Information processing system, information processing method, and program

Abstract

The aim is to reduce the workload of medical professionals in creating medical documents. [Solution] According to the present invention, there is provided an information processing system for assisting in the creation of medical documents, comprising a data acquisition unit and a medical document generation unit, the data acquisition unit configured to acquire conversation data and medical data, the conversation data including the content of a conversation between a medical professional and a patient, the medical data having pharmaceutical data, the pharmaceutical data having new pharmaceutical data and old pharmaceutical data, the new pharmaceutical data corresponding to the conversation in the conversation data, the old pharmaceutical data being data that precedes the new pharmaceutical data, and the medical document generation unit configured to generate the medical document based on the conversation data and the pharmaceutical data.

Description

[Technical field]

[0001] The present invention relates to an information processing system, an information processing method, and a program. [Background technology]

[0002] In addition to dispensing medicines prescribed by doctors, medical professionals (e.g., pharmacists at a pharmacy) also have the task of acquiring and managing patient data through conversations with patients. Conventionally, technologies to support the work of medical professionals have been proposed, and Patent Document 1 discloses a system for improving the quality of medication instruction work. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-047624 Summary of the Invention [Problem to be solved by the invention]

[0004] The work of medical professionals includes the task of creating medical documents to record interactions with patients. For example, in situations where medical professionals have other tasks that require their attention, such as dispensing medicines and providing medication instructions, creating medical documents is a burden for them, but it is a necessary task that cannot be omitted.

[0005] The present invention aims to reduce the workload of medical professionals in creating medical documents. [Means for solving the problem]

[0006] [1] According to the present invention, there is provided an information processing system for assisting in the creation of medical documents, comprising a data acquisition unit and a medical document generation unit, the data acquisition unit is configured to acquire conversation data and medical data, the conversation data includes the content of a conversation between a medical professional and a patient, the medical data has pharmaceutical data, the pharmaceutical data has new pharmaceutical data and old pharmaceutical data, the new pharmaceutical data corresponds to the conversation of the conversation data, the old pharmaceutical data is data that precedes the new pharmaceutical data, and the medical document generation unit is configured to generate the medical document based on the conversation data and the pharmaceutical data.

[0007] According to the present invention, medical documents can be generated using conversation data between medical professionals and patients and pharmaceutical data from medical data, thereby reducing the workload of medical professionals in creating medical documents.

[0008] Various embodiments of the present invention are exemplified below. The embodiments shown below can be combined with each other. [2] Preferably, there is provided an information processing system according to [1], in which the conversation data is voice conversation data, the data acquisition unit is configured to be able to acquire text conversation data from a transcription unit, the transcription unit is configured to generate the text conversation data in which the content of the conversation is transcribed using the voice conversation data, and the medical text generation unit is configured to generate the medical text based on the text conversation data and the pharmaceutical data. [3] Preferably, there is provided an information processing system as described in [2], wherein the medical sentence generation unit has an output generation unit, which generates a sentence generation prompt to be output to a large-scale language model, the sentence generation prompt having the text conversation data and the pharmaceutical data, and having instructions to generate the medical sentence, and the data acquisition unit acquires the medical sentence as output from the large-scale language model in response to the sentence generation prompt. [4] Preferably, there is provided an information processing system as described in [3], wherein the output generation unit generates an accuracy improvement prompt to be output to the large-scale language model, the accuracy improvement prompt has pre-processed character data and the pharmaceutical data, and has an instruction to generate processed character data by referring to the pharmaceutical data, the pre-processed character data is data obtained by transcribing the spoken conversation data, the processed character data is the text conversation data used by the medical text generation unit when generating the medical text, and the data acquisition unit acquires the processed character data as output from the large-scale language model in response to the accuracy improvement prompt. [5] Preferably, there is provided an information processing system as described in [4], wherein the accuracy improvement prompt has instructions for generating the processed character data by distinguishing between what was spoken by the medical professional and what was spoken by the patient, and the character conversation data distinguishes between what was spoken by the medical professional and what was spoken by the patient. [6] Preferably, there is provided an information processing system as described in [1], wherein the conversation data is text data and the content spoken by the medical professional and the content spoken by the patient are distinguished. [7] Preferably, there is provided an information processing system as described in any one of [1] to [6], wherein the medical data further includes basic data of the patient, the basic data including the patient's name and gender, and the medical text generation unit generates the medical text based on the conversation data, the pharmaceutical data and the basic data. [8] Preferably, there is provided an information processing system as described in any one of [1] to [7], wherein the medical data further includes medical history data of the patient, and the medical text generation unit generates the medical text based on the conversation data, the pharmaceutical data and the medical history data. [9] Preferably, there is provided an information processing system according to any one of [1] to [8], in which the medical text is generated by distinguishing between subjective information, objective information, evaluations, and plans.

[10] According to another aspect of an embodiment of the present invention, there is provided an information processing method for assisting in the creation of medical documents, the information processing method comprising an acquisition step and a generation step, in which in the acquisition step, conversation data and medical data are acquired, the conversation data including the content of a conversation between a medical professional and a patient, the medical data having pharmaceutical data, the pharmaceutical data having new pharmaceutical data and old pharmaceutical data, the new pharmaceutical data corresponding to the conversation of the conversation data, the old pharmaceutical data being data that precedes the new pharmaceutical data, and in the generation step, the medical document is generated based on the conversation data and the pharmaceutical data.

[11] According to another aspect of the embodiment of the present invention, there is provided a program for causing a computer to execute the information processing method described in

[10] . [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 shows an example of a system configuration of an information processing system 100 according to the embodiment. [Diagram 2] FIG. 2 is a block diagram showing a hardware configuration of the information processing device 1. As shown in FIG. [Diagram 3] FIG. 3 is a functional block diagram of the control unit 12 shown in FIG. [Figure 4] Figure 4A is an explanatory diagram of data output from the pharmacy to the information processing device 1, and Figure 4B is an explanatory diagram that diagrammatically shows an example of a type of storage data DT0 as a database stored in the memory unit 11 of the information processing device 1 shown in Figure 4A. [Diagram 5] FIG. 5 is an explanatory diagram of data exchanged between the information processing device 1, the large-scale language model server 2, and the speech-to-text server 3. As shown in FIG. [Figure 6]FIG. 6 is an explanatory diagram of data exchanged between the information processing device 1 and the large-scale language model server 2 after the data exchange shown in FIG. [Figure 7] Figure 7A shows an example of medical data with a date later than the data shown in Figure 7B in terms of time series, and Figure 7B shows an example of medical data with a date earlier than the data shown in Figure 7A in terms of time series. In Figures 7A and 7B, data shown in gray is non-output medical data (data that is not output to the large-scale language model server 2). [Figure 8] FIG. 8A shows an example of pre-processed character data d2, and FIG. 8B shows an example of processed character data d3. [Figure 9] FIG. 9 is an example of medical text MT1. [Figure 10] FIG. 10 is a flowchart showing an example of an information processing method of the information processing system 100 according to the embodiment. [Figure 11] FIG. 11A is an example of a system configuration of an information processing system 100 according to Modification 1, and FIG. 11B is a functional block diagram of a control unit 12 of the information processing device 1 shown in FIG. 11A. [Figure 12] FIG. 12 shows an example of a system configuration of an information processing system 100 according to the second modification. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Various characteristic features shown in the following embodiment can be combined with each other. Also, each characteristic feature can be an invention independently.

[0011] 1. System configuration of information processing system 100 The information processing system 100 can support the creation of medical documents that need to be created by medical professionals such as pharmacists. In other words, the information processing system 100 has a function of reducing the workload of medical professionals who create medical documents by generating medical documents.

[0012] As shown in Fig. 1, the information processing system 100 includes an information processing device 1, a large-scale language model server 2 (an example of a large-scale language model), a speech-to-text server 3, a terminal 4, and a speech acquisition device 5. These are connected to each other so as to enable the exchange of information via a communication network 6 (e.g., the Internet, etc.). Note that the communication network 6 may be a closed network, partly or entirely separated from the Internet. Each component of the information processing system 100, such as the information processing device 1, has one or more functions (functional units), and each component may be configured as a single device as shown in Fig. 1, or may be configured as multiple independent devices configured to be able to exchange information. The same applies to each functional unit, such as the control unit 12 of the information processing device 1, which will be described later. Each component included in the information processing system 100 will be further described below.

[0013] 1-1. Information processing device 1 2, the information processing device 1 has a communication unit 10, a storage unit 11, a control unit 12, an output unit 13, and an input unit 14, and these components are electrically connected inside the information processing device 1 via a communication bus 15. The control unit 12 has a data acquisition unit 121, a medical text generation unit 122, and an encoding unit 123. Furthermore, the medical text generation unit 122 has an output generation unit 122a and an adjustment unit 122b.

[0014] Each of the above components may be realized by software or by hardware. When realized by software, various functions can be realized by a CPU executing a computer program. The program may be stored in a non-transitory recording medium that can be read by a computer, or may be provided so as to be downloadable from an external server, or may be realized by so-called cloud computing, which reads out a program stored in an external storage unit and realizes the function. When realized by hardware, it can be realized by various circuits such as ASIC, FPGA, or DRP. In the embodiment, various information and concepts that include the information are handled, and these are represented by high and low signal values ​​or quantum bits as a binary bit collection consisting of 0 or 1, and communication and calculation can be performed by the above software or hardware aspects. The software may be a general-purpose OS or a dedicated OS.

[0015] The communication unit 10 may employ a wired communication means such as USB, IEEE1394, Thunderbolt (registered trademark), wired LAN network communication, etc. The communication unit 10 may employ a configuration in which it is connected to the communication network 6 via a wireless communication means such as wireless LAN network communication, mobile communication such as 3G / LTE / 5G, Bluetooth (registered trademark) communication, etc. The communication unit 10 may also employ a configuration in which the above-mentioned wired communication means and wireless communication means are used in combination.

[0016] The storage unit 11 stores, for example, various programs, constants, variables, and various values ​​such as setting values ​​of the information processing device 1 executed by the control unit 12. The storage unit 11 may be a storage device such as a solid state drive (SSD), or a random access memory (RAM) that stores temporarily required information (arguments, arrays, etc.) related to the calculation of a program. In addition to the storage unit 11, the information processing device 1 may also use an external storage unit (for example, an external storage medium, a cloud, etc.).

[0017] The control unit 12 is configured to execute processing and control related to information processing of the information processing device 1. The control unit 12 can be configured, for example, with a central processing unit (CPU), and in the embodiment, the control unit 12 is an example of a processor capable of executing programs related to each step of a flowchart described later. The control unit 12 realizes various functions related to the information processing device 1, for example, by reading out a program stored in the storage unit 11. Furthermore, the information processing of the software in the information processing device 1 is realized, for example, by the control unit 12 as hardware processing various programs stored in the storage unit 11.

[0018] The output unit 13 is, for example, a display unit of the information processing device 1. The output unit 13 may be, for example, included in the housing of the information processing device 1 or may be externally attached. The output unit 13 displays a screen of a graphical user interface (GUI) that can be operated by a user. For example, the output unit 13 may be a display device such as a CRT display, a liquid crystal display, an organic EL display, a plasma display, or an electronic paper display, as well as a display device such as a light that can be turned on or a projector. It is optional whether or not the information processing device 1 includes the output unit 13. For example, the output of the information processing device 1 may be displayed on a display unit that is located at a location independent of the location where the information processing device 1 is installed. The output unit 13 may also have a device that outputs audio.

[0019] The input unit 14 is configured to receive, for example, an operation input made by a user of the information processing device 1. The input unit 14 may be included in the housing of the information processing device 1, or may be attached externally. For example, a touch panel, a switch button, a mouse, a keyboard, etc. may be adopted as the input unit 14. It is optional whether or not the information processing device 1 includes the input unit 14. For example, the information processing device 1 may receive an operation input to the information processing device 1 via an information processing terminal located at a location independent of and separate from the location where the information processing device 1 is installed.

[0020] 1-2. Large-scale language model server 2 and speech-to-text server 3 The large scale language model server 2 and the speech-to-text server 3 are communicatively connected to the information processing device 1 via a communication network 6 . The large-scale language model server 2 has a sentence generation unit 21. This sentence generation unit 21 has, for example, a neural network trained using a large amount of teacher data, and has a function of generating a response to data (prompt) output from the information processing device 1 and outputting it to the information processing device 1. The large-scale language model server 2 is not particularly limited, but for example, ChatGPT by OpenAI Inc. can be adopted.

[0021] The voice-to-text server 3 includes a transcription unit 31. The transcription unit 31 is configured to generate text conversation data in which the contents of the conversation are transcribed using voice conversation data. The transcription unit 31 includes, for example, a neural network trained using a large amount of teacher data (voice teacher data). The transcription unit 31 has a function of receiving data (voice data) output from the information processing device 1 and outputting the data as text data to the information processing device 1. Note that, in the embodiment, the transcription unit 31 has been described as being configured as a trained neural network, but is not limited thereto. For example, the transcription unit 31 may be configured to analyze and transcribing voice data according to a predetermined algorithm. The speech-to-text server 3 is not particularly limited, but may be, for example, Whisper (registered trademark) by OpenAI Corporation.

[0022] 1-3. Terminal 4 and voice capture device 5 The terminal 4 is an information processing device (e.g., a personal computer) placed in, for example, a pharmacy, and is configured to allow a medical professional to input patient data. Also, the terminal 4 is configured to display the patient data on a display unit (not shown) so that the medical professional can obtain the patient data using the terminal 4. The patient data exchanged in the information processing system 100 will be described in detail later. The voice acquisition device 5 has a device such as a microphone, and is configured to acquire conversation data between a medical professional and a patient at a pharmacy. The conversation data is voice conversation data. The voice acquisition device 5 may be integrated with the terminal 4 or may be separate. For example, the voice acquisition device 5 may be a dedicated voice recorder or a mobile phone (e.g., a smartphone) equipped with a voice acquisition application.

[0023] In one example of the embodiment, when the voice acquisition device 5 acquires conversation data (voice data), the voice data is transmitted from the voice acquisition device 5 to the terminal 4. Then, the terminal 4 outputs the patient data and the conversation data together to the information processing device 1. This prevents the patient data related to the current prescription and the conversation data related to the current prescription from being separated, and allows the information processing device 1 to process them appropriately. Note that the configuration of data transmission is not limited to this, and when the voice acquisition device 5 acquires the conversation data, it may be configured so that the voice acquisition device 5 outputs the conversation data to the information processing device 1 via the communication network 6, and the information is compiled on the information processing device 1 side.

[0024] 2. Functional configuration The functional configuration of the information processing device 1 according to this embodiment will be described with reference to Fig. 3. Information processing by software stored in the storage unit 11 is specifically realized by the control unit 12, which is an example of hardware, and each functional unit included in the control unit 12 is executed.

[0025] The data acquisition unit 121 is configured to acquire various data from the information processing device 1, the large-scale language model server 2, the speech-to-text server 3, etc. In the embodiment, the data acquisition unit 121 is configured to acquire conversation data and medical data. The data acquisition unit 121 is also configured to acquire text conversation data from the transcription unit 31. These data will be described in detail later.

[0026] The medical text generation unit 122 is configured to generate medical text based on conversation data (text conversation data) and drug data. The output generation section 122a of the medical sentence generation section 122 has a function of generating various prompts (for example, sentence generation prompts and accuracy improvement prompts, which will be described later). The adjustment unit 122b of the medical text generation unit 122 has a post-processing function. That is, the adjustment unit 122b has a function of adjusting the content of the medical text MT1 by adding information to the medical text MT1 generated by the large-scale language model server 2 (for example, adding information such as a name that can identify an individual to the medical text MT1). Note that it is optional whether the information processing device 1 has the adjustment unit 122b or not.

[0027] The encoding unit 123 has a function of encoding a part or all of the non-output medical data among the data related to the patient. In other words, the encoding unit 123 has a function of encoding the non-output medical data so that the contents of the non-output medical data are anonymized or abstracted. Note that the non-output medical data is data among the data of the patient that is not output to the large-scale language model server 2. The information processing device 1 prevents individuals from being identified by outputting the encoded data. In other words, since data relating to patients includes, for example, data that is personal information, in the embodiment, data that is not output to the large-scale language model server 2 (non-output medical data) is predetermined, but the information processing device 1 is capable of encoding the non-output medical data and including it in the output medical data.

[0028] 3. Data exchanged in the information processing system 100 In the information processing system 100 according to the embodiment, various data are exchanged between the information processing device 1, the large-scale language model server 2, the speech-to-text server 3, and the terminal 4. The data will be described with reference to FIGS. 4A to 8B.

[0029] The data exchanged in the information processing system 100 includes medical data, conversation data, and medical documents. In one embodiment, the types of medical data include items shown in data D1 to data D19 in Fig. 4B. The numbers of data D1 to data D19 are given for the sake of convenience so that these items can be distinguished. Also, not all of data D1 to data D19 are necessarily required items.

[0030] 3-1.Medical Data In the information processing system 100 according to the embodiment, the storage unit 11 has a function as a database, and therefore all data exchanged in the information processing system 100 is accumulated in the storage unit 11. First, types of medical data among this all data will be listed.

[0031] As shown in FIG. 4B, in an embodiment, the medical data includes items such as ID (data D1), name (data D2), gender (data D3), date of birth (data D4), age (data D5), generation (data D5t), health insurance card information (data D6), allergy history (data D7), history of current illness (data D8), medical history (data D9), history of side effects from medication (data D10), history of concomitant use of medications (data D11), information related to problems with taking the medication (data D12), pharmacy name (data D13), name of prescribing physician (data D14), hospital name (data D15), prescription date (data D16), drug name (data D17), drug quantity, usage, and dosage (data D18), and medical history (data D19).

[0032] The ID (data D1) is a unique symbol assigned to each individual, and is represented by, for example, numbers or letters of the alphabet. The generation (data D5t) may be acquired from the terminal 4 on the pharmacy side, but the information processing device 1 may have a function of calculating the generation based on the age, and may acquire the generation by executing this function. This function can be realized by the encoding unit 123, for example. The insurance card information (data D6) includes an insurer number, a symbol, and a number. Information relating to problems with taking a medicine (data D12) includes, for example, information about circumstances other than allergies, such as an individual's inability to tolerate a medicine or an inability to tolerate a medicine (for example, a desire to avoid Chinese herbal medicine because it does not agree with the taste).

[0033] The drug name (data D17) is an example of drug data. The drug name may be the exact name of the drug, or an abbreviated name. Name (data D2), gender (data D3), date of birth (data D4), age (data D5), and health insurance card information (data D6) are examples of basic data of a patient. The medication history (data D19) is an example of medication history data.

[0034] <Stored data DT0 and medical data DT1> In one example of the embodiment, all of these 20 items of data are stored as a database in the storage unit 11 of the information processing device 1. Then, when a patient visits a pharmacy and a medical professional receives, for example, a prescription, the medical professional inputs the prescription data into the terminal 4. As a result, the prescription data (medical data DT1) is output to the information processing device 1, as shown in FIG. 4A. For example, if the medical data DT1 is prescription data, it includes at least data such as the pharmacy name (data D13), the drug name (data D17), and the amount, usage, and dosage of the drug (data D18).

[0035] Furthermore, for example, if the medical data DT1 is prescription data, it includes data such as the name of the medicine (data D17) and the amount, usage, and dosage of the medicine (data D18). Furthermore, for example, if the medical data DT1 is pharmaceutical data, data such as the name of the pharmaceutical (data D17) is included. The types of data included in the medical data DT1 are not limited to prescription data, dispensing data, and drug data, and can be changed in various ways. In other words, the medical data DT1 can include all or some of the above-mentioned 20 items of data, and various settings are possible. The information processing system 100 is preferably configured to be able to change the type of data included in the medical data.

[0036] When the medical data DT1 is output to the information processing device 1, the contents of the stored data DT0 as a database are updated each time. Also, for example, when there is a change in the patient's status (e.g., name), data related to the change is input to the terminal 4, and the stored data DT0 is updated. Note that the storage unit 11 also holds data before the update (old data). In this way, the storage unit 11 collectively stores data exchanged in the information processing system 100, and the data accumulates.

[0037] Incidentally, medical data can be divided into the above-mentioned non-output medical data and output medical data. The non-output data is data that is not output to the large-scale language model server 2 because it contains personal information, and the output data is data that is output to the large-scale language model server 2. These types of data will be explained below.

[0038] 3-1-1.Non-printable medical data The non-output medical data will be described with reference to Fig. 7A. In the embodiment, the non-output medical data may include, for example, name (data D2), date of birth (data D4), age (data D5), health insurance card information (data D6), pharmacy name (data D13), prescribing doctor name (data D14), hospital name (data D15), and prescription date (data D16). The data is not limited to those exemplified here, and for example, information that is considered to be relatively more anonymous than the name, etc. (date of birth and age) may be included in the output medical data, and may be changed as appropriate.

[0039] 3-1-2. Output medical data The output medical data will be described with reference to FIG. 7A. In the embodiment, the output medical data may include, for example, ID (data D1), sex (data D3), age (data D5t), allergy history (data D7), current illness history (data D8), medical history (data D9), side effects of medicine (data D10), history of concomitant use of medicine (data D11), information on problems with taking medicine (data D12), name of medicine (data D17), quantity, usage, and dosage of medicine (data D18), and medical history (data D19). The data is not limited to the data exemplified here, and other information may be added or may be reduced as necessary.

[0040] <Medical Data DT2> Here, the medical data DT2 shown in FIG. 5 and the medical data DT3 shown in FIG. 6 will be described. 5 shows how medical data DT2 is output to a large-scale language model server 2 in order to improve the accuracy (precision) of text in raw character data d2, which will be described later. The medical data DT2 may include all or a part of the output medical data described above. The medical data DT2 includes drug names (data D17). This is because drug names are often long and not commonly used, making them difficult to transcribe appropriately. Preferably, the medical data DT2 includes at least one of the amount, dosage, and administration of the medicine (data D18) and the medical history (data D19). In addition, the medical data DT2 may include not only the contents of the current prescription (see FIG. 7A) but also the contents of past prescriptions (see FIG. 7B). The contents of the past prescriptions may include those of multiple times (for example, the last time and the time before that).

[0041] <Medical Data DT3> 6 shows how medical data DT3 is output to the large-scale language model server 2 together with processed character data d3, which will be described later, in order to generate medical text MT1, which will be described later. The medical data DT3 may include all or a part of the output medical data described above. The medical data DT3 includes a drug name (data D17), which is an example of drug data, in order to appropriately generate the medical text MT1 output from the large-scale language model server 2. Preferably, the medical data DT3 includes at least one of the amount, dosage, and administration of the medicine (data D18) and the medical history (data D19). More preferably, the medical data DT3 includes at least one of ID (data D1), gender (data D3), age (data D5t), allergy history (data D7), current illness history (data D8), medical history (data D9), history of drug side effects (data D10), and history of concomitant drug use (data D11).

[0042] In addition, the medical data DT3 may include not only the contents related to the current prescription (see FIG. 7A) but also the contents related to past prescriptions (see FIG. 7B) in the same manner as the medical data DT3. The contents related to the past prescriptions may include those for multiple times (for example, the last time and the time before that). Specifically, a drug name (an example of drug data) may have new drug data and old drug data. The new drug data corresponds to the conversation in the conversation data. In other words, the new drug data is data on the drug name written on, for example, a prescription submitted by the patient this time. The old drug data is data that precedes the new drug data. In other words, the old drug data is data on a drug written on, for example, a prescription submitted by the patient before this time. "Before this time" may be data relating to a prescription one prescription before the prescription in the new drug data, or data relating to a prescription two prescriptions before, or even earlier than this time. The old drug data may also have multiple pieces of data that precede this time.

[0043] 3-2. Conversation data The conversation data includes the contents of the conversation between the medical staff and the patient. The types of conversation data include voice conversation data d1 (voice data acquired by the voice acquisition device 5 in the embodiment) shown in FIG. 4A and FIG. 5, and text conversation data. The types of text conversation data include pre-processed text data d2 shown in FIG. 5, and processed text data d3 shown in FIG. 5 and FIG. 6. Here, the pre-processed text data d2 is data obtained by transcribing the voice conversation data by the transcription unit 31. The processed text data d3 is text conversation data used by the medical text generation unit 122 when generating medical text. In the embodiment, the processed text data d3 is text data obtained after the pre-processed text data d2 is processed by the text generation unit 21 so as to improve the accuracy of the contents.

[0044] Specifically, the pre-processed character data d2 may contain errors in the name of a medicine, etc., as shown in the example of FIG. 8A (see the underlined portion). However, in the processed character data d3, as shown in the example of FIG. 8B (see the underlined portion), the errors have been corrected, improving the accuracy of the text. In addition, the processed character data d3 distinguishes between what was said by the medical professional and what was said by the patient. Since such processed character data d3 is output to the large-scale language model server 2, the output from the large-scale language model server 2 (medical text MT1) is optimized and the accuracy is improved.

[0045] 3-3. Medical documents In the embodiment, the medical text is divided into medical text MT1 output from the large-scale language model server 2 to the information processing device 1 and medical text MT2 output from the information processing device 1 to the pharmacy. The medical text MT1 has a predetermined format as shown in Fig. 9. An example of this predetermined format is SOAP. Here, SOAP refers to a recording method (recording format) used in medical care and other fields, where subjective information (Subject), objective information (Object), assessment, and plan are differentiated. In other words, the medical text MT1 is generated by distinguishing between subjective information, objective information, assessment, and plan.

[0046] The medical text MT2 is data configured to identify an individual by adding at least one of the following data to the medical text MT1: name (data D2), gender (data D3), date of birth (data D4), age (data D5), and health insurance card information (data D6). In other words, the medical text MT2 is data to which elements of basic data have also been added. The medical text MT2 may not be generated by the information processing device 1, and the medical text MT1 itself may be output to the pharmacy. In this case, for example, a process to identify the individual may be performed by the terminal 4 on the pharmacy side.

[0047] 4. Information Processing Flow of Information Processing System 100 The program according to the embodiment executes each step (information processing method) described below. This information processing method is a process for supporting the creation of medical documents. An example of the information processing method according to the embodiment will be described with reference to FIG.

[0048] <Start: Step S001> The medical worker receives a prescription from the patient and starts dispensing according to the prescription. Once dispensing is complete, a conversation with the patient begins to hand over the medicine to the patient, but prior to this conversation, the medical worker has started up the voice capture device 5. Medical workers at pharmacies are required to leave records of medical procedures such as dispensing and conversations as medical documents, which is time-consuming. The information processing system 100 according to the embodiment can reduce this time-consuming process by automatically generating all or part of the medical documents, thereby reducing the burden on medical workers.

[0049] <Step S002> When the medical worker completes the conversation with the patient, he stops the voice capture device 5 and transmits voice conversation data d1 from the memory of the voice capture device 5 to the memory of the terminal 4, and the terminal 4 thereby acquires the medical data DT1. The medical worker also inputs the contents of the current prescription as the medical data DT1 into the terminal 4, and the memory of the terminal 4 acquires this medical data DT1. Note that the medical data DT1 may also be acquired by scanning an image of the prescription and reading information such as the characters on the prescription from the scanned data.

[0050] In the description here, the patient has already used one of the pharmacies utilized by the information processing system 100, and the patient's basic data (such as name (data D2)) has already been stored in the memory unit 11 of the information processing device 1 as a database. Therefore, in the example of the medical data DT1 in step S001, it is not necessary to input all of the basic data (such as age (data D5)).

[0051] <Step S003> The data acquisition unit 121 of the information processing device 1 acquires the voice conversation data d1 and the medical data DT1 transmitted from the pharmacy side, and stores them in the storage unit 11.

[0052] This step S003 is an example of an acquisition step.

[0053] <Step S004> The control unit 12 of the information processing device 1 outputs the voice conversation data d1 stored in the memory unit 11 to the voice-to-text server 3. The transcription unit 31 of the voice-to-text server 3 transcribes the voice conversation data d1 to generate raw character data d2, which is character data.

[0054] <Step S005> The speech-to-text server 3 outputs the raw character data d2 to the information processing device 1. The data acquisition unit 121 of the information processing device 1 acquires the raw character data d2 and stores it in the storage unit 11.

[0055] <Step S006> The control unit 12 of the information processing device 1 outputs the pre-processing character data d2 and the medical data DT2 stored in the storage unit 11 to the large-scale language model server 2, and generates an output prompt for this purpose. Specifically, the output generation unit 122a of the medical text generation unit 122 generates an accuracy improvement prompt to be output to the large-scale language model server 2. The accuracy improvement prompt has the pre-processing character data d2 and the medical data DT2, and has an instruction to generate processed character data d3 by referring to the medical data DT2. In addition, the accuracy improvement prompt has an instruction to generate processed character data d3 by distinguishing between what the medical professional said and what the patient said (see FIG. 8B), which is expected to have the effect of improving the generation accuracy of the medical text MT1.

[0056] The control unit 12 of the information processing device 1 outputs this accuracy improvement prompt to the large-scale language model server 2. The number of characters of the accuracy improvement prompt is specifically, for example, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 1200, 1400, 1600, 1800, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000, 12000, 14000, and may be within a range between any two of the numerical values ​​exemplified here. The accuracy improvement prompt may also be composed of multiple divided prompts.

[0057] The sentence generation unit 21 of the large-scale language model server 2 refers to the accuracy improvement prompt, generates processed character data d3 by correcting the contents of the raw character data d2 so as to improve the accuracy of the raw character data d2, and outputs the processed character data d3 to the information processing device 1.

[0058] <Step S007> The data acquisition unit 121 of the information processing device 1 acquires the processed character data d3 as an output from the large-scale language model in response to the accuracy improvement prompt, and stores it in the storage unit 11.

[0059] <Step S008> Subsequently, the control unit 12 of the information processing device 1 outputs the processed character data d3 and the medical data DT3 stored in the storage unit 11 to the large-scale language model server 2, and generates an output prompt for this purpose. Specifically, the output generation unit 122a of the medical text generation unit 122 generates a text generation prompt to be output to the large-scale language model server 2. The text generation prompt has the processed character data d3 and the medical data DT3, and has an instruction to generate medical text MT1. In the embodiment, the prompt has an instruction to create the contents of the processed character data d3 and the medical data DT3 in a predetermined format (SOAP in the embodiment). The prompt may also include supplemental data, such as a definition of what the SOAP is, or a limit on the number of characters in the medical text MT1 to be generated. It should be noted that the medical data DT3 in this prompt has information that could identify an individual (for example, name, which is an example of basic data) removed, but the information may be included as coded data (abstracted data) that has been coded (abstracted) by the coding unit 123.

[0060] The control unit 12 of the information processing device 1 outputs this sentence generation prompt to the large-scale language model server 2. The number of characters in the sentence generation prompt is specifically, for example, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 1200, 1400, 1600, 1800, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000, 12000, 14000, and may be within a range between any two of the numerical values ​​exemplified here. The sentence generation prompt may also be composed of multiple divided prompts.

[0061] The sentence generation unit 21 of the large-scale language model server 2 refers to the sentence generation prompt, generates the medical sentence MT1, and outputs it to the information processing device 1.

[0062] <Step S009> The data acquisition unit 121 of the information processing device 1 acquires medical text MT1 as an output from the large-scale language model in response to a text generation prompt, and stores it in the memory unit 11. Here, the medical data DT3 output to the large-scale language model server 2 is processed so as not to include personal information. For this reason, the adjustment unit 122b of the medical text generation unit 122 generates medical text MT2 by adding basic data (such as name) to the medical text MT1 so that an individual can be identified, and stores it in the memory unit 11. In other words, the medical text generation unit 122 generates medical text MT2 based on the conversation data (processed character data), pharmaceutical data (medical data), and the basic data.

[0063] Steps S008 and S009 are an example of a generating step.

[0064] <Step S010> The medical staff acquires the medical text MT2 via the terminal 4, and the medical text MT2 is stored in the memory of the terminal 4. If the information processing system 100 does not have the adjustment unit 122b, the terminal 4 acquires the medical text MT1.

[0065] <End: Step S011> By using Medical Text MT2, medical professionals can reduce the effort required to create medical documents.

[0066] 5. Variations 5-1. Variation 1 In the embodiment, a configuration has been described in which medical text is generated using servers external to the information processing device 1, such as the large-scale language model server 2 and the voice-to-text server 3, but the present invention is not limited to this. As shown in Fig. 11A and Fig. 11B, the information processing device 1 may have the functions of the large-scale language model server 2 and the voice-to-text server 3. In other words, the information processing device 1 may have a text generation unit 21 and a transcription unit 31.

[0067] 5-2. Variation 2 In the first modification, the information processing system 100 is described as a system in which information is exchanged between a medical institution (pharmacy) and an information processing device 1 located in a place independent of the medical institution, but the present invention is not limited to this. As shown in Fig. 12, the information processing device 1 may be located in the medical institution (pharmacy), and data may be stored within the medical institution (pharmacy).

[0068] 5-3.Other modifications In the embodiment, the medical institution is described as a pharmacy, but the present invention is not limited to this. For example, the present invention can be applied to a place that provides medicines installed in a hospital. The present invention can also be applied to fields where documents in a predetermined format must be created. For example, the present invention can be applied to work in a hospital, since SOAPs are created not only in places where medicines are prescribed, but also in hospitals.

[0069] The text generation unit 21 may adopt a configuration having a dedicated neural network for generating medical text. Dedicated means specialized in a specific field, for example, in the field of medicine and medical care when applied to a pharmacy as in the embodiment. For example, the text generation unit 21 can be configured to perform calculations based on a predetermined learning model. Here, the learning model is a model that trains a model using a large amount of teacher data and makes it possible to predict future output. For example, a large amount of medical data (medicine names, etc.) can be used as input data, and a large amount of medical text can be used as output data. This allows the text generation unit 21 to generate medical text based on a learning model that inputs medical data and outputs medical text.

[0070] In the embodiment, the information processing system 100 has the transcription unit 31, but the present invention is not limited to this. In other words, the transcription unit 31 may exist outside the information processing system 100, and the information processing system 100 may adopt conversation data that has already been transcribed externally. In this case, the information processing device 1 acquires conversation data that has been transcribed externally, and includes the acquired conversation data in the medical data DT3, thereby generating a sentence generation prompt.

[0071] Although the embodiments have been described above, they are presented as examples and are not intended to limit the scope of the invention. The novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made. The embodiments and their modifications are included in the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0072] 100: Information processing systems 1: Information processing equipment 10: Communications Department 11: Storage section 12: Control section 121: Data acquisition section 122: Medical text generation department 122a: Output generation section 122b: Adjustment section 123: Encoding section 13: Output section 14: Input section 15: Communication bus 2: Large-scale language model server 21: Sentence generation section 3: Speech to text server 31: Transcription Department 4: Terminal 5: Audio capture device 6: Communication network DT0: Stored data DT1: Medical Data DT2: Medical Data DT3: Medical Data MT1: Medical Texts MT2: Medical Texts d1: Voice conversation data d2: Unprocessed character data d3: Processed character data

Claims

1. An information processing system for supporting the creation of medical documents, comprising: A data acquisition unit and a medical text generation unit are provided, The data acquisition unit is configured to be able to acquire character conversation data and medical data, The text conversation data includes the content of a conversation between a medical professional and a patient, the medical data comprises pharmaceutical data; The pharmaceutical data includes new pharmaceutical data and old pharmaceutical data, the new drug data corresponds to the conversation in the text conversation data; The old drug data is data prior to the new drug data, The medical text generation unit is configured to generate the medical text by inputting the text conversation data and the pharmaceutical data into a learning model and outputting the medical text from the learning model.

2. 2. The information processing system according to claim 1, The medical text generation unit has an output generation unit, The output generation unit generates sentence generation prompts to be output to the large-scale language model as the learning model; The sentence generation prompt includes the text conversation data and the medicine data, and includes instructions for generating the medical sentence; The medical sentence generation unit is configured to generate the medical sentence by inputting the sentence generation prompt into the large-scale language model and outputting the medical sentence from the large-scale language model.

3. The information processing system according to claim 1, Further comprising a transcription unit, The transcription unit is configured to convert voice conversation data corresponding to the content of the conversation into text data to generate the text conversation data; An information processing system, wherein the data acquisition unit acquires the text conversation data from the transcription unit.

4. 4. The information processing system according to claim 3, The medical text generation unit has an output generation unit, The output generation unit generates precision improvement prompts to be output to the large-scale language model as the learning model; the accuracy improvement prompt includes pre-processed character data and the pharmaceutical data, and includes an instruction to generate processed character data by referring to the pharmaceutical data, and an instruction to improve accuracy of the generation of the medical text; The unprocessed character data is the text conversation data acquired from the transcription unit, the processed text data is the text conversation data output from the large-scale language model based on the refinement prompts; The medical text generation unit is configured to generate the processed character data by inputting the accuracy improvement prompt to the large-scale language model and outputting the processed character data from the large-scale language model.

5. An information processing system according to claim 4, the output generator generates sentence generation prompts for output to the large-scale language model; the sentence generation prompt includes the text conversation data and the medicine data of the processed text data, and includes an instruction to generate the medical sentence; The medical sentence generation unit is configured to generate the medical sentence by inputting the sentence generation prompt into the large-scale language model and outputting the medical sentence from the large-scale language model.

6. 5. The information processing system according to claim 4, The instruction for improving the accuracy of the generation of the medical text in the accuracy improvement prompt includes an instruction for generating the processed character data by distinguishing between the content spoken by the medical professional and the content spoken by the patient; An information processing system, wherein the text conversation data of the processed text data is differentiated between what was spoken by the medical staff and what was spoken by the patient.

7. 2. The information processing system according to claim 1, An information processing system, wherein the text conversation data distinguishes between what was spoken by the medical professional and what was spoken by the patient.

8. An information processing system according to any one of claims 1 to 7, The medical data further comprises basic data of the patient; The basic data comprises the patient's name and the patient's sex, The medical text generation unit further generates the medical text based on the basic data.

9. An information processing system according to any one of claims 1 to 7, The medical data further comprises medical history data of the patient; The medical text generation unit generates the medical text by inputting the text conversation data, the pharmaceutical data, and the medical history data into the learning model and outputting the medical text from the learning model.

10. An information processing system according to any one of claims 1 to 7, An information processing system in which the medical documents are generated by distinguishing between subjective information, objective information, evaluations, and plans.

11. An information processing method for supporting the creation of medical documents, comprising: The method includes an acquisition step and a generation step, In the acquiring step, character conversation data and medical data are acquired, The text conversation data includes the content of a conversation between a medical professional and a patient, the medical data comprises pharmaceutical data; The pharmaceutical data includes new pharmaceutical data and old pharmaceutical data, the new drug data corresponds to the conversation in the text conversation data; The old drug data is data prior to the new drug data, An information processing method, in which the generating step generates the medical text by inputting the text conversation data and the pharmaceutical data into a learning model and outputting the medical text from the learning model.

12. A program causing a computer to execute the information processing method according to claim 11.

Images