Support device, support method and program

Abstract

To provide technique for efficiently recording medication history information on a patient.SOLUTION: The present invention relates to a support device that has an acquisition part which acquires data on a conversation between a pharmacist and a patient in medication guidance; a model interface part which inputs a prompt for generating a summary of the conversation data to a generation model and then acquiring summary data from the generation model; and a support part which provides the summary data for the pharmacist.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present disclosure relates to an assistance device, an assistance method, and a program. [Background technology]

[0002] With the recent advancement of information and communication technology, it has been increasingly used in the medical field. For example, in electronic medical record systems, patient information, diagnosis information, treatment information, medication information, etc. are managed by servers.

[0003] Not only in hospitals but also in dispensing pharmacies, various information and data are now being systematized and organized into databases, and managed on servers, etc. For example, the history of drugs prescribed to patients, i.e., medication history information related to medication history, is managed on a server. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2018-197926 [Patent Document 2] Patent Publication No. 2021-136018 [Patent Document 3] Japanese Patent Publication No. 2022-022129 Summary of the Invention [Problem to be solved by the invention]

[0005] The details of medication instructions given must also be registered in the medication history information. Pharmacists working at dispensing pharmacies use personal computers installed in the pharmacy to record the details of medication instructions as medication history information for patients for whom they have handed over prescription drugs and given medication instructions.

[0006] However, just dispensing prescription drugs, handing them out to many patients, and providing medication instructions takes up a lot of pharmacists' time, and recording the details of each patient's medication instructions as medication history information takes even more time, which can lead to long working hours.

[0007] In view of the above problems, one object of the present disclosure is to provide a technology for efficiently recording a patient's medication history information. [Means for solving the problem]

[0008] One aspect of the present disclosure relates to an assistance device having an acquisition unit that acquires conversation data between a pharmacist and a patient during medication instructions, a model interface unit that inputs prompts for generating a summary of the conversation data into a generative model and acquires summary data from the generative model, and an assistance unit that provides the summary data to the pharmacist. [Effects of the Invention]

[0009] According to the present disclosure, patient medication history information can be efficiently recorded. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a schematic diagram illustrating an assistance device according to one embodiment of the present disclosure. [Figure 2] FIG. 2 is a block diagram illustrating a hardware configuration of the support device according to an embodiment of the present disclosure. [Figure 3] FIG. 3 is a block diagram illustrating a functional configuration of the assistance device according to an embodiment of the present disclosure. [Figure 4] FIG. 4 is a schematic diagram illustrating a process for obtaining summary data according to one embodiment of the present disclosure. [Figure 5] 5A and 5B are diagrams illustrating prompts input to a generative model and output results according to one embodiment of the present disclosure. [Figure 6] FIG. 6 is a diagram illustrating a medication history editing screen according to one embodiment of the present disclosure. [Figure 7]FIG. 7 is a flowchart illustrating a support process according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

[0012] In the following embodiment, a support device is disclosed that supports a pharmacist in recording a patient's medication history information.

[0013] [Summary] In the embodiment described below, when a pharmacist hands over a dispensed prescription drug to a patient, audio or video conversation data of medication instructions given to the patient is collected, and then the support device 100 generates text-based summary data of the conversation content to be input as medication history information from the collected conversation data using the generative model 20. Then, the pharmacist can use or edit the summary data obtained from the support device 100 to record the content of the medication instructions given to the patient as medication history information.

[0014] 1 is a schematic diagram illustrating a support device according to an embodiment of the present disclosure. As shown in FIG. 1, an information processing terminal 10 is an information processing device such as a personal computer or tablet that is installed in a pharmacy and used by a pharmacist to perform administrative tasks, and stores conversation data collected by a microphone or video camera during medication instruction between the pharmacist and a patient as audio data or video data.

[0015] For example, medication instructions from a pharmacist to a patient may be given when handing over a prescribed drug at a pharmacy counter. In this case, the microphone and video camera may be installed, for example, at a counter where the pharmacist and patient interact, or may be built into the information processing terminal 10 if the information processing terminal 10 is installed at a counter. Furthermore, in order to better distinguish between what the pharmacist is saying and what the patient is saying, a microphone for collecting the pharmacist's speech and a microphone for collecting the patient's speech may be installed.

[0016] Alternatively, medication instructions to patients by pharmacists may be provided remotely. In this case, video data recording interactions with patients via a remote tool for video conferencing may be stored in the information processing terminal 10.

[0017] When a pharmacist creates medication history information from the contents of medication instructions, the pharmacist operates the information processing terminal 10 to request the support device 100 to generate text-based summary data from the collected conversation data. The support device 100 uses the generative model 20 on the cloud server 200 to acquire the summary data from the conversation data.

[0018] For example, if the generative model 20 is a multimodal type generative model capable of generating text-based summary data from audio data and / or video data, the assistance device 100 transmits conversation data and a prompt for causing the generative model 20 to output summary data from the conversation data to the cloud server 200, and acquires the summary data as a processing result of the generative model 20 on the conversation data. The assistance device 100 may then display the acquired summary data on an editing screen of the information processing terminal 10.

[0019] On the other hand, if the generative model 20 is not a multimodal type generative model but is composed of a speech recognition model 21 that generates text-format conversation data from audio data and / or video data, and a language model 22 that generates processing results from text-based prompts such as ChatGPT (registered trademark) by OpenAI, the assistance device 100 first transmits the conversation data to the cloud server 200 and acquires the text-format conversation data as the speech recognition result by the speech recognition model 21. Upon acquiring the text-format conversation data, the assistance device 100 then generates a prompt to cause the language model 22 to generate summary data from the text-format conversation data, and transmits the generated prompt to the cloud server 200. Upon acquiring the summary data as the processing result by the language model 22 from the cloud server 200, the assistance device 100 may display the acquired summary data on an editing screen of the information processing terminal 10 (see FIG. 4).

[0020] In this way, the support process described in the following examples collects audio or video conversation data exchanged between a pharmacist and a patient, and uses the generative model 20 to obtain text summary data of the conversation content from the collected conversation data. By appropriately editing the obtained summary data, the pharmacist can record the contents of oral medication instructions given to the patient as medication history information, thereby reducing the burden on the pharmacist of manually entering the interactions with the patient during medication instructions.

[0021] Here, the support device 100 is realized by any type of information processing device such as a server, and may have a hardware configuration such as that shown in Fig. 2. That is, the support device 100 has a drive device 101, a storage device 102, a memory device 103, a processor 104, a user interface (UI) device 105, and a communication device 106, which are interconnected via a bus B.

[0022] Programs or instructions for implementing the various functions and processes of the assistance device 100, which will be described later, may be stored on a removable storage medium such as a CD-ROM (Compact Disk-Read Only Memory) or flash memory. When the storage medium is inserted into the drive device 101, the programs or instructions are installed from the storage medium to the storage device 102 or the memory device 103 via the drive device 101. However, the programs or instructions do not necessarily have to be installed from the storage medium; they may be downloaded from an external device via a network or the like. The storage device 102 is implemented by a hard disk drive or the like and stores installed programs or instructions as well as files, data, etc. used to execute the programs or instructions. The memory device 103 is implemented by a random access memory, static memory, etc. and, when a program or instruction is activated, reads and stores the program, instruction, data, etc. from the storage device 102. The storage device 102, the memory device 103, and the removable storage medium may be collectively referred to as a non-transitory storage medium.

[0023] The processor 104 may be realized by one or more central processing units (CPUs), graphics processing units (GPUs), processing circuitry, etc., each of which may be composed of one or more processor cores. The processor 104 executes various functions and processes of the assistance device 100 (described below) in accordance with programs, instructions, and data such as parameters required to execute the programs or instructions stored in the memory device 103. The user interface (UI) device 105 may be composed of input devices such as a keyboard, mouse, camera, and microphone, output devices such as a display, speaker, headset, and printer, and input / output devices such as a touch panel, and implements an interface between the user and the assistance device 100. For example, the user operates the assistance device 100 by manipulating a graphical user interface (GUI) displayed on a display or touch panel using a keyboard, mouse, etc. The communication device 106 is realized by various communication circuits that execute communication processes with external devices and communication networks such as the Internet and a local area network (LAN).

[0024] However, the above-described hardware configuration is merely an example, and the assistance device 100 according to the present disclosure may be realized by any other appropriate hardware configuration. Furthermore, the information processing terminal 10 and the cloud server 200 may also have the illustrated hardware configuration. Furthermore, although the assistance device 100 and the information processing terminal 10 are physically different devices in the embodiments of the present disclosure, the assistance device 100 and the information processing terminal 10 according to the present disclosure may be realized as the same physical device. In this case, for example, the processing and functions of the assistance device 100 described below may be realized by an application program or the like installed on the information processing terminal 10.

[0025] [Support device] Next, an assistance device 100 according to an embodiment of the present disclosure will be described. Fig. 3 is a block diagram showing a functional configuration of the assistance device 100 according to an embodiment of the present disclosure. As shown in Fig. 3, the assistance device 100 includes an acquisition unit 110, a model interface unit 120, and an assistance unit 130. For example, one or more functional units of the acquisition unit 110, the model interface unit 120, and the assistance unit 130 may be realized by one or more processors 104 executing one or more programs or instructions.

[0026] The acquisition unit 110 acquires conversation data between a pharmacist and a patient during medication instruction. For example, the conversation data may be audio data of the content of conversations between the pharmacist and the patient during medication instruction, or may be video data of the conversation between the pharmacist and the patient during medication instruction.

[0027] For example, if the conversation data is voice data, the conversation data can be voice data collected by a microphone or the like. In this case, the voice data collected by the microphone is transmitted to the information processing terminal 10 via wired and / or wireless communication in real time or at any timing, and is stored in the information processing terminal 10. Then, for example, when a pharmacist attempts to create medication history information for the patient, the conversation data of the patient stored in the information processing terminal 10 is transmitted to the assistance device 100, and the acquisition unit 110 acquires the transmitted voice-format conversation data.

[0028] Furthermore, when the conversation data is video data, the conversation data may be video data captured by a video camera or the like. In this case, the video data captured by the video camera is transmitted to the information processing terminal 10 via wired and / or wireless communication in real time or at any timing, and is stored in the information processing terminal 10. Then, for example, when a pharmacist attempts to create medication history information for the patient, the conversation data of the patient stored in the information processing terminal 10 is transmitted to the assistance device 100, and the acquisition unit 110 acquires the transmitted video-format conversation data.

[0029] The model interface unit 120 inputs a prompt for generating a summary of the conversation data to the generative model 20 and acquires the summary data from the generative model 20. Specifically, the generative model 20 may be a multimodal type generative model that can accept input data in not only text format but also image format, video format, and various other formats. In this case, the model interface unit 120 transmits to the cloud server 200 all or part of the audio or video format conversation data acquired by the acquisition unit 110, as well as a prompt instructing the cloud server 200 to generate text format summary data based on the conversation content of the conversation data.

[0030] The cloud server 200 inputs the conversation data and prompts received from the assistance device 100 into the generative model 20, and acquires summary data as a processing result from the generative model 20. Then, the cloud server 200 returns the acquired summary data to the assistance device 100 as a processing result, and the model interface unit 120 acquires the summary data.

[0031] 4, the generation model 20 may be composed of a speech recognition model 21 that extracts speech recognition results from audio data or video data, and a language model 22 that outputs generation results corresponding to text-format prompts. The model interface unit 120 inputs conversation data to the speech recognition model 21, obtains recognition result data of the speech recognition from the speech recognition model 21, inputs a prompt for generating a summary of the recognition result data to the language model 22, and obtains summary data from the language model 22.

[0032] Specifically, the model interface unit 120 first transmits conversation data, such as audio data and video data, that record the exchange of medication instruction to the cloud server 200, and requests the cloud server 200 to perform voice recognition on the conversation data. Upon receiving the conversation data, the cloud server 200 inputs the received conversation data into the voice recognition model 21, obtains recognition result data in text format that describes the content of the utterances in the conversation data from the voice recognition model 21, and transmits the obtained recognition result data to the assistance device 100.

[0033] When the recognition result data in text format is received from the cloud server 200, the model interface unit 120 creates a prompt to cause the language model 22 to generate summary data of the utterance content during medication instruction described in the recognition result data, and transmits the created prompt including the recognition result data to the cloud server 200. Upon receiving this prompt, the cloud server 200 inputs the received prompt into the language model 22 and obtains summary data summarizing the utterance content of the recognition result data from the language model 22. Then, the cloud server 200 transmits the obtained summary data to the assistance device 100.

[0034] An example of a prompt for generating such summary data is shown in Fig. 5A. First, the model interface unit 120 writes a predetermined phrase in the prompt, such as "Please create a summary of the text data below ### in accordance with the following medication history format," and causes the language model 22 to generate a summary of the "text data," which is all or part of the recognition result data, in accordance with the "medication history format" specified below.

[0035] Next, the model interface unit 120 writes the text data to be summarized in "### text data...". Furthermore, the model interface unit 120 writes a predetermined phrase, "### medication history format...", in the prompt to create the summary data in a predetermined format. The medication history format may be specified by each pharmacist, for example, or may be specified as a common format for a pharmacy, pharmacy chain, etc.

[0036] 5A is input to the language model 22, summary data such as that shown in FIG. 5B is output from the language model 22. Upon acquiring the summary data, the model interface unit 120 provides the acquired summary data to the support unit 130.

[0037] It should be noted that the prompts shown are merely examples and may describe any other appropriate instruction content.

[0038] For example, the model interface unit 120 may generate a prompt corresponding to user information of the assistance device 100. Specifically, the model interface unit 120 may change the prompt according to attribute information of a user who is logged in to the information processing terminal 10. For example, the model interface unit 120 may determine which pharmacy or company the logged-in user belongs to, and write a medication history form corresponding to the determined pharmacy to which the user belongs in the prompt. Specifically, if the model interface unit 120 determines that the logged-in user belongs to pharmacy chain X, the model interface unit 120 may write a medication history form adopted by pharmacy chain X in the prompt.

[0039] For example, pharmacy A has a high demand for ophthalmology prescriptions, so a medication history form specialized for ophthalmology may be written in the prompt. Specifically, the medication history form may be defined so that the method of applying eye drops, the number of times, the order of use, etc. can be accurately recorded. In addition, the medication history form may be defined so that eye-related confirmation items are presented as confirmation items for the next visit.

[0040] The model interface unit 120 may also generate a prompt corresponding to the prescription state. The model interface unit 120 may change the prompt depending on the state or situation in which the user logged in to the information processing terminal 10 provided medication instruction. For example, the model interface unit 120 may determine whether the logged in user provided medication instruction in an "outpatient," "home," or "hospital ward" prescription, and write a medication history form corresponding to the determined prescription state in the prompt. Specifically, if it is determined that the logged in user provided medication instruction in a "hospital ward," the model interface unit 120 may write a medication history form corresponding to medication instruction in the "hospital ward" in the prompt.

[0041] Furthermore, the model interface unit 120 may generate a prompt corresponding to the issuer of the prescription (e.g., a hospital, a clinic, etc.) and / or the medical department. For example, if the issuer of the prescription is an obstetrics and gynecology clinic, the model interface unit 120 may consider that the speech data is likely to include medication instructions specific to women or pregnant women, and may describe in the prompt a medication history form including special notes regarding women or pregnant women (e.g., contraindicated medications, supplements, etc.).

[0042] The support unit 130 provides the summary data to the pharmacist. Specifically, when the summary data is acquired from the model interface unit 120, the support unit 130 may display the acquired summary data on the information processing terminal 10. For example, the support unit 130 may display an editing screen for editing the summary data, allowing the pharmacist to edit the summary data output from the generative model 20.

[0043] Specifically, the support unit 130 may display a medication history editing screen as shown in Fig. 6 on the information processing terminal 10, and display the summary data output from the generative model 20 in an editable format on the medication history editing screen. For example, when a pharmacist desires to add or correct the summary data output from the generative model 20, the pharmacist can operate the information processing terminal 10 and edit the summary data displayed on the medication history editing screen. This allows the pharmacist to add patient information that is not reflected in the summary data.

[0044] Furthermore, the support unit 130 may present medication instruction information to the pharmacist corresponding to the drug to be provided to the patient. For example, when a pharmacist is providing medication instruction to a patient while using the information processing terminal 10, the support unit 130 may provide medication instruction information to the information processing terminal 10 that suggests what kind of medication instruction should be given corresponding to the type of drug prescribed to the patient.

[0045] For example, the medication instruction information may include the purpose, content, or precautions of medication instruction for a drug, the name, usage and dosage, or side effects of the drug, precautions to be taken while taking the drug, or points to note regarding the patient's medication. Such medication instruction information may be displayed on the information processing terminal 10 by a pop-up or the like before or during medication instruction for the patient. Furthermore, when a pop-up medication instruction is performed, the pharmacist may be prompted to enter in a check box that the instruction has been performed. If not all of the pop-up medication instruction items have been performed, the support unit 130 may redisplay the unperformed items.

[0046] Furthermore, if the support device 100 holds medication history information for the patient, the support unit 130 may adjust the medication instruction information by referring to the held medication history information. For example, if it is registered in the medication history information that a patient forgot to take a medication during the previous medication instruction, the support unit 130 may include confirmation of the medication status in the medication instruction information for the current medication instruction.

[0047] Furthermore, when the assistance device 100 has acquired prescription information for a patient, the assistance unit 130 may adjust the medication instruction information by referring to the acquired prescription information. Specifically, when a disease is estimated based on the type of medication prescribed to the patient, the assistance unit 130 may adjust the medication instruction information in accordance with the estimated disease. For example, when it is estimated that the patient is suspected of having interstitial pneumonia based on the type of medication prescribed to the patient, the assistance unit 130 may adjust the medication instruction information to check the state of dry cough.

[0048] [Support Processing] Next, an assistance process according to an embodiment of the present disclosure will be described. Fig. 7 is a flowchart showing the assistance process according to an embodiment of the present disclosure. The assistance process is executed by the above-described assistance device 100, and more specifically, may be realized by one or more processors 104 of the assistance device 100 executing one or more programs or instructions stored in one or more memory devices 103.

[0049] 7, in step S101, the support device 100 acquires conversation data between a pharmacist and a patient during medication instruction. Specifically, the pharmacist operates the information processing terminal 10 to transmit conversation data in audio or video format, which records the exchange between the pharmacist and the patient during medication instruction, to the support device 100, and the support device 100 receives the conversation data from the information processing terminal 10 via the network.

[0050] In step S102, the assistance device 100 inputs a prompt for generating a summary of the conversation data to the generative model 20. Specifically, the assistance device 100 creates a prompt for causing the generative model 20 to summarize the conversation content of the conversation data in audio or video format acquired in step S101, and transmits the created prompt to the cloud server 200. The cloud server 200 inputs the received prompt to the generative model 20.

[0051] In step S103, the assistance device 100 acquires summary data from the generative model 20. Specifically, when the cloud server 200 acquires the summary data from the generative model 20 as a processing result corresponding to the prompt, the cloud server 200 transmits the acquired summary data to the assistance device 100.

[0052] In step S104, the support device 100 provides the summary data to the pharmacist. Specifically, when the support device 100 receives the summary data from the cloud server 200, it transmits the received summary data to the pharmacist's information processing terminal 10. At this time, the support device 100 may display the summary data on an editable medication history editing screen on the information processing terminal 10, allowing the pharmacist to add to and / or modify the summary data.

[0053] The above-described support device 100 and support process enable pharmacists to efficiently input the contents of medication instructions into medication history information, thereby improving the work efficiency of pharmacists.

[0054] In addition, the following supplementary notes are provided in relation to the above description. (Appendix 1) an acquisition unit that acquires conversation data between a pharmacist and a patient during medication instruction; a model interface unit that inputs a prompt for generating a summary of the conversation data into a generative model and acquires summary data from the generative model; a support unit that provides the summary data to the pharmacist; An assist device having: (Appendix 2) The generative model is a speech recognition model that extracts speech recognition results from audio data or video data; a language model that outputs a generation result in response to a text prompt; and The model interface unit inputting the conversation data into the speech recognition model, and obtaining recognition result data of the speech recognition from the speech recognition model; 2. The assistance device of claim 1, wherein a prompt for generating a summary of the recognition result data is input to the language model, and summary data is obtained from the language model. (Appendix 3) 3. The support device according to claim 1, wherein the support unit displays an editing screen for editing the summary data. (Appendix 4) 4. The assistance device according to any one of appendices 1 to 3, wherein the model interface unit generates the prompt in accordance with user information of the assistance device. (Appendix 5) 5. The assistance device according to any one of appendices 1 to 4, wherein the model interface unit generates the prompt in accordance with a prescription state. (Appendix 6) 6. The assistance device according to any one of claims 1 to 5, wherein the model interface unit generates a prompt corresponding to an issuer of a prescription or a medical department. (Appendix 7) 7. The support device according to any one of appendices 1 to 6, wherein the support unit presents medication instruction information to the pharmacist in accordance with the medication to be provided to the patient. (Appendix 8) The assistance device described in Appendix 7, wherein the medication instruction information includes the purpose, content, or precautions of the medication instruction for the medication, the name, usage and dosage, or side effects of the medication, precautions to take while taking the medication, or points to note regarding the patient's medication. (Appendix 9) Acquiring conversation data between a pharmacist and a patient during medication instruction; inputting prompts for generating a summary of the conversation data into a generative model and obtaining summary data from the generative model; providing the summary data to the pharmacist; and The computer performs the assisted method. (Appendix 10) Acquiring conversation data between a pharmacist and a patient during medication instruction; inputting prompts for generating a summary of the conversation data into a generative model and obtaining summary data from the generative model; providing the summary data to the pharmacist; and A program that causes a computer to execute the following.

[0055] Although the examples of the present disclosure have been described in detail above, the present disclosure is not limited to the specific embodiments described above, and various modifications and variations are possible within the scope of the gist of the present disclosure as set forth in the claims. [Industrial Applicability]

[0056] The present disclosure is useful for a medication history registration system. [Explanation of symbols]

[0057] 10 Information processing terminal 20 Generative Model 21 Speech Recognition Model 22 Language Models 100 Support equipment 110 Acquisition Department 120 Model Interface Section 130 Support Department

Claims

1. an acquisition unit that acquires conversation data between a pharmacist and a patient during medication instruction; a model interface unit that inputs a prompt for generating a summary of the conversation data into a generative model and acquires summary data from the generative model; a support unit that provides the summary data to the pharmacist; An assist device having:

2. The generative model is a speech recognition model that extracts speech recognition results from audio data or video data; a language model that outputs a generation result in response to a text prompt; and The model interface unit inputting the conversation data into the speech recognition model, and obtaining recognition result data of the speech recognition from the speech recognition model; The assistance device of claim 1 , wherein a prompt for generating a summary of the recognition result data is input to the language model, and summary data is obtained from the language model.

3. The support device according to claim 1 , wherein the support unit displays an edit screen for editing the summary data.

4. The assistance device according to claim 1 , wherein the model interface unit generates the prompt in response to user information of the assistance device.

5. The assistance device according to claim 1 , wherein the model interface unit generates the prompt in response to a prescription state.

6. The assistance device according to claim 1 , wherein the model interface unit generates a prompt corresponding to an issuer of a prescription or a medical department.

7. The support device according to claim 1 , wherein the support unit presents medication instruction information to the pharmacist in accordance with the medication to be provided to the patient.

8. The assistance device of claim 7, wherein the medication instruction information includes the purpose, content, or precautions of the medication instruction for the medication, the name, usage and dosage, or side effects of the medication, precautions to take while taking the medication, or points to note regarding the patient's medication.

9. Acquiring conversation data between a pharmacist and a patient during medication instruction; inputting prompts for generating a summary of the conversation data into a generative model and obtaining summary data from the generative model; providing the summary data to the pharmacist; and The computer performs the assisted method.

10. Acquiring conversation data between a pharmacist and a patient during medication instruction; inputting prompts for generating a summary of the conversation data into a generative model and obtaining summary data from the generative model; providing the summary data to the pharmacist; and A program that causes a computer to execute the following.

Images