Medication history generation support system and medication history generation support program

Abstract

To generate appropriate candidates for a medication history to perform appropriate medication counseling.SOLUTION: A medication history generation support system comprises: a medication data storage unit 32 that stores prescription data and medication history data in association with each other as medication data for every patient and every clinical department; an extraction unit 34 that, on the basis of newly obtained patient's prescription data, extracts, from the medication data, prescription data in the same clinical department as the newly obtained patient's prescription data and medication history data in the same clinical department as the newly obtained patient; a conversation text generation unit 13 that generates text data of a conversation with the patient on the basis of voice data of the conversation; and a medication history data generation unit 16 that causes a generative AI to generate medication history data corresponding to the newly obtained patient's prescription data, on the basis of the extracted prescription data and medication history data, the generated text data, and prompt text data for generating the medication history data.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a medication history creation support system and a medication history creation support program that generate appropriate medication history candidates in order to provide appropriate medication instruction. [Background technology]

[0002] Patients who have received medical treatment at a hospital or clinic bring the prescription issued by the hospital or clinic to a pharmacy, receive medication based on the prescription, and receive instructions on dosage and administration from the pharmacist.

[0003] At this time, there are a wide variety of side effects and precautions for drugs, and new drugs are constantly being released to the market, making it difficult for inexperienced pharmacists or those unfamiliar with new drugs to provide appropriate medication instructions.

[0004] When it comes to lifestyle-related diseases where the medication does not change, inexperienced pharmacists may end up giving the same medication instructions every time.

[0005] Patent Document 1 discloses a technology related to an inference system in which an AI server, a learning database server, and a computer are connected via a network. The AI ​​server includes a processor and a storage device, and infers at least one of a prescription drug's medication instruction and its importance. The processor includes a machine learning unit that generates a trained model by machine learning based on at least one of first training data in which prescription data and medication instruction data are associated with each other and second training data in which prescription data and prescription drug importance data are associated with each other, and an inference unit that infers at least one of the medication instruction and its importance from the inference prescription data input to the trained model. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Patent Publication No. 2021-64126 Summary of the Invention [Problem to be solved by the invention]

[0007] When a patient brings a prescription issued by a hospital or clinic to a pharmacy to receive medication based on the prescription, the pharmacist will often speak with the patient to understand the patient's current condition, inquire about the patient's medication status, such as whether the patient is taking the medication appropriately, and then provide medication instructions. The patient's current condition and medication status are not recorded in prescription data or past medical history, and are often obtained through conversation when the patient brings the prescription to the pharmacy.

[0008] However, in the inference system described in Patent Document 1, a trained model is generated by machine learning based on at least one of first learning data in which prescription data and medication instruction data are associated and second learning data in which prescription data and prescription drug importance data are associated, and at least one of the medication instruction and importance is inferred from the prescription data for inference input into the trained model. Therefore, the patient's current condition and medication situation are not taken into consideration, making it difficult to provide appropriate medication instructions.

[0009] The present invention has been made in consideration of the above problems, and has an object to provide a medication history creation support system and a medication history creation support program that generate appropriate medication history candidates in order to provide appropriate medication instruction. [Means for solving the problem]

[0010] In order to achieve the above object, the first feature of the medication history generation support system according to the present invention is to a medication data storage unit in which prescription data and medication history data are associated with each other for each patient and each medical department and stored as medication data; An extraction unit that extracts, based on the newly acquired prescription data of a patient, prescription data of the same medical department as the newly acquired prescription data of the patient and medication history data of the same medical department as the newly acquired prescription data of the patient from the medication data; a conversation text generating unit that generates text data of the conversation based on voice data of the conversation with the patient; and a medication history data generation unit that causes a generation AI to generate medication history data corresponding to the newly acquired prescription data of the patient, based on the extracted prescription data and medication history data, the generated text data, and prompt text data for generating medication history data. Effect of the Invention

[0011] According to the medication history creation support system and medication history creation support program of the present invention, it is possible to create appropriate medication history candidates for providing appropriate medication instruction. [Brief description of the drawings]

[0012] [Figure 1] 1 is a schematic configuration diagram showing a schematic configuration of a medication history generation support system according to a first embodiment of the present invention. [Diagram 2] FIG. 2 is an explanatory diagram illustrating the processing contents at the time of medication instruction in the medication history creation support system according to the first embodiment of the present invention. [Diagram 3] FIG. 2 is an explanatory diagram illustrating the processing contents when a medication history is written in the medication history generation support system according to the first embodiment of the present invention. [Figure 4] FIG. 11 is an explanatory diagram illustrating the processing contents at the time of medication instruction in the medication history generation support system according to the second embodiment of the present invention. [Diagram 5] FIG. 11 is an explanatory diagram illustrating the processing contents when a medication history is written in the medication history generation support system according to the second embodiment of the present invention. [Figure 6] FIG. 11 is a schematic diagram showing the schematic configuration of a medication history generation support system according to a third embodiment of the present invention. [Figure 7]FIG. 11 is a schematic diagram showing the schematic configuration of a medication history generation support system according to a third embodiment of the present invention. [Figure 8] FIG. 11 is an explanatory diagram for explaining the operation of the medication history generation support system according to the fourth embodiment of the present invention. [Figure 9] FIG. 13 is a schematic diagram showing the schematic configuration of a medication history generation support system according to a fifth embodiment of the present invention. [Figure 10] FIG. 13 is an explanatory diagram illustrating the processing contents at the time of medication instruction in the medication history generation support system according to the fifth embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. The same or equivalent parts and components are designated by the same or equivalent reference numerals throughout the drawings. However, it should be noted that the drawings are schematic and may differ from the actual ones. In addition, the drawings may include parts with different dimensional relationships and ratios.

[0014] The embodiments described below are merely examples of devices for embodying the technical idea of ​​the present invention, and the technical idea of ​​the present invention does not limit the arrangement of each component to that described below. Various modifications can be made to the technical idea of ​​the present invention within the scope of the claims.

[0015] <First embodiment> A medication history creation support system according to a first embodiment of the present invention will be described below. FIG. 1 is a schematic diagram showing the schematic configuration of a medication history creation support system according to a first embodiment of the present invention.

[0016] As shown in Figure 1, the medication history generation support system 100 includes a medication history generation support server 1, front tablets 2A, 2B installed in a store such as a pharmacy, a dispensing server 3, and a generation AI system 4, each of which is connected via a network 5.

[0017] The front tablets 2A and 2B are mobile terminals and have a touch panel. An operator (e.g., a pharmacist) checks candidates for medication instruction text displayed on the front tablets 2A and 2B, and inputs prescription data from the touch panel. The front tablets 2A and 2B also record, for example, a conversation between a pharmacist and a patient, and generate audio data. When the medication history data is transmitted from the medication history generation support server 1 to the front tablets 2A and 2B, the operator (e.g., a pharmacist) checks the contents of the medication history data, edits it as necessary, and transmits the medication history data to the dispensing server 3. The front tablets 2A and 2B may be a notebook computer or a tablet terminal.

[0018] The dispensing server 3 stores data related to dispensing, and transmits the data to the medication history generation support server 1 via the network 5 upon request. The dispensing server 3 functionally comprises a transmission / reception unit 31, a medication data storage unit 32, a drug data storage unit 33, and an extraction unit 34.

[0019] The transmitting / receiving unit 31 receives requests from the medication history generation support server 1 via the network 5, and transmits extraction results by the extracting unit 34 to the medication history generation support server 1 via the network 5.

[0020] Prescription data and medication history data are associated with each other for each patient and each medical department, and are stored as medication data in the medication data storage unit 32. Specifically, patient data, prescription data, and medication history data are associated with each other, and are stored as medication data.

[0021] The patient data includes the patient's attribute information (attribute data) such as the patient ID, patient name, gender, age, current medical history, past medical history, favorite foods, allergies, and special notes.

[0022] The prescription data includes a prescription ID, a patient ID, a medical department, a prescription date and time, and prescription contents. For example, the prescription data includes information such as prescription ID: 0101, patient ID: 0011, medical department: internal medicine, prescription date and time: 2023.8.31, prescription contents: [general] Paroxetine tablets 10 mg, 1 tablet per dose (1 tablet per day), once a day after dinner, 7 days' supply [general] Triazolam tablets 0.25 mg, 1 tablet per dose (1 tablet per day), once a day before bedtime, 7 days' supply.

[0023] Medication history data includes medication history ID, patient ID, medical department, test item, test value, and SOAP instruction content. For example, medication history ID: 0201, patient ID: 0011, medical department: internal medicine, test item: CRP (0.2 mg / dL), SOAP instruction content: "S: Blood pressure was 178 / 105 mmHg at health check. Cholesterol was also high. O: 46 years old (not elderly), dyslipidemia. A: Grade II hypertension, risk tier 2, so high risk. Therefore, starting treatment with normal doses of antihypertensive drugs is appropriate. P: Next time, ask about obesity, history of organ damage / cerebrovascular disease, and family history of cerebrovascular disease. Monitor blood pressure and side effects."

[0024] Here, "SOAP" is a problem-solving method for managing medication history, in which information obtained by talking with patients is classified and organized into categories according to content, and then analyzed in four categories: Subjective Data, Objective Data, Assessment, and Plan.

[0025] S (Subjective Data) is information from the patient, what the patient said and their main complaint. O (Objective Data) is the patient's objective condition from the pharmacist's perspective, and includes the patient's physical condition, allergy history, side effect history, test results, etc. A (Assessment) is what the pharmacist thinks about from the pharmaceutical knowledge of the pharmacist based on the information in S and O, and the reason for deciding to implement the Plan. P (Plan) is the content of the instructions given by the pharmacist, i.e. what the pharmacist actually did and said.

[0026] Furthermore, when new medical history data is transmitted from the front tablets 2A, 2B, the medication data storage unit 32 associates it with the patient data and stores it as medication data.

[0027] The drug data storage unit 33 stores drug data by associating a drug ID for identifying a drug, a drug name, a contraindication condition, and a symptom. A contraindication condition indicates a symptom that should be avoided, such as information on heart disease, gastrointestinal ulcer, high blood pressure, thyroid disease, diabetes, insomnia, epilepsy, etc. A symptom is a symptom for which a drug is expected to be effective, such as cold, rhinitis, constipation, antipyretic analgesia, diarrhea, itching, lower back pain, oral disease, etc. More specifically, it includes medical drug data, data to check for contraindications for concomitant use and overlapping administration, data to check for applicable disease names, data to check patient attributes, and OTC (Over The Counter) drug data. Medical drug data includes drug ID, drug name, electronic prescription claim code, drug price listed name, furigana, generic name, drug price, administration route, Ministry of Health, Labour and Welfare code, pharmacological classification, poison and narcotic classification, pharmacopoeia classification, transitional measure date, discontinuation date, YJ code, dosage form, color, dosage form identification code, unit name, generic flag, usage limit, long-term administration limit, anti-tuberculosis drug classification, biological product classification, high-risk preparation, warning statement, package insert, photograph, etc. The data for checking for contraindications for combined use and overlapping administration is used to check for contraindications for combined use between prescription drugs, between prescription drugs and OTC drugs, and between OTC drugs, as well as overlapping ingredients and medicinal effects. The applicable disease name check data is a master data that systematizes searches for applicable disease names for drugs, contraindicated disease names, standard disease name master, ICD-10 compatible codes, etc. It is used to check whether a prescribed drug is applicable to a disease and whether it is being administered to a patient with a disease for which it should not be used. The patient attribute check data is used to check whether the drug is administered to pregnant women, lactating women, newborns, infants, children, elderly people, etc., based on the attributes listed in the "Contraindications" section of the package insert. In addition to basic information such as OTC drug names and package inserts, OTC (Over The Counter) drug data includes functions and data that can be used to select OTC drugs based on the patient's symptoms and attributes, as well as check boxes for "things not to do" and "things to ask about" when providing medication instructions.

[0028] Based on the newly acquired prescription data of the patient supplied from the medication history generation support server 1 via the network 5, the extraction unit 34 extracts the prescription data of the same medical department as the newly acquired prescription data of the patient, and the medication history data of the same medical department as the newly acquired prescription data of the patient, from the medication data stored in the medication data storage unit 32. The newly acquired prescription data of the patient also includes the prescription ID, the patient ID, the medical department, the prescription date and time, and the prescription contents.

[0029] The generation AI system 4 generates candidate instruction sentences corresponding to the newly acquired prescription data of the patient based on the prescription data and the medication history data, and the prompt text data for generating candidate instruction sentences, supplied from the medication history generation support server 1 via the network 5. Here, the prompt text data is a command sentence that issues an instruction to the generation AI.

[0030] In addition, the generation AI system 4 generates medical history data corresponding to the newly acquired prescription data of the patient based on the prescription data and medical history data, conversation text data, and prompt text data for generating medical history data supplied from the medical history generation support server 1 via the network 5.

[0031] The medication history generation support server 1 is composed of a computer, and generates appropriate medication history data candidates to provide appropriate medication instruction. The medication history generation support server 1 functionally comprises a transmission / reception unit 11, a voice data acquisition unit 12, a conversation text generation unit 13, a medication data acquisition unit 14, a prompt generation unit 15, and a medication history data generation unit 16.

[0032] The transmitter / receiver unit 11 receives voice data from the front tablets 2A, 2B via the network 5, and transmits prescription data, medical history data, conversation text data, prompt text data for generating medical history data, prompt text data for generating candidate instruction sentences, and the like to the generation AI system 4 via the network 5.

[0033] The voice data acquisition unit 12 acquires voice data from the front tablets 2A, 2B via the network 5. For example, this voice data is data in which a conversation between a patient and a pharmacist is recorded.

[0034] The conversation text generating unit 13 converts the voice data acquired by the voice data acquiring unit 12 into text, extracts data related to the current condition and medication status of the patient from the text, and generates the extracted data as text data (character data). Specifically, keywords related to the patient's condition and medication status may be stored in advance in an internal memory, and sentences including the keywords may be extracted as text data, or AI (artificial intelligence) may be used to learn keywords related to the patient's condition and medication status, and text data related to the current condition and medication status of the patient may be extracted from the conversation text data.

[0035] When the medication data acquisition unit 14 acquires new prescription data of a patient from the front tablets 2A, 2B via the network 5, it transmits the newly acquired prescription data of the patient to the dispensing server 3 via the network 5, and requests transmission of the prescription data of the same medical department as the newly acquired prescription data of the patient, and the newly acquired medication history data of the same medical department as the patient, from the medication data. Then, the medication data acquisition unit 14 receives the prescription data and the medication history data from the dispensing server 3.

[0036] The prompt generation unit 15 internally stores example prompt sentences for generating medication instruction sentences, and generates a prompt based on the example prompt sentences. For example, the example prompt sentence stored is "You are a pharmacist. When providing medication instruction to patient XX, generate multiple candidates for medication instruction sentences to be given to patient XX from the prescription data and medical history data of patient XX." In this case, the patient name stored in the prescription data and medical history data is inserted into patient XX to generate a prompt.

[0037] The prompt generating unit 15 also stores sample prompts for classifying into SOAP and generating medical history data, and generates a prompt based on the sample prompts. For example, a sample prompt is: "You are an experienced pharmacist. We have information on patient ●●'s previous visit, the prescription, and the past medical history (please check the date to determine whether it is past). A patient has come this time. Please write the current medical history in SOAP format based on the prescription. At that time, please pay attention to the difference between the prescription at the previous visit and the prescription at this visit, and consider why there is a difference in the prescription and write the SOAP. At that time, please also refer to the previous medical history. Note that [general] is an abbreviation for the generic name of a medical drug. If a high-risk drug is listed on the prescription, please pay attention to that when writing the SOAP.

[0038] Also, please write down any precautions that should be taken regarding the patient's medication in relation to their lifestyle. There may be hints in the conversation between the pharmacist and the patient, so please take that into consideration when writing the SOAP. Also, when answering, do not use polite language, but end your sentences in an assertive manner using "desu" or "da" as you would write in a paper. Other points to keep in mind when writing SOAP include the following: 1. Patient characteristics: Age, sex, height, weight, pregnancy status, allergy and side effect history, constitution, OTC drugs, health foods, luxury items, daily life, occupation, ADL, 2. Symptoms and medical history: Symptoms of the current disease, complications, visits to other hospitals or departments, other treatments or restrictions, etc. 3. Medication Adherence 4. Changes in symptoms Please enter all answers in the SOAP fields." In this case, the patient name stored in the prescription data and medication history data is entered in Patient XX, and a prompt is generated.

[0039] The medication history data generation unit 16 causes the generation AI system 4 to generate, via the network 5, candidate medication instruction sentences corresponding to the newly acquired prescription data for the patient, based on the prescription data and medication history data extracted by the extraction unit 34 and acquired by the medication data acquisition unit 14, and on prompt text data for generating candidate medication instruction sentences.

[0040] In addition, the medication history data generation unit 16 causes the generation AI system 4 to generate medication history data corresponding to the newly acquired prescription data of the patient via the network 5, based on the prescription data and medication history data extracted by the extraction unit 34 and acquired by the medication data acquisition unit 14, the conversation text data generated by the conversation text generation unit 13, and the prompt text data for generating medication history data. In addition, the medication history data generation unit 16 may cause the generation AI to generate candidate medication instruction sentences corresponding to the newly acquired prescription data of the patient based on the prescription data and medication history data extracted by the extraction unit 34 and acquired by the medication data acquisition unit 14, and prompt text data for generating candidate medication instruction sentences, and then cause the generation AI to generate medication history data corresponding to the newly acquired prescription data of the patient based on the extracted prescription data and medication history data, text data generated based on audio data of a conversation with the patient in line with the candidate medication instruction sentences, and the prompt text data for generating medication history data.

[0041] FIG. 2 is an explanatory diagram illustrating the processing contents during medication instruction in the medication history generation support system 100 according to the first embodiment of the present invention.

[0042] As shown in FIG. 2, the medication history generation support server 1 transmits the patient data D101, past prescription data D103, past medication history data D102, prompt text data D105 for generating candidates for medication instruction text, and current prescription data D104 received from the dispensing server 3 to the generation AI system 4 via the network 5. Here, the past prescription data is the prescription data of the same medical department as the prescription data of the current patient received from the dispensing server 3. The past medication history data is the medication history data of the same medical department as the current patient received from the dispensing server 3. Note that, although the medication history generation support server 1 transmits the patient data D101 here, the patient data D101 can be omitted.

[0043] The generation AI system 4 generates one or more medication instruction sentence candidates D201 corresponding to the current prescription data D104 based on the received patient data D101, past prescription data D103, past medication history data D102, prompt text data D105, and current prescription data D104, and transmits the medication instruction sentence candidates D201 to the medication history generation support server 1. The medication history generation support server 1 transmits the medication instruction sentence candidates D201 to one of the front tablets 2A, 2B that has transmitted the current prescription data D104.

[0044] This allows an operator (e.g., a pharmacist) who operates the front tablets 2A and 2B to select from the candidate medication instruction sentences D201 while checking the candidate medication instruction sentences D201 displayed on the front tablets 2A and 2B, and provide medication instruction to the patient, so that appropriate medication instruction can be provided without omissions or duplications. In addition, the conversation between the pharmacist and the patient during medication instruction can be recorded. The voice data recorded at this time is transmitted to the medication history generation support server 1 and converted into text data of the conversation.

[0045] FIG. 3 is an explanatory diagram illustrating the processing contents when the medication history generation support system 100 according to the first embodiment of the present invention writes a medication history.

[0046] As shown in FIG. 3, the medication history generation support server 1 transmits the patient data D101, past prescription data D103, past medication history data D102, prompt text data D106 for generating medication history data, current prescription data D104, and text data D107 of a conversation between a pharmacist and a patient, which are received from the dispensing server 3, to the generation AI system 4 via the network 5. Here, the past prescription data is the prescription data of the same medical department as the prescription data of the current patient received from the dispensing server 3. The past medication history data is the medication history data of the same medical department as the current patient received from the dispensing server 3.

[0047] The generation AI system 4 generates medical history data D202 corresponding to the current prescription data D104 based on the received patient data D101, past prescription data D103, past medical history data D102, prompt text data D106, current prescription data D104, and text data D107 of the conversation between the pharmacist and the patient, and transmits the medical history data D202 to the medical history generation support server 1. The medical history generation support server 1 transmits the medical history data D202 to one of the front tablets 2A and 2B that has transmitted the current prescription data D104.

[0048] When the medication history data is transmitted from the medication history generation support server 1 to the front tablets 2A, 2B, an operator (e.g., a pharmacist) checks the contents of the medication history data, edits it if necessary, and then transmits the medication history data to the dispensing server 3.

[0049] This allows an operator (e.g., a pharmacist) operating the front tablets 2A, 2B to register the medical history data by checking the medical history data D202 displayed on the front tablets 2A, 2B and editing it as necessary, thereby eliminating the need to generate medical history data from scratch and enabling appropriate medical history data to be registered.

[0050] <Second embodiment> The following describes the medication history generation support system according to the second embodiment of the present invention. The medication history generation support system 100 according to the second embodiment of the present invention includes a medication history generation support server 1, front tablets 2A and 2B, a dispensing server 3, and a generation AI system 4, which are connected to each other via a network 5, in the same manner as the second embodiment of the present invention. Therefore, the description of the configuration will be omitted.

[0051] FIG. 4 is an explanatory diagram illustrating the processing contents during medication instruction in the medication history generation support system 100 according to the second embodiment of the present invention.

[0052] As shown in Fig. 4, the medication history generation support server 1 generates possible risks as risk information D110 based on patient data D101, past medication history data D102, current prescription data D104, and medication data. For example, the extraction unit 34 extracts possible risks from the patient's attribute information and new prescription data, and refers to the medication data in the medication data storage unit of the dispensing server 3 to check the combination of the patient's illness and the prescribed medication, suggests possible side effects of the prescribed medication, and extracts risks such as interactions (contraindications for combined use) between OTC (Over The Counter) medications taken by the patient and the prescription.

[0053] The extraction unit 34 of the medication history generation support server 1 extracts prescription difference data between the past medication history data D102 and the current prescription data D104. For example, as the current / previous prescription difference data D111, "●● medication newly added for ● days, ●● medication added for number of days 14 days ⇒ 28 days, ●● medication deleted" is extracted.

[0054] Then, the medication history generation support server 1 transmits the risk information D110, past prescription data D103, prompt text data D105 for generating candidate medication instruction text, the current prescription data D104, and prescription difference data D111 to the generation AI system 4 via the network 5.

[0055] The generation AI system 4 generates one or more medication instruction sentence candidates corresponding to the current prescription data D104 as medication instruction sentence candidates D201 based on the risk information D110, past prescription data D103, prompt text data D105 for generating medication instruction sentence candidates, the current prescription data D104, and prescription difference data D111, and transmits the medication instruction sentence candidates D201 to the medication history generation support server 1. The medication history generation support server 1 transmits the medication instruction sentence candidates D201 to one of the front tablets 2A, 2B that has transmitted the current prescription data D104.

[0056] This makes it possible to obtain candidate medication instruction sentences D201 that take into consideration the risks to the patient, and an operator (e.g., a pharmacist) operating the front tablets 2A, 2B can select from the candidate medication instruction sentences D201 while checking the candidate medication instruction sentences D201 that take into consideration the risks displayed on the front tablets 2A, 2B, and provide medication instructions to the patient.

[0057] FIG. 5 is an explanatory diagram illustrating the processing contents when a medication history is written by the medication history generation support system 100 according to the second embodiment of the present invention.

[0058] As shown in FIG. 5, the medication history generation support server 1 generates risks considered from the patient data and prescription data as risk information D110 based on the patient data D101 and past medication history data D102 received from the dispensing server 3.

[0059] The extraction unit 34 of the medication history generation support server 1 extracts prescription difference data between the past medication history data D102 and the current prescription data D104.

[0060] Then, the medication history generation support server 1 transmits the risk information D110, past prescription data D103, prompt text data D106 for generating candidates for medication history data, the current prescription data D104, text data D107 of the conversation between the pharmacist and the patient, and prescription difference data D111 to the generation AI system 4 via the network 5.

[0061] The generation AI system 4 generates medical history data D202 corresponding to the current prescription data D104 based on the risk information D110, past prescription data D103, prompt text data D106 for generating medical history data, the current prescription data D104, text data D107 of the conversation between the pharmacist and the patient, and prescription difference data D111, and transmits the medical history data D202 to the medical history generation support server 1. The medical history generation support server 1 transmits the medical history data D202 to one of the front tablets 2A and 2B that has transmitted the current prescription data D104.

[0062] When the medication history data is transmitted from the medication history generation support server 1 to the front tablets 2A, 2B, an operator (e.g., a pharmacist) checks the contents of the medication history data, edits it if necessary, and then transmits the medication history data to the dispensing server 3.

[0063] This allows an operator (e.g., a pharmacist) operating the front tablets 2A, 2B to register medical history data by checking the medical history data D202 displayed on the front tablets 2A, 2B, which takes into account the patient's risks, and editing it as necessary, thereby eliminating the need to generate medical history data from scratch and enabling more appropriate medical history data to be registered.

[0064] In addition, prescription data for each patient and each medical department, medication history data for each patient and each medical department, and text data of conversations with patients may be associated with each other and used as training data to be learned by the generation AI system 4 in advance, and the generation AI may then be made to generate medication history data corresponding to the newly acquired prescription data for the patient based on prompt text data for generating medication history data.

[0065] In addition, the extraction unit 34 may extract prescription data from the same medical department as the newly acquired prescription data of the patient, and medication history data from the same medical department as the newly acquired patient, from the medication data based on the newly acquired prescription data of the patient, and may have the generation AI system 4 learn this extracted prescription data, medication history data, and text data of conversation with the patient as teacher data in advance, and then have the generation AI generate medication history data corresponding to the newly acquired prescription data of the patient based on prompt text data for generating medication history data.

[0066] In addition, the medication history generation support server 1 may be configured to transmit customer data D101, past medication history data D102, past prescription data D103, prompt text data D106 for generating medication history data candidates, current prescription data D104, and text data D107 of the conversation between the pharmacist and the patient to the generation AI system 4 via the network 5, or may be configured to transmit risk information D110, past prescription data D103, prompt text data D106 for generating medication history data, current prescription data D104, and text data D107 of the conversation between the pharmacist and the patient to the generation AI system 4 via the network 5.

[0067] <Third embodiment> A medication history creation support system according to the third embodiment of the present invention will now be described. FIG. 6 is a schematic diagram showing the schematic configuration of a medication history creation support system according to the third embodiment of the present invention.

[0068] 6, the medication history generation support system 100 includes a medication history generation support server 1A, front tablets 2A, 2B installed in a store such as a pharmacy, a dispensing server 3A, and a generation AI system 4, all of which are connected via a network 5. Of these components, the front tablets 2A, 2B and the generation AI system 4 have the same configuration as the front tablets 2A, 2B and the generation AI system 4 of the first embodiment of the present invention, and therefore a description thereof will be omitted.

[0069] The dispensing server 3A stores data related to dispensing, and transmits the data to the medication history generation support server 1 via the network 5 in response to a request. The dispensing server 3A functionally comprises a transmission / reception unit 31, a medication data storage unit 32, a drug data storage unit 33, an extraction unit 34, a text conversion unit 35, and a data update unit 36. Of these components, the transmission / reception unit 31, the medication data storage unit 32, the drug data storage unit 33, and the extraction unit 34 have the same configurations as the transmission / reception unit 31, the medication data storage unit 32, the drug data storage unit 33, and the extraction unit 34 of the first embodiment of the present invention, and therefore a description thereof will be omitted.

[0070] When the paper prescription is imaged by the imaging means provided in the front tablets 2A and 2B and the imaged image data is transmitted via the network 5, the text conversion unit 35 generates prescription text data by performing character recognition on the image data of the image of the prescription. At this time, the prescription text data may be generated by performing character recognition on the image data of the image by an OCR (Optical Character Recognition) function, or the generation AI system 4 may be caused to generate the prescription text data from the image data.

[0071] The data updating unit 36 ​​judges whether or not the test items included in the prescription text data generated by the text conversion unit 35 are appropriate, based on the test items included in the medication history data stored in the medication data storage unit 33. Note that as the medication history data stored in the medication data storage unit 33, all test items may be included in the medication history data itself, or a test item master that is a list of all test items related to the medication history data may be used. When the test items included in the prescription text data generated by the text conversion unit 35 are determined to be appropriate, the data update unit 36 ​​adds the test items and test values ​​included in the prescription text data to the medication history data in association with the patient, without modifying the test items included in the generated prescription text data.

[0072] Furthermore, when the text conversion unit 35 determines that the test items included in the generated prescription text data are not appropriate, the data update unit 36 ​​causes the generation AI system 4 to correct the test items included in the generated prescription text data, and adds the corrected test items and the test values ​​included in the generated prescription text data to the medication history data. In this case, the generation AI system 4 may be made to learn the drug data as training data in advance.

[0073] For example, if the test item included in the prescription text data is "CRP" and the test item included in the medication history data is "CRP", the test item names match, so the data update unit 36 ​​adds the test value of "CRP" to the medication history data. If the test item included in the prescription text data is "HBAIC" and the test item included in the medication history data is "HbA1c", "HBAIC" and "HbA1c" are similar, so the data update unit 36 ​​infers that "HBAIC" is "HbA1c" and adds the test value as "HbA1c" to the medication history data. If the test item included in the prescription text data is "ALT" and there is no test item included in the medication history data that matches or is similar to "ALT", the data update unit 36 ​​adds the test value to the medication history data together with the test item of "ALT".

[0074] In this way, since the dispensing server 3A is equipped with the text conversion unit 35 and the data update unit 36, even if the format in which test items and test values ​​are written is not standardized when a paper prescription is issued from a hospital, the user does not need to input the test items and test values ​​while looking at the paper prescription, thereby improving work efficiency.

[0075] The medication history generation support server 1A is composed of a computer, and generates candidates for appropriate medication history data to provide appropriate medication instruction. The medication history generation support server 1A functionally includes a transmission / reception unit 11, a voice data acquisition unit 12, a conversation text generation unit 13A, a medication data acquisition unit 14, a prompt generation unit 15, a medication history data generation unit 16, a summary generation unit 17, and a display control unit 18. Among these components, the transmission / reception unit 11, the voice data acquisition unit 12, the medication data acquisition unit 14, the prompt generation unit 15, and the medication history data generation unit 16 have the same configurations as the transmission / reception unit 11, the voice data acquisition unit 12, the medication data acquisition unit 14, the prompt generation unit 15, and the medication history data generation unit 16 of the first embodiment of the present invention, and therefore a description thereof will be omitted.

[0076] Like the conversation text generation unit 13, the conversation text generation unit 13A converts the voice data acquired by the voice data acquisition unit 12 into text, extracts data related to the patient's current condition and medication status from the text, and generates the extracted data as text data (character data). Furthermore, the conversation text generation unit 13A generates conversation text data based on the voice data of a conversation with a patient who has recently obtained a prescription, extracts information related to attribute data included in the patient data from the generated text data, and adds it to the attribute data.

[0077] For example, text data of the conversation between a patient and a pharmacist can be generated based on audio data of the conversation, and the generation AI system 4 can be made to extract keywords to be included in the attribute data from the generated text data and add the extracted keywords to the attribute data. Alternatively, a list of keywords related to keywords related to the attribute data can be stored in advance, and keywords stored in the keyword list can be extracted from the generated text data and the extracted keywords can be added to the attribute data.

[0078] In this way, since the conversation text generating unit 13A is provided, for example, a conversation between a pharmacist and a patient can be automatically added to the attribute data of the patient, thereby improving the work efficiency of the pharmacist.

[0079] The summary generation unit 17 generates a summary of the medication history data generated by the medication history data generation unit 16, or the medication history data generated by the medication history data generation unit 16, modified by a user such as a pharmacist, and stored in the medication data storage unit 32. For example, the summary generation unit 17 uses the generation AI system 4 to simplify SOAP text and generate it as a summary.

[0080] The display control unit 18 switches between displaying the medication history data generated by the medication history data generation unit 16, or the medication history data generated by the medication history data generation unit 16 and modified by a user such as a pharmacist and stored in the medication data storage unit 32, or the summary generated by the summary generation unit 17, and displays them on the screens of the front tablets 2A, 2B or on the screen of a monitor (not shown) connected to the medication history generation support server 1A.

[0081] For example, a concise display button and a detailed display button are displayed on the screen, and when the user presses the concise display button, the display control unit 18 displays a summary, and when the user presses the detailed display button, the display control unit 18 displays the original medication history data.

[0082] In this way, by having the summary generation unit 17 and the display control unit 18, the viewer can switch between viewing the medication history data and its summary depending on how detailed the viewer has viewed the medication history data, thereby providing high usability.

[0083] <Fourth embodiment> A medication history creation support system according to a fourth embodiment of the present invention will now be described. FIG. 7 is a schematic diagram showing the schematic configuration of a medication history creation support system according to the fourth embodiment of the present invention, and FIG. 8 is an explanatory diagram illustrating the operation of the medication history creation support system according to the fourth embodiment of the present invention.

[0084] 7, the medication history generation support system 100 includes a medication history generation support server 1B, front tablets 2A, 2B installed in a store such as a pharmacy, a dispensing server 3B, and a generation AI system 4, all of which are connected via a network 5. Of these components, the front tablets 2A, 2B and the generation AI system 4 have the same configuration as the front tablets 2A, 2B and the generation AI system 4 of the first embodiment of the present invention, and therefore a description thereof will be omitted.

[0085] The dispensing server 3B stores data related to dispensing, and transmits the data to the medication history generation support server 1 via the network 5 in response to a request. The dispensing server 3B functionally comprises a transmission / reception unit 31, a medication data storage unit 32, a drug data storage unit 33, an extraction unit 34B, and an index information storage unit 37. Of these components, the transmission / reception unit 31, the medication data storage unit 32, and the drug data storage unit 33 have the same configurations as the transmission / reception unit 31, the medication data storage unit 32, and the drug data storage unit 33 of the first embodiment of the present invention, and therefore a description thereof will be omitted.

[0086] The index information storage unit 37 stores keywords classified by patient condition as index information. The classifications can be, for example, "symptoms," "diseases," "allergies," "conditions specific to women," "matters related to drowsiness," "age," "product classification (JICFS classification)," and the like.

[0087] The extraction unit 34B generates prescription data for the patient based on recommended drug data generated by the recommended drug data generation unit 19 described later, and extracts, based on the generated prescription data, prescription data for the same medical department as the generated prescription data for the patient, and medication history data for the same medical department as the patient of the generated prescription data, from the medication data.

[0088] The medication history generation support server 1B is composed of a computer, and generates candidates for appropriate medication history data to provide appropriate medication instruction. The medication history generation support server 1B functionally includes a transmission / reception unit 11, a voice data acquisition unit 12, a conversation text generation unit 13, a medication data acquisition unit 14, a prompt generation unit 15, a medication history data generation unit 16, and a recommended drug data generation unit 19. Among these components, the transmission / reception unit 11, the voice data acquisition unit 12, the conversation text generation unit 13, the medication data acquisition unit 14, the prompt generation unit 15, and the medication history data generation unit 16 have the same configurations as the transmission / reception unit 11, the voice data acquisition unit 12, the conversation text generation unit 13, the medication data acquisition unit 14, the prompt generation unit 15, and the medication history data generation unit 16 of the first embodiment of the present invention, and therefore the description thereof will be omitted.

[0089] The recommended drug data generation unit 19 causes the generation AI to extract keywords related to the index information included in the text data based on the text data generated by the conversation text generation unit 13 and the index information stored in the index information storage unit 37. Then, the recommended drug data generation unit 19 extracts drugs corresponding to the extracted keywords based on the drug data stored in the drug data storage unit 33 to generate recommended drug data.

[0090] Specifically, as shown in Fig. 8, the conversation text generating unit 13 converts the voice data acquired by the voice data acquiring unit 12 into text, extracts data related to the patient's current condition and medication status from the text, and generates the extracted data as text data (character data). Here, the text data extracted is "Is there a cold medicine that I can take even if my blood pressure is high?"

[0091] Based on the text data "Is there any cold medicine that I can take even if my blood pressure is high?" generated by the conversational text generation unit 13 and the index information stored in the index information storage unit 37, the generation AI system 4 extracts "high blood pressure" as a keyword related to the index information contained in the text data.

[0092] Then, based on the drug data stored in the drug data storage unit 33, the recommended drug data generation unit 19 extracts "cold medicine ABC" as a drug corresponding to the keyword extracted by the generation AI system 4, and generates recommended drug data as recommended drug data.

[0093] Thus, according to the fourth embodiment of the present invention, the medication history generation support system 100 has an extraction unit 34B, an index information storage unit 37, and a recommended medication data generation unit 19, and therefore can automatically extract appropriate medications for a patient from, for example, a conversation between a pharmacist and a patient.

[0094] <Fifth embodiment> A medication history creation support system according to the fifth embodiment of the present invention will be described below. FIG. 9 is a schematic diagram showing the schematic configuration of a medication history generation support system according to the fifth embodiment of the present invention.

[0095] 9, the medication history generation support system 100 includes a medication history generation support server 1C, front tablets 2A, 2B installed in a store such as a pharmacy, a dispensing server 3C, and a generation AI system 4, all of which are connected via a network 5. Of these components, the front tablets 2A, 2B and the generation AI system 4 have the same configuration as the front tablets 2A, 2B and the generation AI system 4 of the first embodiment of the present invention, and therefore a description thereof will be omitted.

[0096] The dispensing server 3C stores data related to dispensing, and transmits the data to the medication history generation support server 1C via the network 5 upon request. The dispensing server 3C functionally comprises a transmission / reception unit 31, a medication data storage unit 32, a drug data storage unit 33, an extraction unit 34, and a side effect extraction unit 38. Of these components, the transmission / reception unit 31, the medication data storage unit 32, the drug data storage unit 33, and the extraction unit 34 have the same configurations as the transmission / reception unit 31, the medication data storage unit 32, the drug data storage unit 33, and the extraction unit 34 of the first embodiment of the present invention, and therefore a description thereof will be omitted.

[0097] The side effect extraction unit 38 extracts the newly acquired patient's previous prescription data based on the medication data stored in the medication data storage unit 32, and extracts side effect information when the newly acquired patient takes medication based on the extracted previous prescription data, drug data, and patient data.

[0098] The medication history generation support server 1C is composed of a computer, and generates candidates for appropriate medication history data to provide appropriate medication instruction. The medication history generation support server 1C functionally comprises a transmission / reception unit 11, a voice data acquisition unit 12, a conversation text generation unit 13, a medication data acquisition unit 14, a prompt generation unit 15, and a medication history data generation unit 16C. Of these components, the transmission / reception unit 11, the voice data acquisition unit 12, the conversation text generation unit 13, the medication data acquisition unit 14, and the prompt generation unit 15 have the same configurations as the transmission / reception unit 11, the voice data acquisition unit 12, the conversation text generation unit 13, the medication data acquisition unit 14, and the prompt generation unit 15 of the first embodiment of the present invention, and therefore a description thereof will be omitted.

[0099] The medication history data generation unit 16C causes the generation AI to generate candidate medication instruction sentences regarding side effects that correspond to the newly acquired prescription data for the patient, based on the side effect information extracted by the side effect extraction unit 38 and prompt text data for generating candidate medication instruction sentences regarding side effects.

[0100] Thereafter, the medication history data generation unit 16C causes the generation AI system 4 to generate medication history data corresponding to the newly acquired prescription data of the patient via the network 5 based on the prescription data and medication history data extracted by the extraction unit 34 and acquired by the medication data acquisition unit 14, the conversation text data generated based on the audio data of the conversation with the patient in line with the candidate medication instruction sentences regarding side effects generated by the conversation text generation unit 13, and the prompt text data for generating medication history data.

[0101] FIG. 10 is an explanatory diagram illustrating the processing contents at the time of medication instruction in the medication history generation support system 100 according to the fifth embodiment of the present invention.

[0102] As shown in FIG. 10, the dispensing server 3C extracts side effect information D302 when the newly acquired patient takes medication based on the newly acquired patient's previous prescription data D104, the newly acquired patient's patient data D101, and the drug data D301, which are acquired in response to a request from the front tablet 2A.

[0103] Here, as side effect information D302, "side effects related to the two drugs amlodipine and vinprolol fumarate may occur", the patient's hobby "cycling", and the current season "spring" are extracted.

[0104] Then, the medication history data generation unit 16C causes the generation AI system 4 to generate candidates for medication instruction sentences D304 regarding side effects that correspond to the newly acquired prescription data of the patient, based on the side effect information D302 extracted by the side effect extraction unit 38 and the prompt text data D303 for generating candidates for medication instruction sentences regarding side effects.

[0105] Here, the following medication instruction text D304 regarding side effects has been generated: "Did you experience any itchy skin after taking amlodipine last time? It's spring now, so it's a good season for cycling. However, since a new medication has been released, please be careful of any changes in your physical condition."

[0106] Thus, according to the medication history generation support system 100 which is the fifth embodiment of the present invention, since it is equipped with a side effect extraction unit 38 and a medication history data generation unit 16C, for example, when a user (pharmacist) provides medication instructions, the user can properly check all side effects of the previously prescribed medication without missing anything, thereby enabling more appropriate medication instructions to be provided and appropriate medication history data to be created based on that.

[0107] The above-described embodiment can also be realized by executing a medication history creation support program installed in a computer. The medication history creation support program includes a medication support program installed in a server and a terminal program installed in a terminal, and the medication history creation support system is realized by installing each program.

[0108] Sixth embodiment A medication history generation support system 100 according to a sixth embodiment of the present invention will be described below. In the medication history generation support system 100, which is the first embodiment of the present invention, the prompt generation unit 15 internally stores prompt example sentences for classifying into SOAP and generating medication history data, and generates prompt text data based on the prompt example sentences, and the generation AI system 4 generates medication history data based on the generated prompt text data.

[0109] At this time, the medication history data classified as SOAP is streamed out in the order of blocks S→O→A→P, which follows the original inference process of SOAP medication history, so the medication history data for the next block is generated taking into account the medication history data output in the previous block.

[0110] Therefore, if the generation AI system 4 performs evaluation or consideration when generating the A block medical history data, it will generate the P block medical history data by also referring to the A block medical history data on which it performed this evaluation or consideration, and there is a possibility that the P block medical history data will include instructions or explanations that are not included in the conversation content.

[0111] Here, as described above, S (Subjective Data) in SOAP is information from the patient, and is what the patient said and the main complaint. That is, the patient's words are extracted from the conversation content, and the contents that the generation AI system 4 judges should be recorded in the medical history are recorded. O (Objective Data) is the objective state of the patient from the pharmacist's perspective, and includes the patient's physical constitution, allergy history, side effect history, and test results. That is, objective facts such as prescription differences, remaining medication, and results of checks for contraindications for combination use are recorded. A (Assessment) is what the pharmacist thinks from his / her pharmaceutical knowledge based on the information in S and O, and the reason for deciding to implement the plan. That is, evaluations and considerations from the pharmacist's perspective are recorded based on the conversation content, patient information, prescription differences, etc. P (Plan) is the content of instruction by the pharmacist, that is, what the pharmacist actually did and said. That is, the content of the instruction this time from the conversation content and the schedule for follow-up observation from the next time onwards are recorded.

[0112] For example, consider the following conversation between a patient and a pharmacist: Pharmacist: How have you been feeling lately? Patient: "I've been suffering from summer fatigue lately and I get tired easily." In this case, if the medical history data classified in SOAP is generated in the order of S, O, A, P, it will look like this, for example: S: I've been suffering from severe summer fatigue lately. O: The drug MZ was newly prescribed. A: Because this is a new prescription for the drug MZ, it is expected to regulate thyroid hormones, but caution is required regarding side effects. P: Explain the effects and side effects of the drug MZ In this way, the medication history data in Block A has been evaluated and considered by the generation AI system 4, and therefore the medication history data in Block P includes information about the effects and side effects of the drug MZ, which was not mentioned in the conversation.

[0113] Therefore, in the medication history generation support system 100, which is the sixth embodiment of the present invention, the prompt generation unit 15 internally stores prompt example sentences for classifying the medication history data into SOAP and generating the medication history data, and when generating prompt text data based on the prompt example sentences, it generates prompt text data including instructions for generating the medication history data classified into SOAP in the order of S, P, O, and A.

[0114] In this way, by generating the medical history data classified into SOAP in the order of S, P, O, and A, and leaving the A part, which must be entered by guesswork, for the last block, it is possible to prevent fictitious content from being entered in other blocks. This allows the medical history data in the P block to be output based on more facts, and conversely, the medical history data in the A block can be written by working backwards from the contents of the medical history data in the P block, making this an inference process that the generative AI system 4 is good at.

[0115] The output image of medical history data classified into SOAP generated in the order of S, P, O, A is as follows. S: Extract the patient's lines from the conversation and record what the Generative AI System 4 determines should be recorded in the medical history. P: Based on the conversation, patient information, prescription differences, etc., write down the instructions given this time, as well as the schedule for follow-up observation from the next time onwards. O: Record objective facts such as prescription differences, remaining medications, and results of checks for contraindications for concomitant use. A: Working backwards from the content of P, write down the evaluations and considerations that led to the instruction and planning.

[0116] For example, assume that the following conversation takes place between a patient and a pharmacist: Pharmacist: How have you been feeling lately? Patient: "I've been suffering from summer fatigue lately and I get tired easily." In this case, if the medical history data classified in SOAP is generated in the order of S, P, O, A, it will look like this, for example: S: I've been suffering from severe summer fatigue lately. P: Check on recent health condition. Next time, check on summer fatigue and the effects and side effects of the drug MZ. O: The drug MZ was newly prescribed. A: Summer fatigue may also be caused by thyroid hormones, so caution is required. It is necessary to check the effectiveness and side effects of the drug MZ.

[0117] In this way, the medical history data in the P block will be output based on more facts, and conversely, the medical history data in the A block will be written by working backwards from the contents of the medical history data in the P block, making this an inference process that the generative AI system 4 is good at.

[0118] (Additional Note) This application discloses the following inventions. (Appendix 1) a medication data storage unit in which prescription data and medication history data are associated with each other for each patient and each medical department and stored as medication data; An extraction unit that extracts, based on the newly acquired prescription data of a patient, prescription data of the same medical department as the newly acquired prescription data of the patient and medication history data of the same medical department as the newly acquired prescription data of the patient from the medication data; a conversation text generating unit that generates text data of the conversation based on voice data of the conversation with the patient; A medical history data generation unit that causes a generation AI to generate medical history data corresponding to the newly acquired prescription data of the patient based on the extracted prescription data and medical history data, the generated text data, and prompt text data for generating medical history data; A medication history creation support system comprising: (Appendix 2) The text data is text data regarding the current condition and medication status of the patient. 2. A medication history generation support system according to claim 1. (Appendix 3) The prompt text data includes instructions for generating the medical history data by classifying the prompt text data into SOAP. 2. A medication history generation support system according to claim 1. (Appendix 4) The prompt text data includes an instruction to generate the medical history data classified into SOAP in the order of S, P, O, and A. 2. A medication history generation support system according to claim 1. (Appendix 5) A drug data storage unit is further provided in which drugs and contraindications for the drugs are associated with each other and stored as drug data, The medication data storage unit stores patient data including patient attribute data, prescription data, and medication history data in association with each other as medication data, The extraction unit extracts a risk of the newly acquired patient taking medication based on the newly acquired prescription data of the patient, the newly acquired attribute data of the patient, the newly acquired medical history data of the same medical department as the patient, and the drug data; The medical history data generation unit causes a generation AI to generate medical history data corresponding to the newly acquired prescription data of the patient based on the extracted risk, the extracted prescription data, the generated text data, and prompt text data for generating medical history data. 2. A medication history generation support system according to claim 1. (Appendix 6) The extraction unit extracts a difference between the newly acquired prescription data of the patient and the extracted prescription data of the newly acquired patient and prescription data of the same medical department as prescription difference data, The medical history data generation unit causes a generation AI to generate medical history data corresponding to the newly acquired prescription data of the patient based on the extracted prescription data and medical history data, the extracted prescription difference data, the generated text data, and prompt text data for generating medical history data. 2. A medication history generation support system according to claim 1. (Appendix 7) The medication data storage unit stores patient data including patient attribute data, prescription data, and medication history data in association with each other as medication data, The conversation text generating unit generates conversation text data based on voice data of a conversation with a patient who has newly obtained a prescription, extracts information related to the attribute data from the generated text data, and adds the information to the attribute data. 2. A medication history generation support system according to claim 1. (Appendix 8) The medical history data includes test items and test values, a text conversion unit that generates prescription text data by performing character recognition on image data of an image of a prescription; A data update unit that determines whether or not the test items included in the generated prescription text data are appropriate based on the test items included in the medication history data, corrects the test items included in the generated prescription text data if they are not appropriate, and adds the corrected test items and the test values ​​included in the generated prescription text data to the medication history data; 2. The medication history creation support system according to claim 1, further comprising: (Appendix 9) a summary generator for generating a summary of the generated medication history data; a display control unit that switches between displaying the generated medication history data or displaying the summary on a display unit; 2. A medication history generation support system according to claim 1, comprising: (Appendix 10) a medication data storage unit in which prescription data and medication history data are associated with each other for each patient and each medical department and stored as medication data; An extraction unit that extracts, based on the newly acquired prescription data of a patient, prescription data of the same medical department as the newly acquired prescription data of the patient and medication history data of the same medical department as the newly acquired prescription data of the patient from the medication data; a conversation text generating unit that generates text data of the conversation based on voice data of the conversation with the patient; a medication history data generation unit that causes a generation AI to generate medication instruction sentence candidates corresponding to the newly acquired prescription data of the patient based on the extracted prescription data and medication history data and prompt text data for generating medication instruction sentence candidates, and then causes the generation AI to generate medication history data corresponding to the newly acquired prescription data of the patient based on the extracted prescription data and medication history data, the text data generated based on audio data of a conversation with the patient along the medication instruction sentence candidates, and prompt text data for generating medication history data; A medication history creation support system comprising: (Appendix 11) a medication data storage unit in which prescription data and medication history data are associated with each other for each patient and each medical department and stored as medication data; a drug data storage unit in which drugs are associated with contraindications and symptoms for the drugs and stored as drug data; an index information storage unit that stores keywords classified according to patient conditions as index information; a conversation text generating unit that generates text data of the conversation based on voice data of the conversation with the patient; A recommended drug data generation unit that causes a generation AI to extract keywords related to the index information included in the text data based on the generated text data and the index information, and extracts drugs corresponding to the extracted keywords based on the drug data to generate recommended drug data; an extraction unit that generates prescription data for a patient based on the generated recommended drug data, and extracts, from the medication data based on the generated prescription data, prescription data for the same medical department as the generated prescription data for the patient, and medication history data for the same medical department as the patient of the generated prescription data; A medical history data generation unit that causes a generation AI to generate medical history data corresponding to the generated prescription data of the patient based on the extracted prescription data and medical history data, the generated text data, and prompt text data for generating medical history data; A medication history creation support system comprising: (Appendix 12) a medication data storage unit in which patient data including patient attribute data, prescription data, and medical history data are associated with each other and stored as medication data; a drug data storage unit in which drugs and side effect information of the drugs are associated with each other and stored as drug data; a side effect extraction unit that extracts the newly acquired previous prescription data of the patient based on the medication data, and extracts side effect information when the newly acquired patient takes medication based on the extracted previous prescription data, the drug data, and the patient data; an extraction unit that extracts, based on the newly acquired prescription data of the patient, prescription data of the same medical department as the newly acquired prescription data of the patient, and medication history data of the same medical department as the newly acquired prescription data of the patient, from the medication data; a conversation text generating unit that generates text data of the conversation based on voice data of the conversation with the patient; a medication history data generation unit that causes a generation AI to generate candidate medication instruction sentences regarding side effects corresponding to the newly acquired prescription data of the patient based on the extracted side effect information and prompt text data for generating candidate medication instruction sentences regarding side effects, and then causes the generation AI to generate medication history data corresponding to the newly acquired prescription data of the patient based on the extracted prescription data and medication history data, the text data generated based on audio data of a conversation with the patient along the candidate medication instruction sentences regarding side effects, and prompt text data for generating medication history data; A medication history creation support system comprising: (Appendix 13) a medication data storage step of associating the prescription data with the medical history data for each patient and each medical department and storing the data as medication data in a medication data storage unit; An extraction step of extracting, based on the newly acquired prescription data of the patient, prescription data of the same medical department as the newly acquired prescription data of the patient and medication history data of the same medical department as the newly acquired prescription data of the patient from the medication data; a conversation text generation step of generating text data of the conversation based on voice data of the conversation with the patient; A medical history data generation step of causing a generation AI to generate medical history data corresponding to the newly acquired prescription data of the patient based on the extracted prescription data and medical history data, the generated text data, and prompt text data for generating medical history data; A medication history generation support program that causes a computer to execute the above steps. [Explanation of symbols]

[0119] 1, 1A, 1B, 1C Medication history generation support server 2A,2B Front Tablet 3,3A,3B,3C Pharmacy Server 4. Generative AI System 5. Network 11 Transmitter / receiver 12 Audio data acquisition unit 13,13A Conversational text generation unit 14 Medication Data Acquisition Department 15 Prompt Generation 16,16C Medication history data generation unit 17 Summary generator 18 Display control section 19 Recommended Drug Data Generation Unit 31 Transmitter / receiver 32 Medication data storage unit 33 Drug data storage unit 34,34B Extraction part 36 Data Update Section 37 Index information storage unit 38 Side effect extraction part 100 Medication History Generation Support System

Claims

1. a medication data storage unit in which prescription data and medication history data are associated with each other for each patient and each medical department and stored as medication data; An extraction unit that extracts, based on the newly acquired prescription data of a patient, prescription data of the same medical department as the newly acquired prescription data of the patient and medication history data of the same medical department as the newly acquired prescription data of the patient from the medication data; a conversation text generating unit that generates text data of the conversation based on voice data of the conversation with the patient; A medical history data generation unit that causes a generation AI to generate medical history data corresponding to the newly acquired prescription data of the patient based on the extracted prescription data and medical history data, the generated text data, and prompt text data for generating medical history data; A medication history creation support system comprising:

2. The text data is text data regarding the current condition and medication status of the patient. The medication history generation support system according to claim 1 .

3. The prompt text data includes instructions for generating the medical history data by categorizing the data into SOAP. The medication history generation support system according to claim 1 .

4. The prompt text data includes an instruction to generate the medical history data classified into SOAP in the order of S, P, O, and A. The medication history generation support system according to claim 1 .

5. A drug data storage unit is further provided in which drugs and contraindications for the drugs are associated with each other and stored as drug data, The medication data storage unit stores patient data including patient attribute data, prescription data, and medical history data in association with each other as medication data, The extraction unit extracts a risk of the newly acquired patient taking medication based on the newly acquired prescription data of the patient, the newly acquired attribute data of the patient, the newly acquired medical history data of the same medical department as the patient, and the drug data; The medical history data generation unit causes a generation AI to generate medical history data corresponding to the newly acquired prescription data of the patient based on the extracted risk, the extracted prescription data, the generated text data, and prompt text data for generating medical history data. The medication history generation support system according to claim 1 .

6. The extraction unit extracts a difference between the newly acquired prescription data of the patient and the extracted newly acquired prescription data of the patient and prescription data of the same medical department as prescription difference data, The medical history data generation unit causes a generation AI to generate medical history data corresponding to the newly acquired prescription data of the patient based on the extracted prescription data and medical history data, the extracted prescription difference data, the generated text data, and prompt text data for generating medical history data. The medication history generation support system according to claim 1 .

7. The medication data storage unit stores patient data including patient attribute data, prescription data, and medical history data in association with each other as medication data, The conversation text generating unit generates conversation text data based on voice data of a conversation with a patient who has newly obtained a prescription, extracts information related to the attribute data from the generated text data, and adds the information to the attribute data. The medication history generation support system according to claim 1 .

8. The medical history data includes test items and test values, a text conversion unit that generates prescription text data by performing character recognition on image data of an image of a prescription; A data update unit that determines whether or not the test items included in the generated prescription text data are appropriate based on the test items included in the medication history data, corrects the test items included in the generated prescription text data if they are not appropriate, and adds the corrected test items and the test values ​​included in the generated prescription text data to the medication history data; The medication history generation support system according to claim 1, further comprising:

9. a summary generator for generating a summary of the generated medication history data; a display control unit that switches between displaying the generated medication history data or displaying the summary on a display unit; The medication history generation support system according to claim 1, further comprising:

10. a medication data storage unit in which prescription data and medication history data are associated with each other for each patient and each medical department and stored as medication data; An extraction unit that extracts, based on the newly acquired prescription data of a patient, prescription data of the same medical department as the newly acquired prescription data of the patient and medication history data of the same medical department as the newly acquired prescription data of the patient from the medication data; a conversation text generating unit that generates text data of the conversation based on voice data of the conversation with the patient; a medication history data generation unit that causes a generation AI to generate medication instruction sentence candidates corresponding to the newly acquired prescription data of the patient based on the extracted prescription data and medication history data and prompt text data for generating medication instruction sentence candidates, and then causes a generation AI to generate medication history data corresponding to the newly acquired prescription data of the patient based on the extracted prescription data and medication history data, the text data generated based on audio data of a conversation with the patient along the medication instruction sentence candidates, and prompt text data for generating medication history data; A medication history creation support system comprising:

11. a medication data storage unit in which prescription data and medication history data are associated with each other for each patient and each medical department and stored as medication data; a drug data storage unit in which drugs are associated with contraindications and symptoms for the drugs and stored as drug data; an index information storage unit that stores keywords classified according to patient conditions as index information; a conversation text generating unit that generates text data of the conversation based on voice data of the conversation with the patient; A recommended drug data generation unit that causes a generation AI to extract keywords related to the index information included in the text data based on the generated text data and the index information, and extracts drugs corresponding to the extracted keywords based on the drug data to generate recommended drug data; an extraction unit that generates prescription data for a patient based on the generated recommended drug data, and extracts, from the medication data based on the generated prescription data, prescription data for the same medical department as the generated prescription data for the patient, and medication history data for the same medical department as the patient of the generated prescription data; A medical history data generation unit that causes a generation AI to generate medical history data corresponding to the generated prescription data of the patient based on the extracted prescription data and medical history data, the generated text data, and prompt text data for generating medical history data; A medication history creation support system comprising:

12. a medication data storage unit in which patient data including patient attribute data, prescription data, and medical history data are associated with each other and stored as medication data; a drug data storage unit in which drugs and side effect information of the drugs are associated with each other and stored as drug data; a side effect extraction unit that extracts the newly acquired previous prescription data of the patient based on the medication data, and extracts side effect information when the newly acquired patient takes medication based on the extracted previous prescription data, the drug data, and the patient data; an extraction unit that extracts, based on the newly acquired prescription data of the patient, prescription data of the same medical department as the newly acquired prescription data of the patient, and medication history data of the same medical department as the newly acquired prescription data of the patient, from the medication data; a conversation text generating unit that generates text data of the conversation based on voice data of the conversation with the patient; a medication history data generation unit that causes a generation AI to generate candidate medication instruction sentences regarding side effects corresponding to the newly acquired prescription data of the patient based on the extracted side effect information and prompt text data for generating candidate medication instruction sentences regarding side effects, and then causes a generation AI to generate medication history data corresponding to the newly acquired prescription data of the patient based on the extracted prescription data and medication history data, the text data generated based on audio data of a conversation with the patient along the candidate medication instruction sentences regarding side effects, and prompt text data for generating medication history data; A medication history creation support system comprising:

13. a medication data storage step of associating the prescription data with the medical history data for each patient and each medical department and storing the data as medication data in a medication data storage unit; An extraction step of extracting, based on the newly acquired prescription data of the patient, prescription data of the same medical department as the newly acquired prescription data of the patient and medication history data of the same medical department as the newly acquired prescription data of the patient from the medication data; a conversation text generation step of generating text data of the conversation based on voice data of the conversation with the patient; A medical history data generation step of causing a generation AI to generate medical history data corresponding to the newly acquired prescription data of the patient based on the extracted prescription data and medical history data, the generated text data, and prompt text data for generating medical history data; A medication history generation support program that causes a computer to execute the above steps.

Images