Program, information processing method, and information processing apparatus.

Abstract

To provide a program or similar tool that can output information related to education for healthcare professionals. [Solution] The program acquires information about healthcare professionals and causes a computer to execute a process that outputs information about education for the healthcare professionals using a language model that utilizes the acquired information about healthcare professionals and information about medicine. Preferably, the program acquires information about the skills of the healthcare professionals and target skill information, and outputs information about education for the healthcare professionals using a language model that utilizes the acquired skill information and target skill information and information about medicine.

Description

Technical Field

[0004] ,

[0006] , , , , , ,

[0005] , , , ,

[0007] , , ,

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

Background Art

[0002] In Patent Document 1, a medical record creation support system is disclosed that creates a SOAP format medical record by classifying text consisting of a plurality of clauses input by medical staff such as doctors into SOAP format sections.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the invention according to Patent Document 1 has a problem that it cannot output information regarding education for medical staff.

[0005] In one aspect, an object is to provide a program or the like that can output information regarding education for medical staff.

Means for Solving the Problems

[0006] In the present invention, (1) the program causes a computer to execute a process of acquiring information regarding medical staff and outputting information regarding education for the medical staff using the acquired information regarding medical staff and a language model regarding medical information.

[0007] In this embodiment of the present invention, (2) the program described in (1) above preferably acquires the skill information of the medical professional and the target skill information, and outputs information regarding education for the medical professional using a language model that uses the acquired skill information and target skill information and medical information.

[0008] (3) The program described in (1) or (2) above preferably outputs information regarding education for healthcare professionals by providing the language model with the acquired skill information and target skill information, and the healthcare information extracted in relation to the skill information and target skill information.

[0009] (4) The program described in any of (1) to (3) above preferably acquires the medical schedule of the medical professional and outputs an educational schedule for the medical professional using a language model that uses the acquired medical schedule, skill information and target skill information and medical information.

[0010] (5) The program described in any of (1) to (4) above preferably generates a prompt that includes the medical schedule, the skill information, the target skill information, the medical information, and the instruction to generate the education schedule, and outputs the education schedule by providing the generated prompt to a language model.

[0011] (6) The program described in any of (1) to (5) above preferably outputs a confirmation test for the healthcare professional using a language model that utilizes the healthcare professional's skill information, target skill information, medical information, and the educational schedule for the healthcare professional.

[0012] (7) The program described in any of (1) to (6) above preferably generates a prompt including the skill information, the target skill information, the medical information, the education schedule, and the instruction to generate the confirmation test, and outputs the confirmation test by providing the generated prompt to the language model.

[0013] (8) The program described in any of (1) to (7) above preferably acquires video or audio related to a medical procedure and provides the acquired video or audio related to the medical procedure to the language model to output a confirmation test for the medical professional or summary information related to the video or audio.

[0014] (9) Preferably, the program described in any of (1) to (8) above generates a prompt that includes the medical video or audio and the instruction to generate the confirmation test or the instruction to generate the summary information, and outputs the confirmation test or the summary information by providing the generated prompt to a language model.

[0015] (10) The program described in any of (1) to (9) above preferably acquires information on the classification of the medical institution to which the medical professional belongs, the treatment record of the medical institution, and the medical care delivery system of the medical institution, and outputs information on education for the medical professional using the language model which uses the acquired information on the classification, treatment record, and medical care delivery system, as well as information on medical care.

[0016] (11) In any of the programs described in (1) to (10) above, it is preferable that the information regarding the medical care delivery system of the medical institution includes the nursing system of the medical institution or the treatment system of the medical institution.

[0017] (12) The program described in any of (1) to (11) above preferably outputs information regarding education for healthcare professionals by providing the language model with the acquired information regarding classification, treatment results and healthcare delivery system, and the information regarding healthcare extracted in relation to the classification, treatment results and healthcare delivery system.

[0018] (13) The program described in any of (1) to (12) above preferably generates prompts that include commands to generate the classification, the treatment results, the information on the medical care delivery system, the information on medical care, and the information on education for medical professionals, and outputs the information on education for medical professionals by providing the generated prompts to a language model.

[0019] (14) Information processing method relating to one aspect of this disclosure acquires information relating to healthcare professionals, Using a language model that utilizes acquired information about healthcare professionals and medical information, information regarding education for said healthcare professionals is output.

[0020] (15) An information processing device relating to one aspect of this disclosure acquires information relating to healthcare workers, The system includes a control unit that performs processing to output information regarding education for healthcare professionals, using a language model that utilizes acquired information about healthcare professionals and information about medical care. [Effects of the Invention]

[0021] One aspect of this is that it becomes possible to output information related to education for healthcare professionals. [Brief explanation of the drawing]

[0022] [Figure 1]It is an explanatory diagram showing the outline of an information processing system. [Figure 2] It is a block diagram showing a configuration example of a server. [Figure 3] It is an explanatory diagram showing an example of the record layout of a medical staff DB. [Figure 4] It is a block diagram showing a configuration example of a tablet. [Figure 5] It is an explanatory diagram showing an example of an input screen. [Figure 6] It is an explanatory diagram showing an example of a prompt. [Figure 7] It is an explanatory diagram showing an example of a screen. [Figure 8] It is a flowchart showing an example of the processing procedure of an information processing system. [Figure 9] It is an explanatory diagram showing an example of a prompt. [Figure 10] It is an explanatory diagram showing an example of a screen. [Figure 11] It is a flowchart showing an example of the processing procedure of an information processing system. [Figure 12] It is an explanatory diagram showing the outline of a second language model. [Figure 13] It is an explanatory diagram showing an example of an acquisition screen. [Figure 14] It is an explanatory diagram showing an example of a prompt. [Figure 15] It is an explanatory diagram showing an example of a prompt. [Figure 16] It is an explanatory diagram showing an example of a screen. [Figure 17] It is a flowchart showing an example of the processing procedure of an information processing system. [Figure 18] It is a flowchart showing an example of the processing procedure of an information processing system. [Figure 19] It is a block diagram showing a configuration example of a server. [Figure 20] It is an explanatory diagram showing an example of the record layout of a medical institution DB. [Figure 21] It is an explanatory diagram showing an example of the record layout of a performance DB. [Figure 22]This is an explanatory diagram showing an example of a prompt. [Figure 23] This is an explanatory diagram showing an example screen. [Figure 24] A flowchart illustrating an example of the processing procedure for an information processing system. [Modes for carrying out the invention]

[0023] (Embodiment 1) Figure 1 is an explanatory diagram illustrating the overview of the information processing system. The information processing system outputs information about healthcare professionals and information about education for healthcare professionals based on healthcare information. The information processing system includes an information processing device 10 and an information processing device 20. The information processing system transmits and receives information via a network.

[0024] The information processing device 10 is an information processing device that processes, stores, and transmits various types of information. The information processing device 10 is, for example, a server device, a personal computer, or a general-purpose tablet PC (personal computer). The information processing device 10 may also be a cloud server device that provides the functions included in the information processing device 10 as a cloud service. In this embodiment, the information processing device 10 will be described as a server 10.

[0025] The information processing device 20 is, for example, a server device, a smartphone, a tablet, a personal computer (hereinafter referred to as "computer"), or a general-purpose tablet PC. In this embodiment, the information processing device 20 will be described as a tablet 20.

[0026] Figure 2 is a block diagram showing an example of a server configuration. Server 10 includes a control unit 11, a storage unit 12, a communication unit 13, a large-capacity storage unit 14, and a read unit 15. Each of the above-mentioned units is interconnected via a bus. The control unit 11 is configured using one or more processors such as a CPU (Central Processing Unit), MPU (Micro-Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-purpose computing on graphics processing units), TPU (Tensor Processing Unit), or AI chip (AI semiconductor). The control unit 11 performs various information processing and control processing related to Server 10 by appropriately executing a control program 12P (program product) stored in the storage unit 12.

[0027] The storage unit 12 includes RAM (Random Access Memory) or ROM (Read Only Memory), etc. The storage unit 12 stores various data necessary for the control program 12P executed by the control unit 11. The storage unit 12 temporarily stores data generated when the control unit 11 executes the control program 12P.

[0028] Furthermore, the memory unit 12 stores the language model M1. The language model M1 is a general-purpose large language model (LLM: Large Language Model) constructed by performing unsupervised pre-training using, for example, a large set of texts and images. Examples of language models M1 include GPT (Generative Pre-trained Transformer)-3, GPT-3.5, GPT-4, RWKV (Receptance Weighted Key Value), PaLM2, or LLaMa (Large Language Model Meta AI).

[0029] When language model M1 receives information about healthcare professionals and information about medical care as input, it performs calculations to generate information about education for healthcare professionals and outputs the generated information about education for healthcare professionals. Additionally, when language model M1 receives information about healthcare professionals, information about medical care, and medical-related schedules concerning healthcare professionals (hereinafter referred to as medical-related schedules) as input, it performs calculations to generate information about education for healthcare professionals and outputs the generated information about education for healthcare professionals.

[0030] Information concerning healthcare professionals includes information on the healthcare professional's skills (hereinafter referred to as "skill information"), information on the healthcare professional's target skills (hereinafter referred to as "target skill information"), the classification of the medical institution to which the healthcare professional belongs, the medical institution's treatment record, or information on the medical care delivery system of the medical institution. In Embodiment 1, information concerning healthcare professionals is described as "skill information" and "target skill information."

[0031] Medical schedules include, for example, the work schedules of medical professionals, the surgical schedules of catheterization labs that medical professionals are responsible for, the schedules of the medical institutions to which medical professionals belong, or the training schedules of medical professionals who are the subjects of learning. Information regarding education for medical professionals includes, for example, the education schedules for medical professionals (hereinafter referred to as "education schedules"), the assessment tests for medical professionals, or the training plans for the medical institutions to which medical professionals belong.

[0032] In Embodiment 1, information regarding education for healthcare professionals is described as an "educational schedule." In Embodiments 2 and 3, information regarding education for healthcare professionals is described as a "confirmation test." Information regarding medical care will be described later.

[0033] The language model M1 may be constructed by combining multiple algorithms. Instead of storing the language model M1 in the storage unit 12, the control unit 11 may access an external server (not shown) that stores the language model M1 and use it.

[0034] The communication unit 13 is a communication module that sends and receives information to and from the tablet 20 via a network. The large-capacity storage unit 14 includes RAM or ROM, etc. The large-capacity storage unit 14 stores the medical professional database 141 and the medical information file 142. The medical professional database 141 and the medical information file 142 will be described later.

[0035] In this embodiment, the storage unit 12 and the large-capacity storage unit 14 may be configured as a single storage device. The large-capacity storage unit 14 may be composed of multiple storage devices. The large-capacity storage unit 14 may also be an external storage device connected to the server 10.

[0036] The reading unit 15 reads information stored in the portable storage medium 1a. The portable storage medium 1a is, for example, a CD (Compact Disc), DVD (Digital Versatile Disc), USB (Universal Serial Bus) memory, or SD (Secure Digital). The reading unit 15 reads the control program 12P from the portable storage medium 1a. The control unit 11 stores the read control program 12P in the storage unit 12.

[0037] The control unit 11 may download the control program 12P from another computer via the network. In that case, the control unit 11 stores the downloaded control program 12P in the storage unit 12. Alternatively, the control unit 11 may store the read control program 12P in the large-capacity storage unit 14.

[0038] In this embodiment, server 10 may be composed of multiple servers. Server 10 may be a virtual machine virtually constructed by software within a single device. Server 10 may be a local server installed in the facility where server 10 is located. Server 10 may be a cloud server connected via a network. Furthermore, the control program 12P may run on a single server, or it may be distributed and run on multiple servers interconnected via a network.

[0039] Figure 3 is an explanatory diagram showing an example of the record layout of the healthcare professional database. Healthcare professional database 141 stores healthcare professional ID, skill information, and target skill information. Skill information indicates, for example, the healthcare professional's current work performance ability and work performance status, and includes subjects taught, years of experience, work content, performance status, and remarks. Target skill information indicates, for example, the ideal state that the healthcare professional should aim for, and includes common goals for all employees (hereinafter referred to as common goals) and individual goals.

[0040] In this embodiment, an example is described in which the information system acquires individual goals via the input unit 24 of the tablet 20 (see Figure 4) and reads skill information and other target skill information (common goals) from the medical professional database 141, but it is not limited to this. The information system may read all skill information and all target skill information from the medical professional database 141. Alternatively, the information system may acquire all skill information and all target skill information via the input unit 24 of the tablet 20.

[0041] The healthcare professional database 141 includes columns for healthcare professional ID, skills information, and target skills information. The healthcare professional ID column stores the healthcare professional ID to identify the healthcare professional. The skills information column includes columns for assigned subjects, years of experience, job content, performance status, and remarks. The assigned subjects column stores the subjects assigned to the healthcare professional. The years of experience column stores the healthcare professional's years of experience. The job content column stores the healthcare professional's current job content. The performance status column stores the performance status of the healthcare professional's work. The performance status of a healthcare professional's work is, for example, whether or not the healthcare professional can perform a particular task. The remarks column stores remarks regarding the healthcare professional's skills information. The common goals column stores the common goals of the healthcare professionals.

[0042] The record for healthcare worker ID "I001" in Figure 3 contains the following information: Department of responsibility: "Main duty: Emergency and Critical Care Medicine, concurrent duty: Catheterization Laboratory", Years of experience: "4 years (Catheterization Laboratory)", Job description: "Performs nursing duties in the catheterization laboratory about twice a month. Basically, only performs nursing duties for emergency patients in the catheterization laboratory.", and Status of performance: "Can perform basic nursing duties according to the manual on their own. However, cannot perform nursing duties for emergency patients without instructions from a supervising nurse. Because there are emergency patients they have no experience with, they perform nursing duties in the catheterization laboratory with anxiety."

[0043] Furthermore, the record for healthcare professional ID "I001" stores the following notes: "Has learned basic nursing tasks through on-the-job training (OJT). However, has not received OJT in the catheterization lab because it is infrequently used for procedures there," and the common goal: "Can respond appropriately to sudden changes in a patient's condition during catheterization procedures, in accordance with the doctor's response and the patient's situation. Can also respond flexibly to irregular cases." In addition, the healthcare professional database 141 may store the healthcare professional's name, age, gender, etc., associated with the healthcare professional ID.

[0044] The medical information file 142 stores medical information, the chunks into which the medical information is divided, and the vectorized chunks. A chunk is a component of medical information, such as text, paragraphs, sentences, clauses, and words. The medical information includes knowledge data on drugs, knowledge data on complications, knowledge data on medical procedures, knowledge data on medical devices, medical glossaries, manuals for medical devices, or performance data on education.

[0045] Knowledge data on drugs includes the effects, side effects, and methods of use of drugs used in medical institutions (e.g., antithrombotic agents, acetylcholine, and iodine contrast agents). Knowledge data on complications includes the mechanisms of occurrence, risks, and management methods of complications that occur in medical institutions (e.g., wound infection, pneumonia, and delirium). Knowledge data on medical procedures includes the types of medical procedures performed in medical institutions (e.g., blood sampling, anesthetic injection, and catheter removal), the methods of performing such procedures, precautions for such procedures, and image information of such procedures.

[0046] Knowledge data on drugs, knowledge data on complications, and knowledge data on medical procedures, for example, consist of several chapters (e.g., chapters 1 to 10) totaling several hundred pages, and are organized by field. Knowledge data on medical devices includes, for example, image information of medical devices used in medical settings, the names of each part of the medical device, and instructions and precautions for using the medical device. This knowledge data on medical devices also consists of several hundred pages and is organized by field.

[0047] The medical glossary is a glossary that explains specialized medical terms (e.g., MI and HAV) and consists of several hundred pages. The manual on medical devices includes, for example, the types and usage methods of medical devices used in medical institutions (imaging devices, stent grafts, and ECMO) and consists of several chapters and several hundred pages. The data on educational achievements includes, for example, the classification of medical institutions that have implemented educational schedules in the past, the implementation system, results, and teaching materials used.

[0048] This document describes a method for storing medical information, the divided chunks of medical information, and the vectorized chunks in a medical information file. Specifically, a medical glossary, the divided chunks of the medical glossary, and the vectorized chunks are stored in a medical information file 142. The control unit 11 reads the medical glossary via the communication unit 13 or the reading unit 15. The control unit 11 sets hyperlinks to each page of the read medical glossary. The control unit 11 divides the medical glossary into a number of chunks. Each divided chunk may include overlapping portions.

[0049] When the control unit 11 divides the medical glossary into numerous chunks, it associates each chunk with the hyperlink to the page in which that chunk is described. It may also associate each chunk with the hyperlink to the chapter in which that chunk is described. By associating each chunk with the hyperlink to the page in which that chunk is described, the control unit 11 can identify the page in which that chunk is described and the original text of that page. The control unit 11 vectorizes each divided chunk. The control unit 11 stores the medical glossary, each divided chunk of the medical glossary, and each vectorized chunk in association in the medical information file 142. The control unit 11 performs similar processing on medical information other than the medical glossary.

[0050] Figure 4 is a block diagram showing an example of the tablet's configuration. The tablet 20 includes a control unit 21, a storage unit 22, a communication unit 23, an input unit 24, and a display unit 25. The above-mentioned units are interconnected via a bus. The control unit 21 is configured using one or more processors such as a CPU, MPU, or GPU. The storage unit 22 includes RAM or ROM. The storage unit 22 stores various data necessary for the control program 22P (program product) executed by the control unit 21. The storage unit 22 temporarily stores data generated when the control program 22P is executed. The control unit 21 performs information processing and control processing related to the tablet 20 by appropriately executing the control program 22P stored in the storage unit 22.

[0051] The communication unit 23 is a communication module that sends and receives information to and from the server 10 via a network. The display unit 25 is, for example, a liquid crystal panel or an organic EL (electro-luminescence) panel. The input unit 24 is stacked on the display unit 25. The tablet 20 may also have the display unit 25 and the input unit 24, such as a mouse, separately.

[0052] In the following description, we will use as an example a case in which a medical professional uses a tablet 20 with the program of this embodiment already installed. When the control unit 21 executes the program of this embodiment, it displays an input screen for the medical professional ID, etc., on the display unit 25. Figure 5 is an explanatory diagram showing an example of an input screen. The input screen d1 is a screen that accepts input from a medical professional, such as a medical professional ID, target skill information, or medical-related schedule, and includes an ID field d11, a personal goal field d12, a registration button b1, a schedule addition button b2, and a schedule field d13.

[0053] The ID field d11 is where the healthcare professional's ID is entered. In Figure 5, the ID field d11 is entered as "I001". The name of the healthcare professional may also be entered in the ID field d11. The following explanation will describe an example where the healthcare professional ID "I001" is entered in the ID field d11. The personal goal field d12 is where the personal goal included in the target skill information is entered. In Figure 5, the personal goal field d12 is entered as "To be able to act as a nurse without the help of a supervising nurse. To be able to respond confidently even if a sudden change occurs during a procedure."

[0054] The Add Schedule button b2 is used to add medical-related schedules. When the Add Schedule button b2 is selected, the newly added medical-related schedule is displayed in the schedule field d13. In Figure 5, the schedule field d13 displays "Catheterization Lab Surgery Schedule.txt" and "Work Schedule.xlsx". If the Add Schedule button b2 is not selected, nothing is displayed in the schedule field d13. The Register button b1 is used to send the entered or selected medical professional ID, personal goals, and medical-related schedule to the server 10.

[0055] If the Add Schedule button b2 is selected, a medical-related schedule stored in the storage unit 12 or the large-capacity storage unit 14 may be read. The process in this case will be described below. When the Add Schedule button b2 is selected, the control unit 21 sends a notification to the server 10 indicating that the schedule will be read (hereinafter referred to as the read notification) and the medical professional ID. The control unit 11 receives the read notification and medical professional ID sent from the tablet 20. The control unit 11 reads the medical-related schedule corresponding to the medical professional ID from the storage unit 12 or the large-capacity storage unit 14. The control unit 11 sends the read medical-related schedule to the tablet 20. The control unit 21 receives the medical-related schedule sent from the server 10. The control unit 21 displays the received medical-related schedule in the schedule field d13.

[0056] In this embodiment, an example has been described in which the schedule addition button b2 and the schedule field d13 are placed on the input screen d1, but this is not limited to this. The schedule addition button b2 and the schedule field d13 do not necessarily have to be placed on the input screen d1.

[0057] The control unit 21 determines whether or not the selection of registration button b1 has been accepted. If the control unit 21 determines that the selection of registration button b1 has not been accepted, it again accepts input of the medical professional ID, personal goals, or medical-related schedule. If the control unit 21 determines that the selection of registration button b1 has been accepted, it sends the medical professional ID, personal goals, and medical-related schedule to the server 10.

[0058] In the example where the schedule addition button b2 and schedule field d13 are not placed on the input screen d1, the control unit 21, upon determining that the registration button b1 has been selected, sends the medical professional ID and personal goal to the server 10.

[0059] The control unit 11 receives the medical professional ID, personal goals, and medical-related schedule transmitted from the tablet 20. The control unit 11 reads the assigned subject, years of experience, job content, implementation status, remarks, and common goals corresponding to the medical professional ID "I001" from the medical professional DB 141. The control unit 11 vectorizes the skill information and target skill information acquired through input or read. The control unit 11 calculates the similarity between the vectorized skill information and target skill information and each vectorized chunk stored in the medical information file 142. For calculating the similarity between vectors, for example, cosine similarity or k-nearest neighbors is used.

[0060] The control unit 11 identifies vectorized chunks whose calculated similarity is equal to or greater than a predetermined threshold. The control unit 11 extracts medical information corresponding to the identified chunks from the medical information file 142. If the control unit 11 extracts multiple pieces of medical information, it may re-rank the multiple pieces of medical information. In this case, for example, Maximum Marginal Relevance (MMR) may be used for re-ranking. Through re-ranking, the control unit 11 extracts the top few items (e.g., the top 3 items) of medical information that are more highly relevant to the skill information and target skill information. By the above method, the control unit 11 extracts medical information related to the skill information and target skill information from multiple pieces of medical information.

[0061] The control unit 11 generates a prompt 30 (hereinafter referred to as "prompt 30") which includes acquired or extracted skill information, target skill information, medical information, and an instruction to generate an educational schedule. The instruction to generate an educational schedule is pre-stored in the storage unit 12 or the large-capacity storage unit 14. The instruction to generate an educational schedule can be modified as appropriate to suit the embodiment.

[0062] Figure 6 is an explanatory diagram showing an example of a prompt. Prompt 30 includes a skill information field 31, a target skill information field 32, a medical information field 33, a schedule field 34, and a generation command field 35. The skill information field 31 contains the skill information of the medical professional. The medical professional's skill information is read from the medical professional DB 141. The content written in the skill information field 31 is formed by applying the read skill information to a pre-prepared template.

[0063] Figure 6, in the skills information section 31, states: "The skills information of the medical professional is as follows: The primary duty is in the emergency and critical care department, with concurrent duties in the catheterization lab. The number of years of experience in the catheterization lab is 4 years. Nursing work in the catheterization lab is performed about twice a month. Basically, only nursing work for emergency patients is performed in the catheterization lab. Basic nursing work according to the manual can be performed alone, but nursing work for sudden changes in condition cannot be performed without instructions from a supervising nurse. Because there are emergency patients that the nurse has no experience with, she performs nursing work in the catheterization lab with anxiety. She is learning basic nursing work through on-the-job training (OJT). However, because she is assigned to the catheterization lab infrequently, she has not received OJT in the catheterization lab."

[0064] The Target Skill Information field 32 contains the target skill information of healthcare professionals. Common goals included in the target skill information are read from the healthcare professional database 141. Individual goals included in the target skill information are obtained through input by healthcare professionals. The content written in the Target Skill Information field 32 is composed of applying the target skill information obtained through input or retrieval to a pre-prepared template.

[0065] In Figure 6, under the section 32 for target skills information, it states: "The target skills for healthcare professionals are as follows: To be able to respond appropriately to sudden changes in a patient's condition during catheterization procedures, in accordance with the doctor's response and the patient's situation. To be able to respond flexibly to irregular cases. To be able to respond as a nurse without the need for assistance from senior nurses. To be able to respond confidently to sudden changes in a patient's condition during procedures."

[0066] Section 33, which contains medical information, includes medical information extracted in relation to skill information and target skill information. Section 33 of the medical information section in Figure 6 states: "The medical information is as follows: In on-the-job training (OJT) in the catheterization lab, the video manual "***" is frequently used. In emergency response in the catheterization lab, "***" is often implemented. The combination of lectures and simulated experiences increases the retention rate of knowledge in OJT."

[0067] The schedule column 34 displays the medical-related schedule selected via the input screen d1. If no medical-related schedule has been received from the tablet 20, the schedule column 34 will be blank. In Figure 6, the schedule column 34 displays "Catheterization Lab Surgery Schedule.txt" and "Work Schedule.xlsx". In Figure 6, the schedule column 34 may also display the training schedule of the medical professional who is the learning target.

[0068] The generation command field 35 contains instructions for the language model M1 to output an educational schedule. The contents of the generation command field 35 are pre-stored in the memory unit 12 or the large-capacity memory unit 14. The generation command field 35 in Figure 6 contains the following: "As a supervising nurse, please present the educational schedule and the necessary teaching materials for that schedule. When presenting the educational schedule and necessary teaching materials, please consider the following: The educational schedule should be 3 to 5 months long. Please refer to the attached medical schedule. Do not include the lunch break (12:00-13:00). Please make efficient use of any free time during working hours."

[0069] The control unit 11 inputs the generated prompt 30 to the language model M1, thereby obtaining the educational schedule generated by the language model M1. The control unit 11 stores the skill information, target skill information, medical information, and educational schedule in the large-capacity storage unit 14, associating them with the medical professional ID. The control unit 11 transmits the acquired educational schedule to the tablet 20.

[0070] The control unit 21 receives the educational schedule transmitted from the server 10. The control unit 21 displays the received educational schedule on the display unit 25. Figure 7 is an explanatory diagram showing an example screen. Screen d2 in Figure 7 includes areas d21, d22, and d23. Area d21 displays the educational schedule output by the language model M1. The educational schedule includes the days of the week, time slots, content, and related teaching materials.

[0071] The time slots for implementing the training schedule are, for example, during breaks in working hours. The content of the training schedule includes, for example, review tests and summary tests conducted during the training schedule's implementation time slots. The review tests and summary tests include multiple formats such as multiple choice, fill-in-the-blank, written, and image selection. The related materials for the training schedule are, for example, materials related to the review tests and summary tests. Also, if the training materials displayed in area d21 are part of the medical information referenced from medical information file 142, such materials are indicated by a double dashed line and a number. Area d22 displays the related materials for the training schedule. The display of related materials may be a brief text description of the material's content, or it may be image data such as a photograph of related medical equipment or a photograph showing the usage of said medical equipment.

[0072] Area d23 displays hyperlinks to show the source of educational materials, etc., indicated by double dashed lines and numbers in area d21. When a hyperlink is selected, the control unit 21 sends an instruction to the server 10 to request information about the source medical materials. The control unit 11 receives the instruction from the tablet 20 to request information about the source medical materials. The control unit 11 reads the information about the source medical materials from the medical information file 142. The control unit 11 sends the read information about the source medical materials to the tablet 20. The control unit 21 receives the information about the source medical materials sent from the server 10. The control unit 21 overlays the information about the source medical materials onto screen d2. The control unit 21 may split the display unit 25 to show the information about the source medical materials and screen d2 separately.

[0073] Figure 8 is a flowchart showing an example of the processing procedure of the information processing system. When the program of this embodiment is executed, the control unit 21 displays the input screen d1 for the medical professional ID, etc., on the display unit 25 (step S101). The control unit 11 acquires the medical professional ID and personal goals entered by the medical professional (step S102). The control unit 11 determines whether or not the selection of the schedule addition button b2 has been accepted (step S103). If the control unit 11 determines that the schedule addition button b2 has been accepted (step S103: YES), it acquires the medical-related schedule selected by the medical professional (step S104).

[0074] After step S104 is completed, or if it is determined that the selection of the schedule addition button b2 has not been accepted (step S103: NO), the control unit 11 determines whether or not the selection of the registration button b1 has been accepted (step S105). If the control unit 11 determines that the selection of the registration button b1 has not been accepted (step S105: NO), it returns to step S102. If the control unit 11 determines that the selection of the registration button b1 has been accepted (step S105: YES), it sends the acquired medical professional ID, personal goals, and medical-related schedule to the server 10 (step S106).

[0075] Furthermore, if the Add Schedule button b2 is not selected, the control unit 11 may send the acquired medical professional ID and personal goal to the server 10. The following flowchart will explain an example where the Add Schedule button b2 is selected.

[0076] The control unit 11 receives the healthcare professional ID, personal goals, and medical-related schedule (step S201). The control unit 11 reads the skill information and common goals corresponding to the healthcare professional ID from the healthcare professional DB 141 (step S202). The control unit 11 vectorizes the skill information and target skill information acquired through input or readout (step S203).

[0077] The control unit 11 calculates the similarity between the vectorized skill information and target skill information and each vectorized chunk stored in the medical information file 142 (step S204). The control unit 11 identifies vectorized chunks whose calculated similarity is above a predetermined threshold (step S205). The control unit 11 extracts medical information corresponding to the identified chunks from the medical information file 142 (step S206).

[0078] The control unit 11 generates a prompt 30 by combining the acquired or extracted skill information, target skill information, medical information, medical schedule, and education schedule generation commands (step S207). The control unit 11 inputs the generated prompt 30 to the language model M1 to acquire the education schedule generated by the language model M1 (step S208). The control unit 11 stores the skill information, target skill information, medical information, and education schedule in the large-capacity storage unit 14, associating them with the medical professional ID (step S209). The control unit 11 transmits the acquired education schedule to the tablet 20 (step S210).

[0079] The control unit 21 receives the training schedule transmitted from the server 10 (step S107). The control unit 21 displays the received training schedule on the display unit 25 (step S108).

[0080] According to Embodiment 1, the information processing system can acquire information about medical professionals and output information about education for those medical professionals using a language model that utilizes the acquired information about medical professionals and information about medical care.

[0081] According to Embodiment 1, the information processing system acquires skill information of healthcare professionals and target skill information, and can output information related to education for healthcare professionals using a language model that utilizes the acquired skill information and target skill information, along with information related to healthcare.

[0082] According to Embodiment 1, the information processing system can output information related to education for healthcare professionals by providing the acquired skill information and target skill information, along with medical information extracted in relation to the skill information and target skill information, to a language model.

[0083] According to Embodiment 1, the information processing system can acquire medical-related schedules for medical professionals and output an educational schedule using a language model that utilizes the acquired medical-related schedules, skill information, and target skill information, along with medical-related information.

[0084] According to Embodiment 1, the information processing system can output an educational schedule by generating prompts that include medical-related schedules, skill information, target skill information, medical-related information, and commands to generate an educational schedule, and by providing the generated prompts to a language model.

[0085] (Embodiment 2) Embodiment 2 describes an example in which an information processing system generates a confirmation test (hereinafter referred to as "confirmation test") for healthcare professionals based on the healthcare professional's skill information, target skill information, medical information, and training schedule.

[0086] The control unit 21 obtains instructions for generating a medical professional ID and a confirmation test (hereinafter referred to as the confirmation test generation instruction) based on the instructions of the medical professional. Specifically, the control unit 21 obtains the medical professional ID and confirmation test generation instruction by displaying a screen (not shown) for obtaining the medical professional ID and confirmation test generation instruction, or by accepting input of the medical professional ID and confirmation test generation instruction via the input unit 24. The control unit 21 transmits the obtained medical professional ID and confirmation test generation instruction to the server 10.

[0087] The control unit 11 receives the medical professional ID and the instruction to generate a confirmation test transmitted from the tablet 20. The control unit 21 reads the skill information, target skill information, medical information, and training schedule corresponding to the medical professional ID from the large-capacity storage unit 14. The control unit 21 generates a prompt 40 (hereinafter referred to as prompt 40) that includes the read skill information, target skill information, medical information, and training schedule, as well as the instruction to generate a confirmation test. The instruction to generate a confirmation test is pre-stored in the storage unit 12 or the large-capacity storage unit 14. The instruction to generate a confirmation test can be modified as appropriate according to the embodiment.

[0088] Figure 9 is an explanatory diagram showing an example of a prompt. The prompt 40 includes a skill information field 31, a target skill information field 32, a medical information field 33, an education schedule field 41, and a second generation command field 42. The contents of the skill information field 31, the target skill information field 32, and the medical information field 33 are the same as in Embodiment 1. The education schedule field 41 contains the education schedule output by the language model M1 (see Figure 7). The second generation command field 42 contains instructions for the language model M1 to output a confirmation test. The contents of the second generation command field 42 are pre-stored in the memory unit 12 or the large-capacity memory unit 14.

[0089] In Figure 9, under the second generation instruction column 42, it states: "Based on the above, as a supervising nurse, please create a confirmation test to check the understanding of healthcare professionals. When creating the confirmation test, please consider the following points: The confirmation test will be administered on the final day of the healthcare professionals' training schedule. The confirmation test should have a total of 20 questions. The time allotted for answering the confirmation test should be 10 minutes."

[0090] The control unit 11 inputs the generated prompt 40 to the language model M1, thereby obtaining the confirmation test generated by the language model M1. The control unit 11 then sends the obtained confirmation test to the tablet 20.

[0091] The control unit 21 receives the confirmation test transmitted from the server 10. The control unit 21 displays the received confirmation test on the display unit 25. Figure 10 is an explanatory diagram showing an example screen. Screen d3 in Figure 10 includes area d31. In area d31, the confirmation test is displayed item by item. The confirmation test displayed in area d31 includes text-format confirmation tests and image-format confirmation tests. Image-format confirmation tests include, for example, image-based confirmation tests to confirm the operation method of medical devices such as ECMO (extracorporeal membrane oxygenation).

[0092] Embodiment 2 describes an example in which the control unit 21 displays a confirmation test, but it is not limited to this. The control unit 11 may also display the combination of the confirmation test and the answers to the confirmation test on the display unit 25. In that case, the second generation command field 42 in Figure 9 will contain the following: "Based on the above, as a supervising nurse, please create a confirmation test to check the understanding of healthcare professionals, and the answers to the confirmation test. When creating the confirmation test and the answers to the confirmation test, please consider the following points. The confirmation test will be administered on the last day of the healthcare professional's training schedule. The total number of questions in the confirmation test should be 20. The time allotted for answering the confirmation test should be 10 minutes."

[0093] Figure 11 is a flowchart showing an example of the processing procedure of the information processing system. The control unit 21 obtains a medical professional ID and a confirmation test generation instruction based on the instructions of the medical professional (step S301). The control unit 21 sends the obtained medical professional ID and confirmation test generation instruction to the server 10 (step S302).

[0094] The control unit 11 receives the medical professional ID and the instruction to generate a confirmation test transmitted from the tablet 20 (step S401). The control unit 21 reads the skill information, target skill information, medical information, and training schedule corresponding to the medical professional ID from the large-capacity storage unit 14 (step S402). The control unit 21 generates the prompt 40 by combining the read skill information, target skill information, medical information, and training schedule with the instruction to generate a confirmation test (step S403).

[0095] The control unit 11 inputs the generated prompt 40 to the language model M1, thereby obtaining the confirmation test generated by the language model M1 (step S404). The control unit 11 then sends the obtained confirmation test to the tablet 20 (step S405).

[0096] The control unit 21 receives the confirmation test sent from the server 10 (step S303). The control unit 21 displays the received confirmation test on the display unit 25 (step S304).

[0097] According to Embodiment 2, the information processing system can output a confirmation test using a language model that utilizes skill information, target skill information, medical information, and an educational schedule.

[0098] According to Embodiment 2, the information processing system can output a confirmation test by generating a prompt that includes skill information, target skill information, medical information, an educational schedule, and a command to generate a confirmation test, and by providing the generated prompt to a language model.

[0099] (Embodiment 3) Embodiment 3 describes an example in which an information processing system acquires video or audio related to a medical procedure and generates a confirmation test or summary information (hereinafter referred to as "summary information") related to the video or audio based on the acquired video or audio.

[0100] Figure 12 is an explanatory diagram illustrating the overview of the second language model. The memory unit 12 stores the second language model M2 in addition to the language model M1. When the second language model M2 receives a prompt that includes video or audio related to a medical procedure and a command (text) to generate a confirmation test, it performs calculations to generate a confirmation test according to the content of the prompt and outputs the generated confirmation test.

[0101] Furthermore, when the second language model M2 receives a prompt that includes video or audio related to medical procedures and a command to generate summary information, it performs calculations to generate summary information according to the content of the prompt and outputs the generated summary information. Video related to medical procedures includes on-the-job training (OJT) related videos, lecture videos, or videos of actual procedures. Audio related to medical procedures includes OJT related audio or lecture audio related to procedures.

[0102] Figure 12 illustrates an example in which an information system inputs a prompt containing video related to a medical procedure and a command to generate a confirmation test into a second language model M2 to obtain a confirmation test. The second language model M2 comprises a first extraction layer C1, a second extraction layer C2, a third extraction layer C3, and a language model M1. The language model M1 is the same as in Embodiment 1. The first extraction layer C1, the second extraction layer C2, and the third extraction layer C3 each comprise a convolutional layer and a pooling layer, respectively.

[0103] The first extraction layer C1 extracts features from the text contained in the prompt. The second extraction layer C2 extracts features from the audio contained in the prompt. The third extraction layer C3 extracts features from the image contained in the prompt. If the prompt contains video, the second language model M2 separates the video into audio and individual images. The second extraction layer C2 extracts features from the separated audio. The third extraction layer C3 extracts features from the separated images.

[0104] The second language model M2 synthesizes a sequence for input to language model M1 by concatenating multiple extracted features. The second language model M2 inputs the synthesized sequence to language model M1. Language model M1 performs calculations to generate a confirmation test from the input sequence and outputs the generated confirmation test. Furthermore, the information system can obtain a confirmation test by inputting a prompt containing audio related to a medical procedure and a command to generate a confirmation test into the second language model M2. In addition, the information system can obtain summary information by inputting a prompt containing video or audio related to a medical procedure and a command to generate summary information into the second language model M2.

[0105] The second language model M2 is not limited to the configuration shown in Figure 12. The second language model M2 may be a multimodal LLM such as GPT-4. In that case, the second language model M2 comprises an input layer, a hidden layer, a fully connected layer, and an output layer. The input layer accepts various input data such as text, video, or audio. The hidden layer extracts features from the various input data such as text, video, or audio. Multiple hidden layers are arranged depending on the type of input data. The fully connected layer combines the features extracted from the multiple hidden layers. The output layer outputs a confirmation test or summary information based on the features combined in the fully connected layer.

[0106] First, the process for obtaining confirmation test results will be explained. Based on instructions from the medical professional, the control unit 21 displays a screen for acquiring video or audio related to the medical procedure on the display unit 25. Figure 13 is an explanatory diagram showing an example of the acquisition screen. The acquisition screen d4 is a screen for receiving video or audio related to the medical procedure and includes a message for the medical professional, an add button b3, a delete button b4, a data field d41, and a send button b5.

[0107] The Add button b3 is used to add video or audio related to medical procedures. When the Add button b3 is selected, the newly added video or audio related to the medical procedure will be displayed in data field d41. In Figure 13, data field d41 displays "Catheterization Room Lecture Video.MP4". If the Add button b3 is not selected, nothing will be displayed in data field d41.

[0108] The delete button b4 is used to delete the video or audio related to the newly added medical procedure. The delete button b4 is positioned corresponding to the video or audio related to the medical procedure displayed in the data field d41. In Figure 13, the delete button b4 corresponding to "Catheterization Room Lecture Video.MP4" is positioned. The send button b5 is used to send the added video or audio related to the medical procedure to the server 10.

[0109] The control unit 21 determines whether or not the selection of the transmit button b5 has been accepted. If the control unit 21 determines that the selection of the transmit button b5 has not been accepted, it accepts the request to add video or audio related to the medical procedure again. If the control unit 21 determines that the selection of the transmit button b5 has been accepted, it sends the video or audio related to the medical procedure to the server 10.

[0110] If the control unit 21 determines that it has received a selection of the send button b5 while no video or audio related to medical procedures has been added to the data field d41, it may display an error message (not shown). In that case, the control unit 21 will continue to display the acquisition screen d4 to accept the addition of video or audio related to medical procedures again.

[0111] The control unit 11 receives video or audio related to a medical procedure transmitted from the tablet 20. The control unit 11 generates a prompt 50a (hereinafter referred to as prompt 50a) which includes the received video or audio related to the medical procedure and a command to generate a confirmation test. The command to generate a confirmation test is pre-stored in the storage unit 12 or the large-capacity storage unit 14. The command to generate a confirmation test can be modified as appropriate according to the embodiment.

[0112] Figure 14 is an explanatory diagram showing an example of a prompt. Prompt 50a includes a video / audio field 51 and a second generation command field 42. The video / audio field 51 contains the video or audio added via the acquisition screen d4. In Figure 14, the video / audio field 51 contains "Catheterization Room Lecture Video.MP4".

[0113] In Figure 14, under the second generation instruction column 42, it states: "Based on the above, as a supervising nurse, please create a confirmation test to check understanding. When creating the confirmation test, please consider the following points: The confirmation test should have a total of 20 questions. The time allotted for answering the confirmation test should be 10 minutes."

[0114] The control unit 11 inputs the generated prompt 50a to the second language model M2. The second language model M2 extracts features from the input prompt 50a using the first extraction layer C1, the second extraction layer C2, and the third extraction layer C3. The second language model M2 synthesizes a sequence for input to the language model M1 by concatenating the extracted features. The second language model M2 outputs a confirmation test by inputting the synthesized sequence to the language model M1. The control unit 11 acquires the confirmation test output by the second language model M2. The control unit 11 transmits the acquired confirmation test to the tablet 20.

[0115] The control unit 21 receives the confirmation test transmitted from the server 10. The control unit 21 displays the received confirmation test on the display unit 25.

[0116] Next, the process for acquiring summary information will be described. Based on the instructions of the medical professional, the control unit 21 displays a screen for acquiring video or audio related to the medical procedure (Figure 13) on the display unit 25. The control unit 21 determines whether or not the selection of the send button b5 has been accepted. If the control unit 21 determines that the selection of the send button b5 has not been accepted, it accepts the addition of video or audio related to the medical procedure again. If the control unit 21 determines that the selection of the send button b5 has been accepted, it sends the video or audio related to the medical procedure added by the add button b3 to the server 10.

[0117] The control unit 11 receives video or audio related to a medical procedure transmitted from the tablet 20. The control unit 11 generates a prompt 50b (hereinafter referred to as prompt 50b) which includes the received video or audio related to the medical procedure and a command to generate summary information. The command to generate summary information is pre-stored in the storage unit 12 or the large-capacity storage unit 14. The command to generate summary information can be modified as appropriate according to the embodiment.

[0118] Figure 15 is an explanatory diagram showing an example of a prompt. Prompt 50b includes a video / audio field 51 and a third generation command field 52. The video / audio field 51 contains the video or audio added via the acquisition screen d4. In Figure 15, the video / audio field 51 contains "Catheterization Room Lecture Video.MP4".

[0119] In Figure 15, under the third generation instruction column 52, it states: "Based on the above, as a supervising nurse, please create a summary of the video or audio related to the medical procedure. When creating the summary, please consider the following points: It should be 400 characters long. Please narrow down the key points to 10 and list them in bullet points."

[0120] The control unit 11 inputs the generated prompt 50b to the second language model M2. The second language model M2 extracts features from the input prompt 50b using the first extraction layer C1, the second extraction layer C2, and the third extraction layer C3. The second language model M2 synthesizes a sequence for input to the second language model M2 by concatenating the extracted features. The second language model M2 outputs summary information by inputting the synthesized sequence to the second language model M2. The control unit 11 acquires the summary information output by the second language model M2. The control unit 11 transmits the acquired summary information to the tablet 20.

[0121] The control unit 21 receives summary information transmitted from the server 10. The control unit 21 displays the received summary information on the display unit 25. Figure 16 is an explanatory diagram showing an example screen. Screen d5 in Figure 16 includes area d51. In area d51, summary information is displayed item by item. The summary information includes key points and summaries of the video related to medical procedures. There are a number of key points specified by prompt 50b and they are written in bullet points. The summary is written with a number of characters specified by prompt 50b. In this embodiment, an example of summary information is shown in which the second language model M2 outputs summary information in text format, but it is not limited to this. The second language model M2 may output summary information in image format or audio format.

[0122] Embodiment 3 describes an example in which the information processing system outputs a confirmation test or summary information by providing a video related to a medical procedure selected by a medical professional to a second language model M2, but it is not limited to this. The information processing system may also output a confirmation test or summary information by providing a video related to medical information or an educational schedule to the second language model M2. The processing in that case will be described below. The large-capacity storage unit 14 stores a video database (not shown). The video database stores combinations of videos related to medical procedures and video titles. The video titles include the original video titles and vectorized video titles.

[0123] The control unit 11 vectorizes the medical information extracted in relation to the skill information and target skill information. The control unit 11 also vectorizes the educational schedule output by the language model M1. The control unit 11 calculates the similarity between the vectorized medical information and the vectorized video titles. The control unit 11 also calculates the similarity between the vectorized educational schedule and the vectorized video titles.

[0124] The control unit 11 identifies vectorized video titles whose calculated similarity is equal to or greater than a predetermined threshold. The control unit 11 extracts videos related to medical procedures corresponding to the identified video titles from the video database. The control unit 11 generates a prompt that includes the extracted videos related to medical procedures and a command to generate a confirmation test. The control unit 11 inputs the generated prompt to the second language model M2 to obtain the confirmation test output by the second language model M2. The control unit 11 may also generate a prompt that includes the extracted videos related to medical procedures and a command to generate summary information. In that case, the control unit 11 can input the generated prompt to the second language model M2 to obtain the summary information output by the second language model M2.

[0125] Figure 17 is a flowchart illustrating an example of the processing procedure of an information processing system. The flowchart shown in Figure 17 is a flowchart of the information processing system outputting a confirmation test. Based on the instructions of the medical professional, the control unit 21 displays the video or audio acquisition screen d4 related to the medical procedure on the display unit 25 (step S501). The control unit 21 determines whether or not the selection of the add button b3 has been accepted (step S502). If the control unit 21 determines that the selection of the add button b3 has been accepted (step S502: YES), it acquires the video or audio related to the newly added medical procedure (step S503).

[0126] The control unit 21 determines whether or not it has received a selection of the transmit button b5 (step S506). If the control unit 21 determines that it has not received a selection of the transmit button b5 (step S506: NO), it returns to step S502. If the control unit 21 determines that it has received a selection of the transmit button b5 (step S506: YES), it transmits video or audio related to the medical procedure to the server 10 (step S507).

[0127] Furthermore, if the control unit 21 determines that it has not accepted the selection of the add button b3 (step S502: NO), it determines whether it has accepted the selection of the delete button b4 (step S504). If the control unit 21 determines that it has not accepted the selection of the delete button b4 (step S504: NO), it returns to step S502. If the control unit 21 determines that it has accepted the selection of the delete button b4 (step S504: YES), it deletes the video or audio related to the medical procedure that is located corresponding to the delete button b4 (step S505). The control unit 11 proceeds to step S506.

[0128] The control unit 11 receives video or audio related to the medical procedure transmitted from the tablet 20 (step S601). The control unit 11 generates a prompt 50a by combining the received video or audio related to the medical procedure with a command to generate a confirmation test (step S602). The control unit 11 inputs the generated prompt 50a to the second language model M2 and obtains the confirmation test output by the second language model M2 (step S603). The control unit 11 transmits the obtained confirmation test to the tablet 20 (step S604).

[0129] The control unit 21 receives the confirmation test sent from the server 10 (step S508). The control unit 21 displays the received confirmation test on the display unit 25 (step S509).

[0130] Figure 18 is a flowchart showing an example of the processing procedure of an information processing system. The flowchart shown in Figure 18 is a flowchart of the process when the information processing system outputs summary information. Furthermore, the flowchart shown in Figure 18 is a modification of the process shown in Figure 17, where the processing from step S601 onwards is changed to steps S701-S703 and steps S801-S802. The explanation of steps that are the same as in Figure 17 is omitted.

[0131] The control unit 11 generates a prompt 50b by combining the received video or audio related to the medical procedure with a command to generate summary information (step S701). The control unit 11 inputs the generated prompt 50b to the second language model M2 and obtains the summary information output by the second language model M2 (step S702). The control unit 11 transmits the obtained summary information to the tablet 20 (step S703).

[0132] The control unit 21 receives summary information transmitted from the server 10 (step S801). The control unit 21 displays the received summary information on the display unit 25 (step S802).

[0133] According to Embodiment 3, the information processing system can acquire video or audio related to medical procedures and output confirmation tests or summary information by providing the acquired video or audio related to medical procedures to a language model.

[0134] According to Embodiment 3, the information processing system can output a confirmation test or summary information by generating a prompt that includes medical-related video or audio, and a command to generate a confirmation test or a command to generate summary information, and by providing the generated prompt to a language model.

[0135] (Embodiment 4) Embodiment 4 describes an example in which an information processing system acquires information on the classification of the medical institution to which a medical professional belongs, the treatment record of the medical institution, and the medical system of the medical institution, and outputs information on education for medical professionals based on the acquired information on classification, treatment record, and medical provision system, as well as medical information.

[0136] Figure 19 is a block diagram showing an example of server configuration. The memory unit 12 stores a second language model M2 instead of language model M1. In Embodiment 4, when information about healthcare professionals and information about medical care is input to the second language model M2, it performs calculations to generate information about education for healthcare professionals and outputs the generated information about education for healthcare professionals. In Embodiment 4, information about healthcare professionals is described as "the classification of the medical institution to which the healthcare professional belongs, the treatment record of the medical institution, and information about the medical care delivery system of the medical institution."

[0137] Furthermore, in Embodiment 4, information regarding education for healthcare professionals is described as a "training plan for the medical institution to which the healthcare professional belongs (hereinafter referred to as the training plan)." The training plan includes a recommended material package, a learning period, a learning schedule, and scenario training.

[0138] The large-capacity storage unit 14 includes a medical institution database 143 and a performance database 144. The medical institution database 143 stores information regarding the classification of medical institutions to which medical professionals belong, the treatment performance of said medical institutions, and the medical care delivery system of said medical institutions. The treatment performance of medical institutions includes annual case numbers and annual number of surgeries, etc. Information regarding the medical care delivery system of medical institutions includes the nursing care system and treatment system of said medical institutions, etc.

[0139] The performance database 144 stores performance data related to education, which is part of the information related to medical care. The performance data related to education in Embodiment 4 further includes information about the medical institution that implemented the training plan and information about the content of the training plan implemented by the medical institution. The information about the medical institution that implemented the training plan includes the name of the medical institution, the classification of the medical institution, the treatment record of the medical institution, and information about the medical care delivery system of the medical institution. The information about the content of the training plan implemented by the medical institution includes the implementation period of the training plan, the educational package used in the training plan, the schedule of the training plan, the guidance system for the training plan, and the results of the training plan implementation.

[0140] Figure 20 is an explanatory diagram showing an example of a record layout in a medical institution database. Medical institution database 143 includes columns for healthcare professional ID, classification, treatment record, and healthcare delivery system. The healthcare professional ID column stores the healthcare professional ID to identify the healthcare professional. The classification column stores the classification of the medical institution to which the healthcare professional belongs. Classifications of medical institutions include, for example, general hospitals, advanced medical care hospitals, regional medical support hospitals, clinical research core hospitals, clinics, and midwifery centers.

[0141] The treatment record column includes the annual number of cases column and the annual number of surgeries column. The annual number of cases column stores the number of cases treated by the medical institution to which the healthcare professional belongs in one year, case by case. The annual number of surgeries column stores the number of surgeries performed by the medical institution to which the healthcare professional belongs in one year, case by case. The medical care delivery system column includes the nursing system column and the treatment system column. The nursing system column stores the nursing system of the medical institution to which the healthcare professional belongs. The nursing system is, for example, the number of primary and secondary healthcare professionals in charge of each subject, or the number of patients each healthcare professional is responsible for. In Embodiment 4, the nursing system is described as "the number of primary and secondary healthcare professionals in charge of each subject."

[0142] The treatment system column stores the treatment system of the medical institution to which the medical professional belongs. The treatment system is, for example, the configuration of equipment and drugs in the operating room (e.g., operating room, catheterization lab, catheterization lab examination room, and angiography room) of the medical institution to which the medical professional belongs. The configuration of equipment and drugs in the operating room is stored as text, or as drawings and images. In Embodiment 4, the treatment system is described as "an image showing the configuration of equipment and drugs in the operating room."

[0143] The record for healthcare professional ID "I001" in Figure 20 contains the following information: classification "General Hospital", annual number of cases "Percutaneous Coronary Intervention (PCI): 250 cases, Peripheral Vascular Therapy (EVT): 223 cases", annual number of surgeries "Cardiovascular Surgery: 218 cases, Neurosurgery: 241 cases", nursing system "Catheterization Lab (3 main staff / 7 part-time staff)", and treatment system "Configuration diagram of equipment in the catheterization lab.JPG".

[0144] Figure 21 is an explanatory diagram showing an example of the record layout of the performance database. The performance database 144 includes columns for facility name, classification, treatment performance, medical care delivery system, implementation period, education package, schedule, training system, and implementation results. The facility name column stores the name of the medical institution that implemented the training plan. The classification column stores the classification of the medical institution that implemented the training plan.

[0145] The treatment results column stores the treatment results of the medical institutions that implemented the training plan, and includes columns for the annual number of cases and the annual number of surgeries. The annual number of cases column stores the annual number of cases for each medical institution that implemented the training plan, item by item. The annual number of surgeries column stores the annual number of surgeries for each medical institution that implemented the training plan, item by item.

[0146] The Healthcare Delivery System column stores the healthcare delivery system of the medical institution where the training plan was implemented, and includes the Nursing System column and the Treatment System column. The Nursing System column stores the nursing system of the medical institution where the training plan was implemented. The Treatment System column stores the healthcare delivery system of the medical institution where the training plan was implemented. The Implementation Period column stores the period during which the training plan was implemented. The Educational Package column stores the educational packages used in the training plan. The Schedule column stores the schedule of the training plan. The Supervision System column stores the supervision system implemented in the training plan. The Implementation Results column stores the implementation results of the training plan. The Performance DB144 may also store each piece of information stored in the Classification column, Treatment Results column, and Healthcare Delivery System column, as well as vectorized versions of each piece of information. In that case, each piece of information and the vectorized versions of each piece of information are stored in correspondence.

[0147] The processing of Embodiment 4 will now be described. Based on the instructions of the medical professional, the control unit 21 obtains a medical professional ID and a training plan generation instruction. Specifically, the control unit 21 obtains a medical professional ID and a training plan generation instruction by displaying a screen (not shown) for obtaining a medical professional ID and a training plan generation instruction, or by inputting a medical professional ID and a training plan generation instruction via the input unit 24. The control unit 21 transmits the obtained medical professional ID and training plan generation instruction to the server 10.

[0148] The control unit 11 receives the medical professional ID and training plan generation instructions transmitted from the tablet 20. The control unit 11 reads from the medical institution DB 143 information regarding the classification of the medical institution to which the medical professional corresponding to the medical professional ID belongs, the treatment record of the medical institution, and the medical care delivery system of the medical institution. Specifically, the control unit 11 reads from the medical institution DB 143 the classification "General Hospital" corresponding to medical professional ID "I001", the annual number of cases "Percutaneous Coronary Intervention (PCI): 250 cases, Peripheral Vascular Therapy (EVT): 233 cases", and the annual number of surgeries "Cardiovascular Surgery: 218 cases, Neurosurgery: 241 cases". The control unit 11 also reads from the medical professional DB 143 the nursing system "Catheterization Room (3 main staff / 7 part-time staff)" and the treatment system "Configuration Diagram of Catheterization Room Equipment, etc..JPG".

[0149] The control unit 11 vectorizes the read classification, treatment record, and medical care delivery system information. The control unit 11 calculates the similarity between the vectorized classification, treatment record, and medical care delivery system information and each vectorized chunk stored in the medical information file 142. For calculating the similarity between vectors, for example, cosine similarity or k-nearest neighbors is used.

[0150] The control unit 11 identifies vectorized chunks whose calculated similarity is equal to or greater than a predetermined threshold. The control unit 11 extracts medical information corresponding to the identified chunks from the medical information file 142. If the control unit 11 extracts multiple pieces of medical information, it may re-rank the multiple pieces of medical information. The method for doing so is the same as in Embodiment 1. By the above method, the control unit 11 obtains medical information extracted in relation to the classification of the medical institution to which the medical professional belongs, treatment results, and information on the medical care delivery system.

[0151] The control unit 11 calculates the similarity for each record between the classification of the medical institution to which the medical professional belongs, the treatment performance of the medical institution, and information regarding the medical care delivery system of the medical institution, and the various information stored in the classification column, treatment performance column, and medical care delivery system column of the performance DB 144. As an example, we will explain an example in which the similarity for each item is calculated between the classification of the medical institution, the treatment performance of the medical institution, and information regarding the medical care delivery system of the medical institution corresponding to the medical professional ID "I001" in the medical institution DB 143, and the classification of the medical institution, the treatment performance of the medical institution, and information regarding the medical care delivery system of the medical institution corresponding to the facility name "X Hospital" in the performance DB 144. Levenshtein distance, cosine similarity, Jacquard coefficient, and sequence matching are used to calculate the similarity.

[0152] When calculating the similarity (hereinafter referred to as classification similarity) between the classification of a medical institution corresponding to the medical professional ID "I001" in Medical Institution DB143 and the classification of a medical institution corresponding to the facility name "X Hospital" in Performance DB144, the classification similarity is indicated by, for example, "0" or "1". If the two classifications are the same, the classification similarity will be "1", and if the two classifications are not the same, the classification similarity will be "0".

[0153] When calculating the similarity (hereinafter referred to as "similarity of treatment results") between the treatment results of the medical institution corresponding to the medical professional ID "I001" in Medical Institution DB143 and the treatment results of the medical institution corresponding to the facility name "X Hospital" in Performance DB144, the similarity of treatment results is shown on a scale of 0 to 1, for example. The more similar the two are, the closer the similarity of treatment results will be to "1," and the less similar the two are, the closer the similarity of treatment results will be to "0."

[0154] When calculating the similarity (hereinafter referred to as the similarity of medical delivery systems) between the information on the medical delivery system of a medical institution corresponding to the medical professional ID "I001" in the medical institution DB143 and the information on the medical delivery system of a medical institution corresponding to the facility name "X Hospital" in the performance DB144, the similarity of medical delivery systems is indicated, for example, on a scale of 0 to 1. The more similar the two are, the closer the similarity of medical delivery systems will be to "1," and the less similar the two are, the closer the similarity of medical delivery systems will be to "0." The control unit 11 calculates the sum of the three types of similarity scores.

[0155] The control unit 11 may also calculate the sum of the three types of similarity using a weighting method. The process in that case will be explained below. The control unit 11 sets several weights (for example, 1 to 3) for the classification similarity, the treatment performance similarity, and the medical care delivery system similarity. Specifically, the control unit 11 sets a weight of "1" for the classification similarity, a weight of "3" for the treatment performance similarity, and a weight of "3" for the medical care delivery system similarity.

[0156] The control unit 11 calculates the weighted similarity by multiplying each similarity by the weight set for each similarity. Specifically, if the similarity of treatment results is "0.8", the control unit 11 calculates the weighted similarity of treatment results as "0.8 × 3 = 2.4". The control unit 11 performs the same process on the similarity of classification and the similarity of the medical care delivery system to calculate the weighted similarity of classification and the weighted similarity of the medical care delivery system. The control unit 11 calculates the sum of the three weighted similarities. Alternatively, the control unit 11 may calculate the average value of the three similarities instead of the sum of the three similarities. Furthermore, the control unit 11 may calculate the average value of the three weighted similarities. In the following explanation, we will use the "sum of the three similarities" from among "sum of the three similarities", "average value of the three similarities", "sum of the three weighted similarities", and "average value of the three weighted similarities".

[0157] The control unit 11 calculates the sum of three types of similarity scores for all records in the performance database 144 by performing the same process for all records (for example, records for Y Hospital). The control unit 11 identifies the record with the largest sum of the three types of similarity scores from the performance database 144. Alternatively, the control unit 11 may identify the top few records (for example, the top three) with the largest sum of the three types of similarity scores from the performance database 144. The control unit 11 extracts the implementation period of the training plan, the educational package used in the training plan, the schedule of the training plan, the guidance system for the training plan, and the results of the training plan stored in the identified record from the performance database 144 as medical information.

[0158] The control unit 11 generates a prompt 60 (hereinafter referred to as "prompt 60") which includes the read classification, treatment record, and medical care provision system information, the extracted medical information, and a training plan generation command. The training plan generation command is pre-stored in the storage unit 12 or the large-capacity storage unit 14. The training plan generation command can be modified as appropriate according to the embodiment.

[0159] Figure 22 is an explanatory diagram showing an example of a prompt. Prompt 60 includes a classification field 61, a treatment performance field 62, a medical care delivery system field 63, a medical information field 64, and a fourth generation command field 65. The classification field 61 contains the classification of the medical institution to which the medical professional belongs. The classification of the medical institution is read from the medical institution DB 143. The content written in the classification field 61 is composed of applying the read classification of the medical institution to a pre-prepared template. In the classification field 61 of Figure 22, it says, "The classification of the medical institution is a general hospital."

[0160] The treatment record column 62 contains the treatment record of the medical institution to which the healthcare professional belongs. The treatment record of the medical institution is read from the medical institution database 143. The content entered in the treatment record column 62 is created by applying the retrieved treatment record of the medical institution to a pre-prepared template. In the treatment record column 62 of Figure 22, it is written that, "The treatment record of the medical institution is as follows. Regarding the number of cases per year, there were 250 cases of percutaneous coronary intervention (PCI) and 223 cases of peripheral vascular treatment (EVT). Regarding the number of surgeries per year, there were 218 cases in cardiovascular surgery and 241 cases in neurosurgery."

[0161] The medical care delivery system section 63 contains information about the medical care delivery system of the medical institution to which the medical professionals belong. This information is retrieved from the medical institution database 143. The content of the medical care delivery system section 63 is created by applying the retrieved information about the medical institution's medical care delivery system to a pre-prepared template. In Figure 22, the medical care delivery system section 63 states, "The nursing system of the medical institution is as follows: There are 3 medical professionals whose primary duty is the catheterization lab. There are 7 medical professionals who also work in the catheterization lab." The medical care delivery system section 63 also contains "The treatment system of the medical institution is as follows," and "Configuration diagram of the catheterization lab equipment, etc..JPG."

[0162] Embodiment 4 describes an example in which "Configuration Diagram of Catheterization Room Equipment, etc..JPG" is included in the medical care provision system section 63, but is not limited to this. The control unit 11 may input "Configuration Diagram of Catheterization Room Equipment, etc..JPG" to the image analysis AI (not shown) before generating the prompt 60. The image analysis AI is stored in the large-capacity storage unit 14. The image analysis AI is, for example, an object detection model such as YOLO (You Only Look Once), and is a trained model that has been trained to output analysis results when an image showing the configuration of equipment and drugs, etc. in an operating room is input. The analysis results include, for example, the types of equipment and drugs, and location information of equipment and drugs, etc. When "Configuration Diagram of Catheterization Room Equipment, etc..JPG" is input, the image analysis AI outputs the analysis results. The control unit 11 includes the analysis results output by the image analysis AI in the medical care provision system section 63.

[0163] The medical information section 64 contains medical information extracted in relation to the classification of the medical institution to which the medical professional belongs, the treatment record of the medical institution, and information regarding the medical care delivery system of the medical institution. The medical information is read from the medical information file 142 and the performance database 144. The medical information section 64 in Figure 22 contains an extraction command that reads, "Refer to the medical information file 142 and the performance database 144 for information related to the information described above." The control unit 11 includes the knowledge data on educational packages and drugs extracted by the extraction process described above in the prompt 60.

[0164] The fourth generation command field 65 contains instructions for the second language model M2 to output a training plan. The contents of the fourth generation command field 65 are pre-stored in the memory unit 12 or the large-capacity memory unit 14. The fourth generation command field 65 in Figure 22 contains the following: "Based on the above, you, as the hospital's education committee chairperson, should present a training plan. When presenting the training plan, please consider the following points: The training plan should include recommended educational packages, recommended duration, and recommended scenario training. If there is no suitable training plan, please explain that fact and the reason."

[0165] The control unit 11 inputs the generated prompt 60 to the second language model M2, thereby acquiring the training plan output by the second language model M2. The control unit 11 then transmits the acquired training plan to the tablet 20.

[0166] The control unit 21 receives the training plan transmitted from the server 10. The control unit 21 displays the received training plan on the display unit 25. Figure 23 is an explanatory diagram showing an example screen. Screen d6 in Figure 23 includes areas d61 and d62. Area d61 displays the components of the training plan, such as the training period, educational package, training content, and guidance system, item by item. Furthermore, if the training period, educational package, training content, and guidance system displayed in area d61 are part of the medical information referenced from the medical information file 142 or the performance DB 144, the relevant training period, educational package, training content, and guidance system are indicated by a double dashed line and a number.

[0167] Area d62 displays hyperlinks to show the sources of information such as training period, education package, training content, and guidance system, which were indicated by double dashed lines and numbers in area d61. When a hyperlink is selected, the control unit 21 sends an instruction to the server 10 to request information about the source medical care. The control unit 11 receives the instruction from the tablet 20 to request information about the source medical care. The control unit 11 reads the information about the source medical care from the medical information file 142 or the performance DB 144. The control unit 11 sends the read information about the source medical care to the tablet 20. The control unit 21 receives the information about the source medical care sent from the server 10. The control unit 21 overlays the information about the source medical care onto screen d6. The control unit 21 may split the display unit 25 to show the information about the source medical care and screen d6 separately.

[0168] Embodiment 4 describes an example in which the information processing system outputs a training plan by providing the second language model M2 with information regarding the classification of the medical institution to which the medical professional belongs, treatment results, and the medical care delivery system, as well as information regarding medical care. However, the system is not limited to this. In addition to information regarding the classification of the medical institution, treatment results, and the medical care delivery system, the information processing system may also output a training plan by providing the second language model M2 with skill information, target skill information, and medical-related schedules.

[0169] The following describes the process in that case. The control unit 11 receives instructions from the tablet 20 to generate the medical professional ID, target skill information (individual goals), medical schedule, and training plan. The control unit 11 reads the classification of the medical professional corresponding to the medical professional ID, the treatment record of the medical institution, and information on the medical institution's medical care delivery system from the medical institution DB 143. The control unit 11 also reads the skill information and target skill information (common goals) corresponding to the medical professional ID from the medical professional DB 141.

[0170] The control unit 11 vectorizes the classification of medical institutions, treatment performance of those medical institutions, information regarding the medical care delivery system of those medical institutions, skill information, and target skill information obtained through reading, etc. The control unit 11 calculates the similarity between the vectorized classification, treatment performance, information regarding the medical care delivery system, skill information, and target skill information and each vectorized chunk stored in the medical information file 142.

[0171] The control unit 11 identifies vectorized chunks whose calculated similarity is equal to or greater than a predetermined threshold. The control unit 11 extracts medical information corresponding to the identified chunks from the medical information file 142. If the control unit 11 extracts multiple pieces of medical information, it may re-rank the multiple pieces of medical information.

[0172] The control unit 11 compares the classification of the medical institution to which the medical professional belongs, the treatment performance of the medical institution, and the medical care delivery system of the medical institution with the various information stored in the classification column, treatment performance column, and medical care delivery system column of the performance DB 144 to calculate the classification similarity, treatment performance similarity, and medical care delivery system similarity for each record. The control unit 11 calculates the sum of the above three types of similarity. The control unit 11 performs the same process for all records in the performance DB 144 to calculate the sum of the three types of similarity for all records. The control unit 11 identifies the record with the largest sum of the calculated three types of similarity from the performance DB 144. The control unit 11 extracts the implementation period of the training plan, the educational package used in the training plan, the schedule of the training plan, the guidance system of the training plan, and the implementation results of the training plan stored in the identified record from the performance DB 144 as medical information.

[0173] The control unit 11 generates a prompt that includes classification, treatment results, information on the medical care delivery system, skill information, target skill information, and medical-related schedules acquired through reading, etc., as well as medical information extracted from the medical information file 142 and a command to generate a training plan. The control unit 11 inputs the generated prompt to the second language model M2, thereby acquiring the training plan output by the second language model M2.

[0174] Figure 24 is a flowchart showing an example of the processing procedure of the information processing system. The control unit 21 obtains a medical professional ID and a training plan generation instruction based on the instructions of the medical professional (step S901). The control unit 21 sends the obtained medical professional ID and training plan generation instruction to the server 10 (step S902).

[0175] The control unit 11 receives the medical professional ID and training plan generation instruction transmitted from the tablet 20 (step S1001). The control unit 11 reads the classification of the medical professional corresponding to the medical professional ID, the treatment record of the medical institution, and information on the medical institution's medical care delivery system from the medical institution DB 143 (step S1002). The control unit 11 vectorizes the read classification, treatment record, and medical care delivery system information (step S1003). The control unit 11 calculates the similarity between the vectorized classification, treatment record, and medical care delivery system information and each vectorized chunk stored in the medical information file 142 (step S1004).

[0176] The control unit 11 identifies vectorized chunks whose calculated similarity is equal to or greater than a predetermined threshold (step S1005). The control unit 11 extracts medical information corresponding to the identified chunks from the medical information file 142 (step S1006). The control unit 11 calculates the similarity for each record between the classification of the medical institution to which the medical professional belongs, the treatment record of the medical institution, and the medical care delivery system of the medical institution, and the various information stored in the classification column, treatment record column, and medical care delivery system column of the record DB 144 (step S1007). The control unit 11 calculates the sum of the three types of similarity (step S1008).

[0177] The control unit 11 determines whether or not it has calculated the sum of the three types of similarity scores for all records in the actual database 144 (step S1009). If the control unit 11 determines that it has not calculated the sum of the three types of similarity scores for all records in the actual database 144 (step S1009: NO), it returns to step S1007. If the control unit 11 determines that it has calculated the sum of the three types of similarity scores for all records in the actual database 144 (step S1009: YES), it identifies the record with the largest sum of the three types of similarity scores from the actual database 144 (step S1010).

[0178] The control unit 11 extracts the implementation period of the training plan, the educational package used in the training plan, the schedule of the training plan, the guidance system for the training plan, and the implementation results of the training plan from the performance DB 144 as medical information (step S1011). The control unit 11 generates a prompt 60 by combining the read classification, treatment performance and medical care provision system information, the extracted medical information, and the training plan generation command (step S1012). The control unit 11 inputs the generated prompt 60 to the second language model M2 and obtains the training plan output by the second language model M2 (step S1013). The control unit 11 transmits the obtained training plan to the tablet 20 (step S1014).

[0179] The control unit 21 receives the training plan transmitted from the server 10 (step S903). The control unit 21 displays the received training plan on the display unit 25 (step S904).

[0180] According to Embodiment 4, the information processing system acquires information on the classification, treatment record, and medical care delivery system of the medical institution to which the medical professional belongs, and can output a training plan using a language model that utilizes the acquired information on classification, treatment record, and medical care delivery system, as well as information on medical care.

[0181] According to Embodiment 4, the information processing system can output a training plan by providing a language model with the acquired information on classification, treatment results, and medical care delivery system, as well as medical information extracted in relation to the information on classification, treatment results, and medical care delivery system.

[0182] According to Embodiment 4, the information processing system can output information regarding education for the medical professionals by generating prompts that include information on the classification of the medical institution to which the medical professionals belong, treatment results, and the medical care delivery system, as well as a command to generate a training plan, and by providing the generated prompts to a language model.

[0183] The matters described in each of the embodiments described above can be combined with one another. Furthermore, the independent claims and dependent claims described in the claims can be combined with one another in any combination, regardless of the form of reference. Moreover, although the claims use a form in which claims referencing two or more other claims (multi-claim form), the claims are not limited to this. A form in which multi-claims referencing at least one multi-claim (multi-multi-claim) may also be used.

[0184] The embodiments disclosed herein should be considered in all respects to be illustrative and not restrictive. The scope of the present invention is indicated by the claims, not in the sense described above, and all modifications within the sense and scope equivalent to the claims are intended. [Explanation of Symbols]

[0185] 10. Information Processing Equipment (Server) 11 Control Unit 12 Storage section 12P control program M1 Language Model M2 Second Language Model C1 1st extraction layer C2 2nd extraction layer C3 3rd extraction layer 13 Communications Department 14 Mass storage 141 Healthcare Professionals Database 142 Medical Information File 143 Medical Institution Database 144 Performance Database 15 Reading section 20. Information Processing Devices (Tablets) 21 Control Unit 22 Memory section 22P Control Program 23 Communications Department 24 Input section 25 Display section 30 prompts 31 Skill Information Section 32. Target Skill Information Section 33. Medical Information Section 34 Schedule section 35 Generated instruction field 40 Prompts 41 Education Schedule Section 42 2nd generated instruction column 50a Prompt 50b Prompt 51 Video / Audio Section 52 3rd generation instruction column 60 prompts 61 Classification column 62 Treatment Results Section 63. Medical Care Delivery System Section 64. Medical Information Section 65 4th generated instruction column d1 Input screen d11 ID field d12 Personal goal column d13 Schedule section d4 acquisition screen d41 Data section b1 Registration button b2 Add Schedule button b3 Add button b4 Delete button b5 Send button 1a Portable storage medium

Claims

1. Obtain information about healthcare workers, Using a language model that utilizes acquired information about healthcare professionals and information about healthcare, information regarding education for said healthcare professionals is output. A program that instructs a computer to perform a process.

2. The skills information of the aforementioned medical professionals and the target skill information are acquired. A language model that uses acquired skill information, target skill information, and medical information outputs information regarding education for the aforementioned medical professionals. The program according to claim 1.

3. Medical information includes knowledge data on drugs, knowledge data on complications, knowledge data on medical procedures, knowledge data on medical devices, medical glossaries, manuals on medical devices, or performance data on education. By providing the language model with the acquired skill information and target skill information, and the medical information extracted in relation to the skill information and target skill information, information regarding education for the medical professionals is output. The program according to claim 2.

4. Obtain the medical schedules of the aforementioned medical professionals, Using a language model that utilizes acquired medical-related schedules, skill information, and target skill information, along with medical-related information, an educational schedule for the aforementioned medical professionals is output. The program according to claim 2 or 3.

5. The system generates prompts including the medical schedule, the skill information, the target skill information, the medical information, and the instruction to generate the educational schedule. By providing the generated prompt to the language model, the aforementioned educational schedule is output. The program according to claim 4.

6. A language model using the aforementioned medical professional's skill information, target skill information, medical information, and the medical professional's training schedule outputs a confirmation test for the medical professional. The program according to claim 4.

7. A prompt is generated that includes the aforementioned skill information, the aforementioned target skill information, the aforementioned medical information, the aforementioned educational schedule, and the command to generate the aforementioned confirmation test. By providing the generated prompt to the language model, the confirmation test is output. The program according to claim 6.

8. We obtain video or audio related to medical procedures. By providing the acquired video or audio related to the medical procedure to the language model, a confirmation test for the medical professional or summary information related to the video or audio is output. The program according to claim 1.

9. The system generates a prompt including the aforementioned medical video or audio, and the command to generate the confirmation test or the command to generate the summary information. By providing the generated prompt to the language model, the confirmation test or the summary information is output. The program according to claim 8.

10. The classification of the medical institution to which the aforementioned medical professional belongs, the treatment record of the said medical institution, and information regarding the medical care delivery system of the said medical institution are obtained. Using the acquired classification, treatment results, and information on the healthcare delivery system, along with information on healthcare, the language model outputs information on education for healthcare professionals. The program according to claim 1.

11. Information regarding the medical care delivery system of a medical institution includes the nursing system or the treatment system of the said medical institution. The program according to claim 10.

12. Medical information includes knowledge data on drugs, knowledge data on complications, knowledge data on medical procedures, knowledge data on medical devices, medical glossaries, manuals on medical devices, or performance data on education. By providing the language model with the acquired information on classification, treatment results, and healthcare delivery systems, and the healthcare information extracted in relation to the classification, treatment results, and healthcare delivery systems, information on education for healthcare professionals is output. The program according to claim 10.

13. A prompt is generated that includes a command to generate the aforementioned classification, the aforementioned treatment results, the aforementioned medical care delivery system information, the aforementioned medical care information, and the aforementioned education information for medical professionals. By providing the generated prompts to the language model, information regarding education for the healthcare professionals is output. The program according to any one of claims 10 to 12.

14. Obtain information about healthcare workers, Using a language model that utilizes acquired information on healthcare professionals and medical information, information regarding education for said healthcare professionals is output. Information processing methods.

15. An information processing device having a control unit, The control unit, Obtain information about healthcare workers, Using a language model that utilizes acquired information on healthcare professionals and medical information, information regarding education for said healthcare professionals is output. Information processing device.

Images