquires a speech recognition result of a conversation regarding the symptom between a medical professional and the patient; a medical record generation unit that inputs the medical treatment information and the speech recognition result into a generation model, acquires from the generation model subjective information complained by the patient about the symptom, objective information that objectively captures the symptom, evaluation information that indicates the evaluation content of the symptom, and a treatment plan for the symptom, and generates a medical record; and A program that allows a computer to function.

Images