Methods for formulating health, medical, and long-term care plans

Abstract

This provides a method for formulating health, medical, and long-term care plans that facilitates policy-making and other related processes. [Solution] The plan formulation method accepts multiple issue-solution datasets, which are datasets of existing issue texts and existing solution texts describing existing issues and existing solutions related to health, medical care, and nursing care in natural language. From these, it extracts issue-solution datasets corresponding to the classified types of target issues. Using a generation AI, it reuses the issue-solution datasets and numerical data or prompts containing instructions given to the generation AI during the processing process as needed to create and output a plan in which the issues of the creator and the solutions to those issues are described in natural language.

Description

Technical Field

[0001] The present invention relates to a method for formulating a plan for healthcare and nursing care.

Background Art

[0002] In response to social issues such as the increase in medical costs associated with the declining birthrate and aging population, the shortage of medical and nursing care personnel, and lifestyle-related diseases, there is a strong demand for improving the QoL (Quality of Life) of inpatients and their families in hospitals and nursing care facilities, and improving the QoW (Quality of Work) by enhancing the work efficiency of facility employees and nursing care workers and enriching the workplace environment to reduce the turnover rate. In addition, with the increasing health awareness among healthy people including the younger generation, the well-being and health-related markets have achieved remarkable growth, and the movement for individuals to actively manage and improve their health, happiness, and QoL has accelerated.

[0003] In response to such trends in the social healthcare and nursing care fields, the inventors of the present invention comprehensively analyze and consider the optimal allocation of human resources such as medical personnel, physical resources such as medical facilities, medical equipment, and pharmaceuticals, financial resources, and information resources across society, in order to provide suggestions useful for medical care on-site in hospitals and nursing care facilities, and for formulating medical and nursing care policies at the national and local government levels.

[0004] As a method for conducting such analysis and consideration, for example, Patent Document 1 discloses a method of estimating the severity of patients by analyzing the medical treatment information of patients in a medical institution and determining the number and allocation of nurses required for that medical institution. However, since this method uses the medical treatment information of patients in a specific medical institution as an information resource, it cannot comprehensively analyze and consider the optimal allocation of human and physical resources across society.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

[0006] The inventors handle a wide variety of data, such as that shown in Figure 1. As shown in Figure 1, not only are the types of data diverse, but the sources from which the data is obtained are also diverse, so the inventors refer to this data as "multi-source, multi-dimensional data." By analyzing this multi-source, multi-dimensional data, it becomes possible to understand the current situation, analyze challenges, and formulate policies in the fields of health, medical care, and nursing care. This, in turn, enables proactive management of comprehensive well-being improvements such as individual health, happiness, and quality of life, as well as the formulation and implementation support of medium- to long-term policies that encompass government, businesses, and individuals.

[0007] However, analyzing multi-source, multi-variable data and using the results to understand the current situation, analyze challenges, and formulate policies in the fields of health, medical care, and nursing care requires advanced analytical techniques that combine multi-source, multi-variable data in complex ways, based on specialized knowledge. Even for experts, this is not easy.

[0008] The problem that this invention aims to solve is to facilitate the current state of human resources, material resources, financial information, and information resources, as well as the analysis of challenges and the formulation of policies in the fields of health, medical care, and nursing care. [Means for solving the problem]

[0009] One aspect of the present invention, made to solve the above problems, is a method for formulating a health, medical care, and nursing care plan, Using a computer, The first reception step receives current status description data that describes the current situation of the plan creator regarding at least one of health, medical, and care aspects, Based on the current situation description data received in the first reception step, the issue setting step sets target issues related to the health, medical care, or nursing care of the creator, A second reception step in which the creation entity receives numerical data related to the target issue set in the issue setting step, A statistical calculation step which calculates a statistical quantity relating to the top group of the entities that created the numerical data received in the second reception step, A target issue classification step involves analyzing the statistics calculated in the aforementioned statistics calculation step and classifying the target issue into one of several types, A third acceptance step accepts multiple problem-and-solution datasets, which are datasets of existing problem statements and existing solution statements, each describing an existing problem related to at least one of health, medical care, and nursing care, and an existing solution to that problem, in natural language. An extraction step in which, from among the multiple problem countermeasure datasets received in the third reception step, a problem countermeasure dataset corresponding to the type of target problem classified in the target problem classification step is extracted, A planning step that uses a generative AI to create a plan in which text describing the issues of the creator and the countermeasures for those issues is written in natural language, using the issue countermeasures dataset and numerical data extracted in the extraction step, A knowledge storage step involves saving, as knowledge data, a problem-solving dataset created in the aforementioned planning step and a pair of prompts containing the instructions given to the generating AI. An output step that outputs the plan created in the aforementioned planning step, This is what it does.

[0010] In the above method for formulating health, medical, and nursing care plans, Using the aforementioned computer, A further management step may be performed to manage the current status description data received in the first reception step and the numerical data received in the second reception step based on a thesaurus related to health, medical care, and nursing care. Here, "managing" includes hierarchically dividing and storing current descriptive data and numerical data in pairs with prompts containing instructions for the generating AI, based on a thesaurus related to health, medical care, and nursing care, the meaning of each data item, and the dependencies between the data items.

[0011] The method for formulating health, medical, and nursing care plans according to the present invention is applicable not only to plans by the national government, local governments, medical service areas, schools, companies, medical facilities, nursing care facilities, and health insurance associations, but also to plans for individuals such as residents of the national government, local governments, and medical service areas, students of schools, employees of companies, users of medical and nursing care facilities, members of health insurance associations, and users of health management apps. In other words, the "organizer of the plan" in this invention refers to an organization such as the national government, local governments, medical service areas, schools, companies, medical facilities, nursing care facilities, and health insurance associations, or an individual.

[0012] Furthermore, the "higher-level group of the creator" is, for example, a region such as Asia, Europe, or Oceania, to which a country and several other countries belong, if the creator is a country; or, for example, a company, to which the company belongs, the industry association to which the company belongs, if the creator is an individual; and if the creator is an individual, the "higher-level group of the creator" is the country or local government, school, company, medical facility, nursing care facility, etc. to which the individual belongs.

[0013] "Current situation description data" can be any data that allows the creator to understand the current situation regarding at least one of the following: health, medical care, and nursing care. This includes data from descriptive materials created by people (workers), such as text, graphs, tables, comics, and drawings.

[0014] The types of issues to be addressed include whether the issue relates to human resources, physical resources, financial resources, or information resources in the health, medical care, or nursing care field; the length of time required to resolve the issue (short-term, medium-term, long-term, etc.); whether the cost required to resolve the issue is high or low; and the urgency and importance of the issue.

[0015] The aforementioned numerical data includes numerical data representing the performance, activities, quantity, type, and geographical distribution of human resources, material resources, and financial resources in the health, medical care, and nursing care fields of the creating entity, as well as numerical data representing demographic and socioeconomic environmental data. Furthermore, if the creating entity is an individual, it includes numerical data representing that individual's economic and health status, and if the creating entity is one of the aforementioned organizations, it includes numerical data representing the economic and health status of individuals belonging to that organization, as well as vital information and well-being information that can grasp the mental health status of individuals. The statistics relating to the top group of entities that create the aforementioned numerical data are typically standard scores or percentiles, but for example, the difference or ratio between the entity's numerical value and the mean, minimum, or maximum value of the top group are also included in the statistics.

[0016] Furthermore, one aspect of the present invention is a computer-based system for formulating health, medical, and nursing care plans, A first storage unit containing current status description data that describes the current status of the plan creator regarding at least one of health, medical, and care, A second memory unit storing numerical data related to the health, medical, and nursing care fields of the above-mentioned creation entity and its superior group, A task setting unit sets target issues related to the health, medical care, or nursing care of the creator based on the current situation description data stored in the first memory unit, A statistical calculation unit obtains numerical data of the top group of creators related to the target task set by the task setting unit from the second storage unit and calculates statistical quantities relating to the top group of the numerical data of the creators, A target issue classification unit analyzes the statistics calculated by the statistics calculation unit and classifies the target issue into one of several types, A third memory unit stores a dataset of existing problem statements and existing countermeasure statements, which are datasets of existing problem statements and existing countermeasure statements, in which one or more existing problems related to at least one of health, medical care, and nursing care, and existing countermeasures for those problems are described in natural language. A planning unit obtains a problem countermeasure dataset from the third memory unit corresponding to the type of target problem classified by the target problem classification unit, and uses a generating AI to create a plan in which the text concerning the problem of the creator and the countermeasures for that problem is written in natural language, using the problem countermeasure dataset corresponding to the type of target problem and the numerical data. A knowledge storage unit stores the problem-solving dataset created by the aforementioned planning unit and a pair of prompts containing the instructions given to the generating AI as knowledge data. An output unit that outputs the plan created by the aforementioned planning unit and It is provided with

Advantages of the Invention

[0017] In the present invention, based on the current situation description, the problems (target problems) of the plan-making entity are set, and the statistical quantities in the upper population of the creation entity of the numerical data representing at least one of the health, medical care, and nursing care of the creation entity related to the target problems are obtained. The statistical quantities are analyzed to classify the types of the target problems. When the types of the target problems are classified, based on a dataset (problem-countermeasure dataset) including existing problem texts and existing countermeasure texts in natural language that describe existing problems and their existing countermeasures corresponding to the types, and prompt data including the content of the instructions given to the generative AI, a plan is created in natural language that describes the problems of the creation entity and the countermeasures for those problems, and the plan is output. Since this plan is generated using a generative AI, a learned model that has learned the patterns and features using a dataset of a large number of existing problem texts and existing countermeasure texts is stored in the knowledge storage unit, and by using the problem-countermeasure dataset as the learning data of the generative AI again, it is possible to easily create a high-quality plan while growing knowledge even without expertise.

Brief Description of the Drawings

[0018] [Figure 1] A diagram showing an example of multi-source and multi-dimensional data. [Figure 2] A block configuration diagram of an example of a health care, medical care, and nursing care plan-making system according to an embodiment of the present invention. [Figure 3] A flowchart showing the operations and processing procedures in a health care, medical care, and nursing care plan-making method using the health care, medical care, and nursing care plan-making system of this embodiment. [Figure 4] A diagram for explaining an example of the flow of the creation work of a provisional plan and a final plan. [Figure 5] An example of a table showing the correspondence between the content of the target problem text included in the plan and the classification of the problems. [Figure 6] An example of a table showing the correspondence between the content of the target countermeasure text included in the plan and the classification of the countermeasures. [Figure 7]An example of a hierarchical or systematic data management method between different data sets. [Figure 8] An example of data needed from a healthcare, medical, and long-term care perspective when generative AI predicts future plans and future social transformations. [Figure 9] An example of data trends (from 2000 to 2050) predicted by a generating AI from the perspective of health, medical care, and nursing care, using the data shown in Figure 8. [Figure 10] A diagram illustrating the configuration of a health, medical, and nursing care planning system according to one embodiment. [Figure 11] A diagram showing the input / output data and processing flow of a health, medical, and nursing care planning system according to one embodiment. [Modes for carrying out the invention]

[0019] Recent advancements in AI (Artificial Intelligence) technology have been remarkable. Generative AI, a branch of machine learning, can generate new documents, images, and audio data from user-input patterns and features by pre-learning the correspondence between various data formats such as documents, images, and audio, and the patterns and features that data represents. Here, "generative AI" refers to artificial intelligence that generates various types of text based on input data, constraints, or instructions. Currently, widely known text generation AIs (also called generative AI) include "ChatGPT" provided by OpenAI, "Copilot" provided by Microsoft, "Gemini" provided by Google, and "Llama" provided by Meta, but are not limited to these.

[0020] Generative AI includes a technology called conversational AI, which is specialized to enable natural conversations with humans. Among these, services that apply large-scale language models (LLMs), which are trained using a large amount of data of various types, are increasingly being offered. Because LLMs are designed to understand and generate text written in natural language, by reinforcement learning on content related to health, medical care, and nursing care, and by fine-tuning the parameters, it becomes possible to create plans or analyze numerical data using LLMs or methods that support decision-making based on unique data from RAG (Retrieval Augmented Generation), which previously required considerable time and effort from experts in the health, medical, and nursing care fields. This is expected to improve the efficiency of plan creation and numerical data analysis. The present invention relates to a method and system that utilizes such generative AI to support plan creation in the health, medical, and nursing care fields.

[0021] In this invention, "plan" refers, for example, to a plan for formulating and implementing policies, measures, and systems related to the health, medical care, and nursing care fields of the national or local government, but may also include management plans for specific health, medical care, and nursing care facilities, associations, or organizations, as well as plans for promoting individual health or improving their lifestyles.

[0022] Hereinafter, an embodiment of the planning support method and system according to the present invention will be described in detail with reference to the drawings.

[0023] [System Configuration] Figure 2 is a schematic block diagram of the planning system of this embodiment. As shown in Figure 2, the planning system of this embodiment includes a terminal 1 connected to the Internet or intranet 2 (hereinafter referred to as "Internet 2" for convenience), a multi-source multiple database (DB) 3, and a generation AI system 4. The generation AI system 4 is usually built on a single large-scale server connected via the Internet, but some of its functions may be distributed to multiple medium-sized servers, or it may be built on a small-scale server on a personal computer (PC) and specialized for unique functions. Alternatively, it is possible to coordinate multiple generation AI systems depending on the application (data volume and analysis function).

[0024] Terminal 1 is an edge computer (EC) such as a PC or smartphone. Software (computer programs) installed on the PC or EC runs on the PC or EC, and in some cases, it also utilizes functions provided by a system (not shown) on the Internet 2 to execute the processing in each functional block described later.

[0025] The multi-source, multi-format DB3 is a database that stores a wide variety of data, including statistical data on health, medical care, and population / socioeconomics, individual health, medical care, and nursing care data, data including geospatial information, and diverse needs such as policies and management plans for customer issues. It may include not only generally available databases (systems) that store demographic data and survey results of health, medical care, and nursing care facilities provided by the national and local governments, but also various databases provided by private companies. The multi-source, multi-format DB3 is not limited to one database, but may consist of multiple databases. In other words, in this embodiment, "multi-source, multi-format data" refers to a wide variety of data related to health, medical care, and nursing care. Furthermore, as shown in Figure 10, the multi-source, multi-format DB3 stores a large dataset of optimal problem statements and countermeasure statements generated by the generation AI, as well as instructions (prompts) given to the generation AI, as knowledge in the knowledge memory unit, and automatically strengthens and grows the knowledge data by using it again (feedback) as learning data for the generation AI.

[0026] Figure 1 illustrates an example of multi-source, multi-format data. For example, in Figure 1, administrative statistics data, health, medical care, and long-term care data, health and lifestyle-related data, various government statistical survey data, and various publicly available data are included in statistical data related to health, medical care, long-term care, and population / socioeconomics. Also in Figure 1, hospital DPC data, health checkup data, medical and long-term care claims data, and related data are included in individual health, medical care, and long-term care data, while transportation data, natural environment and earth observation data, various government statistical survey data, and various publicly available data are included in data containing geospatial information. Furthermore, the needs of national and local government administrative planning support, organizational (hospitals / nursing care facilities / companies) support needs, and individual well-being support needs are included in the diverse needs such as policies and management plans for customer challenges.

[0027] Generative AI System 4 may be the whole or part of a system built by a provider of generative AI services as exemplified above. The generative AI used here is generally called text generation AI. Text generation AI is a model that uses a large amount of web text data existing on the internet, a pre-trained proprietary language model (LLM), or a method of supporting decision-making based on proprietary data from RAG (Retrieval Augmented Generation) as training data, and completes sentences by inferring the word that follows the input word. ChatGPT, one of the well-known GPT (Generative Pre-trained Transformer) algorithms, is a model tuned to output appropriate responses to text, audio, and images input by the user.

[0028] Terminal 1 comprises, as functional blocks, a data input receiving unit 10, a problem (needs) setting unit 11, a statistical calculation unit 12, a problem countermeasure data set extraction unit 13, a target problem classification unit 14, a plan creation unit 15, and a storage unit 16. The storage unit 16 includes a current situation description data storage unit 160, a numerical data storage unit 161, a problem countermeasure data storage unit 162, a provisional plan storage unit 163, and the like. The functions or processing of each functional block are typically realized by the generation AI system 4 described above, but may also be realized by machine learning models or generation AI built into Terminal 1.

[0029] Terminal 1 is connected to an input unit 17 and an output unit 18, which serve as user interfaces. The input unit 17 includes, for example, hardware keys such as a touch panel and keyboard, a pointing device such as a mouse, a camera (image input), and a microphone (voice input). The output unit 18 includes, for example, a display, a touch panel, a speaker, etc. In the following description, the output unit 18 will be described as a display.

[0030] [Overview of the planning process] Next, using the system described above, an example of the procedure for formulating and processing a regional medical care plan for a certain secondary medical area will be explained using the flowchart in Figure 3. In this example, "a certain secondary medical area" is the entity responsible for creating the plan.

[0031] A secondary medical area refers to a regional area established by a prefecture in accordance with the Medical Care Act, a Japanese law, in order to develop a medical care delivery system.

[0032] First, the user (the person responsible for creating the plan at the creation entity) operates the input unit 17 to input current situation description data that describes the current situation regarding at least one of the health, medical, and nursing care services (hereinafter referred to as "health, medical care, and nursing care") in the secondary medical area, and this current situation description data is received by the data input reception unit 10 (Step 1). The current situation description data is typically text data or audio data in which the current health, medical care, and nursing care services are described in natural language, but it may also be graphs, tables, comics, or pictures, or data consisting of a combination of two or more of the text, graphs, tables, comics, and pictures.

[0033] The current situation description data includes the current situation, evaluations, types of issues, content of evaluations and issues, types of countermeasures, and content of countermeasures for the health, medical care, and long-term care fields in any of the prefectures that comprise the secondary medical care areas that are created by the data-creating entity, as well as for each secondary medical care area within that prefecture. The current situation description data is data that can reinforce evaluations, issue setting, and countermeasure formulation for the prefecture as a whole and for each secondary medical care area.

[0034] The current situation description data can be entered by the user directly by operating the input unit 17, by having the user load pre-prepared text data into the terminal 1, or by accessing the multi-source / multi-database DB3 or other databases via the internet 2 and obtaining the data from there. The data input reception unit 10 stores the received current situation description data for the secondary medical area in the current situation description data storage unit 160.

[0035] Next, when the user operates the input unit 17 to instruct the extraction of issues related to health care and nursing in the secondary medical area, the issue setting unit 11 reads the current situation description data from the current situation description data storage unit 160, extracts issues related to health care and nursing, and sets those issues as target issues (Step 2). Issue extraction is performed, for example, by understanding the content of the current situation description and searching for characteristic words included in the current situation description, such as words representing human resources, physical resources, and information resources related to health care and nursing (doctors, caregivers, hospitals, nursing facilities, hospital beds, medicines, etc.), and words representing the qualitative and quantitative characteristics of human resources, physical resources, and information resources (shortages, uneven distribution, aging, etc.). For example, potential issues related to healthcare include "doctor shortage," "doctors' working styles and environment," "doctors' career development," "shortage of medical facilities," and "access to medical facilities." The issue setting unit 11 sequentially displays the set target issues on the output unit 18.

[0036] Next, when the user performs a predetermined operation on the input unit 17, the statistics calculation unit 13 accesses the multi-source / multi-database DB3 via the Internet 2 and retrieves numerical data related to the target issue set in step 2 from among the many numerical data stored in the multi-source / multi-database DB3 (step 3). Numerical data refers to data that visualizes and evaluates the characteristics of any prefecture that includes the secondary medical area that is the creating entity, as well as the characteristics of each secondary medical area within that prefecture, such as the number of doctors, the number of medical and nursing care facilities, the number of medical professionals, population, and income. This data is expressed in text, numbers, and charts as evaluations such as whether it is relatively high or low based on comparisons with the whole country or other secondary medical areas within the prefecture. Numerical data is data that can reinforce evaluations of the entire prefecture and each secondary medical area, as well as the setting of issues and the formulation of countermeasures. Here, an example of retrieving numerical data from the multi-source / multi-database DB3 is shown, but the user may input numerical data prepared in advance by loading it into the terminal 1, or they may access the database via an intranet or the like and retrieve it from there.

[0037] Here, the numerous numerical data stored in the multi-source, multi-variable DB3 are linked to at least information that identifies the source of the numerical data, such as the country or prefecture, and information that identifies the type of numerical data. Furthermore, it is desirable that the numerous numerical data be linked to information that allows for hierarchical and systematic linking and management.

[0038] For example, if the designated target issue is "shortage of doctors," numerical data on the number of doctors in all secondary medical areas included in the prefecture to which the aforementioned secondary medical area belongs is acquired and stored in the numerical data storage unit 161. Alternatively, instead of acquiring numerical data for secondary medical areas included in prefectures, numerical data for secondary medical areas nationwide may be acquired, or numerical data for secondary medical areas in areas between prefectures and the national government (for example, the Kinki region, the Chugoku region, or the Hokuriku region) may be acquired. In other words, prefectures, the national government, and the Kinki region, etc., are higher-level groups relative to the secondary medical areas, which are the entities that create the data.

[0039] Next, the statistical calculation unit 13 statistically processes the acquired numerical data and calculates statistics (e.g., standard scores) for the top group of physicians in the secondary medical area that is the source of the data (Step 4). Then, it analyzes the calculated statistics and classifies the type of target issue (Step 5). For example, if the standard score of the number of physicians in the secondary medical area is within the range of 40-50, it is determined that there is no shortage of physicians in that secondary medical area and it is designated as a "potential physician shortage area." If the standard score of the number of physicians exceeds 50, it is determined that there is a sufficient number of physicians and it is designated as "no physician shortage problem." If the standard score of the number of physicians is less than 40, it is determined that there is a shortage of physicians and it is designated as "physician shortage identified." However, these values ​​are hypothetical, and the numerical values ​​(thresholds) for determining whether or not there is a "physician shortage" can be set as appropriate.

[0040] Next, the user performs a predetermined operation on the input unit 17 and accesses the multi-source / multi-dimensional DB3 via the internet 2 through the problem-solving dataset extraction unit 13. From the numerous problem-solving datasets stored in the multi-source / multi-dimensional DB3, the user obtains one or more problem-solving datasets that match the classification of the target problem certified in step 5 (step 6). A problem-solving dataset refers to a dataset of texts (existing problem texts and existing solution texts) that describe problems and solutions to those problems in natural language, which were included in plans created in the past for the formulation of policies related to health, medical care, and nursing care. The series of operations in the "plan formulation processing procedure" described above are saved in the multi-source / multi-dimensional DB3 as a dataset of instructions (prompts) for the generating AI in programming language (JSON, Python, etc.) format, as pairs of the most suitable problem texts and solution texts, and are used again as a training dataset for the generating AI as needed.

[0041] Figure 10 shows a detailed system configuration diagram of a health, medical, and nursing care planning system according to one embodiment, and Figure 11 shows the basic input / output data and processing flow of a health, medical, and nursing care planning system according to one embodiment. In Figure 10, the user extracts or modifies the optimal set from "language data with numerical values ​​+ System Code set (JSON, Python, etc.)" and saves it as knowledge data in the knowledge memory unit, thereby accumulating the user's knowledge in code format. In the basic relationship of input / output data to the generating AI and the flow of processing instructions to the generating AI shown in Figure 11, in order to obtain more optimal output (problem statement and countermeasure statement), the knowledge memory unit automatically strengthens and grows the knowledge data through instructions (prompts) for organization and efficiency as feedback to the field 3 (generating AI).

[0042] Here, the numerous problem-solving datasets stored in the multi-source, multi-dimensional DB3 include at least information that can identify the creator of the dataset (hereinafter referred to as "identification information"), and in addition to the identification information, they may also include information that allows for the hierarchical and systematic linking and management of the numerous problem-solving datasets (hereinafter referred to as "management information"), and instructions (prompts) for the generating AI.

[0043] This paper describes the data management methods for the numerous problem-solving datasets stored in the multi-source, multi-dimensional DB3 and the data storage unit 160. Because the data stored in the multi-source / multi-format DB3 of this embodiment is large in volume and diverse in type, the data structure of the multi-source / multi-format DB3 has a significant impact on the performance of the planning system in this embodiment (analysis speed, capacity, system efficiency, etc.), and is also closely related to the accuracy of the analysis and prediction of the generating AI. For this reason, personal information is managed according to the attributes of the multi-source / multi-format data, i.e., public nature or individual privacy and wishes. For example, as shown in Figure 7, in the multi-source / multi-format DB3 or the data storage unit 160, each data is classified into folders hierarchically according to the target organization, region, year, data attributes, and data details, and managed hierarchically. Alternatively, depending on the public nature or privacy and wishes of each data, a Flag (L1~Ln) is assigned to the hierarchy of each data, and during statistical processing, it is processed in order of priority (assigned number, etc.) and confidentiality (assigned marker, etc.) according to the Flag. In the process of processing, the prompt dataset containing the instructions given to the generating AI is stored in the knowledge storage unit in programming language (JSON, Python, etc.) format and as a pair of optimal problem statements and countermeasure statements.

[0044] Furthermore, as a method for managing the numerous problem-solving datasets and data storage unit 160, the planning system according to this embodiment hierarchically divides and manages multi-source, multi-dimensional data according to its meaning, thesaurus of health, medical care, and nursing care, or the dependency relationships between data (for example, chapters, sections, items, paragraphs, and content of a document). When the generating AI reads the hierarchically divided and managed current situation description data, or prompts containing instructions given to the generating AI during the processing process, from the current situation description data storage unit 160, extracts issues related to health, medical care, and nursing care, and sets those issues as target issues, it can calculate the meaning of the input sentence, thesaurus or the dependency relationships that can be read from it, the vector distance of words, etc., thereby improving the analysis, prediction accuracy, and speed of the generating AI.

[0045] If the creator is a secondary medical area, the user typically searches for problem-solving datasets in each secondary medical area nationwide using, for example, a word representing the target issue (and / or secondary medical area number, secondary medical area name, etc.) as a search term, and extracts one or more problem-solving datasets that match the type of target issue. The extracted problem-solving datasets may include problem-solving datasets previously created by the user (creator) themselves. If the number of extracted problem-solving datasets is large, the user may use the aforementioned specific information or management information as a search term to narrow down the results to problem-solving datasets in secondary medical areas that are similar to the creator's secondary medical area in terms of population, region, demographic / socioeconomic factors, climate, etc.

[0046] Returning to the flowchart in Figure 3, the problem-solving dataset acquired in step 6 is stored in the problem-solving dataset storage unit 162, along with a prompt containing instructions for the generating AI. In other words, the problem-solving dataset storage unit 162 and the knowledge storage unit correspond to the third storage unit.

[0047] Next, when the user performs a predetermined operation on the input unit 17, the planning unit 15 inputs one or more problem-solving datasets stored in the problem-solving dataset storage unit 162 to the generation AI system 4, as shown in Figure 10. It also identifies the target problem and the corresponding classification of countermeasures from the existing problem documents and countermeasure documents included in the problem-solving dataset, and generates a prompt requesting the generation AI to create documents for the problem and countermeasures, which is then provided to the generation AI system 4. The planning unit 15 manages the exchange of information between the user and the generation AI system 4, for example, by using an API (Application Programming Interface) provided by the generation AI system 4. The prompt is sent to the generating AI system 4, and as shown in Figure 11, the generating AI system 4 creates a provisional plan in which texts relating to the target issue and the corresponding issues and countermeasures (target issue text and target countermeasure text) are written in natural language based on the existing issue text and existing countermeasure text. Based on the user's instructions, the system sequentially displays the prompt dataset containing the instructions given to the generating AI during the processing process in a programming language (JSON, Python, etc.) format and as a pair of the optimal issue text and countermeasure text on the output unit 18 (step 7). The plan creation unit 15 saves the first draft of the created provisional plan to the provisional plan storage unit 163 and sequentially displays it on the output unit 18 (step 8).

[0048] In this configuration, if the input unit 17 is a microphone, the user can instruct the system to create text corresponding to the target issue and its classification from existing issue texts and existing countermeasure texts included in the issue countermeasure dataset using voice commands. With this configuration, through dialogue with the generation AI system 4, the user can provide instructions based on the classification of the target issue and numerical data or expert knowledge, and convert existing issue texts and existing countermeasure texts into content that reinforces the target issue or countermeasure. Furthermore, through dialogue with the generation AI system 4, the generation AI can predict future plans and social changes and present issues to the user, and the user can provide instructions based on expert knowledge to solve those issues, formulated in the form of prompts or code.

[0049] Figure 8 shows an example of the data necessary from a health, medical, and nursing care perspective when a generative AI predicts future plans and future social changes. Figure 9 shows an example of data trends (from 2000 to 2050) predicted by the generative AI from a health, medical, and nursing care perspective using the data shown in Figure 8.

[0050] As shown in Figure 11, the user looks at the contents of the provisional plan displayed sequentially on the output unit 18 to check for any points that need to be corrected or revised. If there are points that need to be corrected or revised, the user performs the prescribed operation on the input unit 17, inputs the content to be corrected or revised, and then issues a sequential plan re-creation (feedback) instruction (if No in step 9). This returns to the process of step 3, and steps 3 to 8 are repeatedly executed until Yes is selected in step 9. Here, the "content to be corrected or revised" entered by the user includes information that identifies the source and type of numerical data received in step 3, and the content of the statistics calculated in step 4. Therefore, in the second and subsequent processing of steps 3 to 8, the type of target problem is classified based on different numerical data and statistics than in the previous processing of steps 3 to 8 (step 5), a problem countermeasure dataset is extracted (step 6), and as a result, the generating AI system 4 creates a provisional plan with revised content different from the previous one (step 8).

[0051] After reviewing the contents of the provisional plan displayed in the output unit 18, if there are no corrections or revisions, the user performs the prescribed operation in the input unit 17 and issues a command to end the plan creation process (Yes in step 9). As a result, the provisional plan is finalized as the final plan and saved in the provisional plan storage unit 163 and the knowledge storage unit, and also displayed in the output unit 18 (step 10).

[0052] Figure 4 shows an example of the process from when "physician shortage" is extracted as a target issue in Step 2, and when it is identified as "physician shortage identification" as a classification of the target issue in Step 5, to when a provisional plan and a final plan are created. In this example, in Step 3, the number of physicians for each secondary medical area is accepted as numerical data, in Step 4, the standard score of the prefecture as a whole, which is the top group of the secondary medical area of ​​the creating entity, is calculated as a statistical quantity, in Steps 5 and 6, the classification of the issues and countermeasures is identified, and in Step 9, points that need to be corrected or revised regarding the content of the provisional plan are entered, and as a result of repeating Steps 3 to 8, a plan is created that includes proposed issues and countermeasures based on multiple combinations of issue classifications and countermeasure classifications.

[0053] Figure 5 shows the classification of issues and the specific content (text) of the issues related to those issues, and Figure 6 shows the classification of countermeasures and the specific content (text) of the countermeasures related to those countermeasures. From Figures 4 to 6, it can be seen that in this example, when "doctor shortage" was identified, the types of issues were classified into four categories, and as a result, four categories of countermeasures were identified, resulting in a regional medical care plan that includes four types of issues and countermeasures. It can be seen that each regional medical care plan is more specific than the provisional plan.

[0054] Furthermore, the content of the target issue documents and target countermeasure documents included in regional medical plans, as well as the classification of issues and countermeasures, vary depending on the size of the population and area of ​​the creating body, geographical conditions, etc., and may also change due to technological advancements, etc. In short, the above embodiment is merely one example of the present invention, and it is natural that any changes, modifications, additions, etc. made as appropriate within the scope of the spirit of the present invention will be included in the claims of this patent application.

[0055] [Aspect] Those skilled in the art will understand that the exemplary embodiments described above are specific examples of the following embodiments.

[0056] (Section 1) One aspect of the present invention is a method for formulating a health, medical care, and nursing care plan, Using a computer, The first reception step receives current status description data that describes the current situation of the plan creator regarding at least one of health, medical, and care aspects, Based on the current situation description data received in the first reception step, the issue setting step sets target issues related to the health, medical care, or nursing care of the creator, A second reception step in which the creation entity receives numerical data related to the target issue set in the issue setting step, The aforementioned data describing the current situation and numerical data are hierarchically divided, managed, and stored according to the meaning, thesaurus, or dependency relationships between each data item, and a storage and management step is performed. A statistical calculation step which calculates a statistical quantity relating to the top group of the entities that created the numerical data received in the second reception step, A target issue classification step involves analyzing the statistics calculated in the aforementioned statistics calculation step and classifying the target issue into one of several types, A third acceptance step accepts multiple problem-and-solution datasets, which are datasets of existing problem statements and existing solution statements, each describing an existing problem related to at least one of health, medical care, and nursing care, and an existing solution to that problem, in natural language. An extraction step in which, from among the multiple problem countermeasure datasets received in the third reception step, a problem countermeasure dataset corresponding to the type of target problem classified in the target problem classification step is extracted, A planning step that uses a generative AI to create a plan in which text describing the issues of the creator and the countermeasures for those issues is written in natural language, using the issue countermeasures dataset and numerical data extracted in the extraction step, A knowledge storage step involves saving the problem-solving dataset created in the aforementioned planning step and a set of prompts containing the instructions given to the generating AI as knowledge data. An output step that outputs the plan created in the aforementioned planning step, This is what it does.

[0057] (Paragraph 10) The health, medical care, and long-term care planning system relating to Paragraph 10 is a computer-based system, A first storage unit containing current status description data that describes the current status of the plan creator regarding at least one of health, medical, and care, A second memory unit storing numerical data related to the health, medical, and nursing care fields of the above-mentioned creation entity and its superior group, A task setting unit sets target issues related to the health, medical care, or nursing care of the creator based on the current situation description data stored in the first memory unit, A statistical calculation unit obtains numerical data of the top group of creators related to the target task set by the task setting unit from the second storage unit and calculates statistical quantities relating to the top group of the numerical data of the creators, A target issue classification unit analyzes the statistics calculated by the statistics calculation unit and classifies the target issue into one of several types, A third memory unit stores a dataset of existing problem statements and existing countermeasure statements, which are datasets of existing problem statements and existing countermeasure statements, in which one or more existing problems related to at least one of health, medical care, and nursing care, and existing countermeasures for those problems are described in natural language. A planning unit obtains a problem countermeasure dataset from the third memory unit corresponding to the type of target problem classified by the target problem classification unit, and uses a generating AI to create a plan in which the text concerning the problem of the creator and the countermeasures for that problem is written in natural language, using the problem countermeasure dataset corresponding to the type of target problem and the numerical data. A knowledge storage unit stores the problem-solving dataset created by the aforementioned planning unit and a set of prompts containing the instructions given to the generating AI as knowledge data. An output unit that outputs the plan created by the aforementioned planning unit and It is equipped with these features.

[0058] According to the method for formulating health, medical, and nursing care plans in paragraph 1, and the system for formulating health, medical, and nursing care plans in paragraph 10, based on a description of the current situation, the challenges (target challenges) of the plan-creating entity are set, statistics are obtained for the top group of numerical data representing at least one of the health, medical, and nursing care attributes of the plan-creating entity related to the target challenges, and these statistics are analyzed to classify the types of the target challenges. Once the types of target challenges are classified, a plan is created in which the challenges of the plan-creating entity and the countermeasures for those challenges are described in natural language, based on a dataset of existing challenge documents and existing countermeasure documents (challenge-countermeasure dataset) corresponding to that type, and the plan is output. Since this plan is generated using generative AI, by creating a trained model that has learned the patterns and features using a large dataset of existing challenge documents and existing countermeasure documents, high-quality plans can be easily created even without specialized knowledge. In the above method and system, challenges or solutions can be presented by using a large amount of existing multi-source and multi-variable data, or by predicting future plans and social changes.

[0059] (Paragraph 2) The method for formulating health, medical, and nursing care plans relating to Paragraph 2 is as described in the method for formulating health, medical, and nursing care plans relating to Paragraph 1, Using the aforementioned computer, A further management step may be performed to manage the current status description data received in the first reception step and the numerical data received in the second reception step, or prompts including instructions given to the AI ​​generated during the processing, based on a thesaurus related to health, medical care, and nursing care.

[0060] According to the health, medical, and nursing care planning method described in paragraph 2, current situation description data, numerical data, and instructions (prompts) given to the generating AI are managed based on a thesaurus related to health, medical, and nursing care. Therefore, in the planning step, the generating AI can improve the accuracy of its analysis of the challenges and countermeasures of the planning entity and shorten the processing time.

[0061] (Paragraph 3) The method for formulating health, medical care and nursing care plans relating to Paragraph 3 shall be the same as the method for formulating health, medical care and nursing care plans relating to Paragraph 1 or Paragraph 2, The aforementioned entity may be an organization including the national government, local governments, communities, schools, companies, medical facilities, nursing care facilities, health insurance associations (such as the National Health Insurance, the Medical Care System for the Elderly, the National Health Insurance Association, and other health insurance societies), or an individual.

[0062] (Paragraph 4) The method for formulating a health, medical, and nursing care plan relating to Paragraph 4 shall be the same as the method for formulating a health, medical, and nursing care plan relating to Paragraph 1 or 2, The aforementioned creator is the national government, a local government, or a local community. The aforementioned numerical data may consist of the number and location information of human or physical resources related to health, medical care, and nursing care that exist within the entity that created the data.

[0063] (Article 5) The method for formulating a health, medical, and nursing care plan relating to Article 5 shall be the same as the method for formulating a health, medical, and nursing care plan relating to any one of Articles 1 to 4. The aforementioned numerical data may include numerical data of geospatial information of the entity that created it.

[0064] (Paragraph 6) The method for formulating a health, medical, and nursing care plan relating to Paragraph 6 is the method for formulating a health, medical, and nursing care plan relating to any one of Paragraphs 1 to 5, Using a computer, A voice or conversation input step that receives voice or conversation input from the user, The system further performs a prompt creation step which involves analyzing the voice or conversation received in the voice or conversation receiving step and creating a prompt that includes questions and / or instructions for the generating AI. The aforementioned planning step is, Using a generation AI, the plan is created by transforming existing problem statements and sentences related to the target problem contained in the problem countermeasure dataset extracted in the extraction step, based on the type of the target problem, the numerical data, or the prompt created in the prompt creation step. By saving the problem-solving dataset created in the aforementioned planning step and the instructions (prompts) given to the generating AI as knowledge data, the user's knowledge is accumulated. If necessary, the created plan dataset can be used again as training data for the generating AI, allowing the knowledge to automatically evolve even without specialized knowledge, and enabling the creation of high-quality plans easily.

[0065] In the method for formulating health, medical, and nursing care plans relating to paragraph 6, the voice or conversational text reception step may include a function to convert the language of the input voice or conversational text into a language of another region, as necessary, depending on the usage environment, user instructions, etc.

[0066] (Paragraph 7) The method for formulating a health, medical, and nursing care plan relating to Paragraph 7 is the same as the method for formulating a health, medical, and nursing care plan relating to any one of Paragraphs 1 to 6. The aforementioned planning step is, The plan can be created by using a generation AI to predict future plans and social changes in healthcare and long-term care, and by transforming existing problem statements and sentences related to the target problem contained in the existing countermeasures statements included in the problem countermeasures dataset extracted in the extraction step.

[0067] (Paragraph 8) The method for formulating a health, medical, and nursing care plan relating to Paragraph 8 is the same as the method for formulating a health, medical, and nursing care plan relating to any one of Paragraphs 1 to 7. The aforementioned problem setting step may involve receiving input prompts that include terms and numerical data related to health, medical care, and nursing care, or instructions given to the AI ​​generated during the processing, and setting the target problem based on the terms, the numerical data, and the current situation description received in the first reception step.

[0068] (Paragraph 9) The plan creation support method relating to Paragraph 9 is a plan creation support method relating to any one of Paragraphs 1 to 8, Using a computer, A voice or conversation input receiving step that accepts voice or conversation input, The system further performs a prompt creation step which involves analyzing the voice or conversation received in the voice or conversation receiving step and creating a prompt that includes questions and / or instructions for the generating AI. In the second reception step, the numerical data is received based on the prompt created in the prompt creation step. The extraction step may involve extracting the problem-solving dataset based on the prompt created in the prompt creation step.

[0069] In the method for formulating health, medical, and nursing care plans related to paragraph 9, the voice or conversational text reception step may include a function to convert the language of the input voice or conversational text into a language of another region, as necessary, depending on the usage environment, user instructions, etc.

[0070] (Paragraph 11) The plan creation support method relating to Paragraph 11 is a plan creation support method relating to any one of Paragraphs 1 through 9, Using a computer, A voice or conversation input receiving step that accepts voice or conversation input, The system further performs a prompt creation step which involves analyzing the voice or conversation received in the voice or conversation receiving step and creating a prompt that includes questions and / or instructions for the generating AI. In the prompt creation step, if a prompt is created that instructs the reproduction of the plan output in the output step, the generating AI can be used to execute the problem setting step, the second reception step, the statistics calculation step, the target problem classification step, the third reception step, the extraction step, and the plan creation step, using the plan as the problem countermeasures dataset.

[0071] (Clause 12) The data management method relating to paragraph 12 manages the current situation description data received in the first reception step of the health, medical care, and nursing care planning method described in any one of claims 1 to 9, or the instructions (prompts) given to the generating AI during the processing, and the numerical data received in the second reception step, according to the attributes of the entity that created the current situation description data and the numerical data.

[0072] (Paragraph 13) The data management method relating to Paragraph 13 involves managing the current situation description data received in the first reception step of the method for formulating health, medical care, and nursing care described in any one of claims 1 to 9, or the instructions (prompts) given to the generating AI during the processing, and the numerical data received in the second reception step, in accordance with a thesaurus relating to health, medical care, and nursing care. [Explanation of symbols]

[0073] 1… Terminal 10...Data Entry Reception Department 11…Problem Setting Department 12…Statistics calculation section 13…Dataset extraction unit for problem-solving 14…Target Issue Classification Section 15…Planning Department 16...Storage section 160...Current status description data storage unit 161... Numerical data storage unit 162... Problem-solving dataset storage unit 163... Provisional Plan Preservation Department 17...Input section 18…Output section 2…Internet 3…Multi-source multi-dimensional DB 4…Generative AI system

Claims

1. A method for formulating a health, medical, and nursing care plan, Using a computer, The first reception step receives current status description data that describes the current situation of the plan creator regarding at least one of health, medical, and care aspects, Based on the current situation description data received in the first reception step, the issue setting step sets the target issues related to the health, medical care, or nursing care of the creator, A second reception step in which the creation entity receives numerical data related to the target issue set in the issue setting step, A statistical calculation step which calculates statistical quantities relating to the top group of the entities that created the numerical data received in the second reception step, A target issue classification step involves analyzing the statistics calculated in the aforementioned statistics calculation step and classifying the target issue into one of several types, A third acceptance step accepts multiple problem-and-solution datasets, which are datasets of existing problem texts and existing solution texts that describe existing problems and existing solutions related to at least one of health, medical care, and nursing care in natural language. An extraction step is performed to extract from the multiple problem countermeasure datasets received in the third reception step the above that a problem countermeasure dataset corresponds to the type of target problem classified in the target problem classification step, A planning step that uses a generation AI to create a plan in which text describing the issues of the creator and the countermeasures for those issues is written in natural language, using the issue countermeasures dataset and numerical data extracted in the extraction step, A knowledge storage step involves saving a set of pairs of problem-solving datasets created in the aforementioned planning step and prompts containing the instructions given to the generating AI as knowledge data. An output step that outputs the plan created in the aforementioned planning step, A method for formulating health, medical, and nursing care plans.

2. In the method for formulating a health, medical, and nursing care plan according to claim 1, Using the aforementioned computer, A method for formulating a health, medical, and nursing care plan according to claim 1, further comprising a management step of managing the current situation description data received in the first reception step and the numerical data received in the second reception step or prompts including instructions given to the generated AI during the processing, based on a thesaurus relating to health, medical care, and nursing care.

3. The method for formulating a health, medical, and long-term care plan as described in claim 1, wherein the entity that creates the plan is an organization or individual that includes the national government, local governments, communities, schools, companies, medical facilities, nursing care facilities, health insurance associations (such as the National Health Insurance, the Medical Care System for the Elderly, the National Health Insurance Association, and other health insurance associations).

4. The aforementioned creator is the national government, a local government, or a local community. The method for formulating a health, medical, and nursing care plan according to claim 1, wherein the numerical data is the number and location information of human or physical resources related to health, medical care, and nursing care that exist within the entity that created the plan.

5. The method for formulating a health, medical, and nursing care plan according to claim 1, wherein the numerical data includes numerical data of geospatial information of the entity that created the plan.

6. Using a computer, A voice or conversation input step that receives voice or conversation input from the user, The system further performs a prompt creation step which involves analyzing the voice or conversation received in the voice or conversation receiving step and creating a prompt that includes questions and / or instructions for the generating AI. The aforementioned planning step is, A method for formulating a health, medical, and nursing care plan according to claim 1, wherein a generating AI is used to create a plan by converting existing problem statements and sentences related to the target problem contained in the problem countermeasure dataset extracted in the extraction step, based on the type of the target problem, the numerical data, or the prompt created in the prompt creation step, and the plan and the instructions (prompts) of the generating AI are saved as a pair of knowledge data. Furthermore, if necessary, a new plan is created by again using the knowledge data as training data for the generating AI.

7. The aforementioned planning step is, The method for creating a plan according to claim 6, wherein the plan is created by using a generation AI to predict future plans and social changes in health, medical care, and nursing care, and by converting existing problem sentences and sentences related to the target problem contained in the existing countermeasures sentences included in the problem countermeasures dataset extracted in the extraction step.

8. The method for formulating a health, medical, and nursing care plan according to claim 1, wherein the task setting step receives input of a prompt including terms and numerical data related to health, medical care, and nursing care, or instructions given to the AI ​​generated during the processing, and sets the target task based on the terms, the numerical data and the current situation description data received in the first reception step.

9. Using a computer, The system further includes a voice or conversation input receiving step, and a prompt creation step, which analyzes the voice or conversation received in the voice or conversation input step to create a prompt that includes questions and / or instructions for the generating AI. In the second reception step, the numerical data is received based on the prompt created in the prompt creation step. The method for formulating a health, medical, and nursing care plan according to claim 1, wherein the extraction step extracts the problem-solving dataset based on the prompt created in the prompt creation step.

10. A computer-based system for formulating health, medical, and nursing care plans, A first storage unit that stores current status description data describing the current situation of the plan creator regarding at least one of health, medical, and care, A second memory unit storing numerical data related to the health, medical, and nursing care fields of the above-mentioned creation entity and its superior group, A task setting unit sets target issues related to the health, medical care, or nursing care of the creator based on the current situation description data stored in the first memory unit, A statistical calculation unit obtains numerical data of the top group of creators related to the target task set by the task setting unit from the second storage unit and calculates statistical quantities relating to the top group of creators' numerical data, A target issue classification unit analyzes the statistics calculated by the statistics calculation unit and classifies the target issue into one of several types, A third memory unit stores a dataset of existing problem statements and existing countermeasure statements, which are datasets of existing problem statements and existing countermeasure statements, describing in natural language one or more existing problems related to at least one of health, medical care, and nursing care, and existing countermeasures for those problems. A planning unit obtains a problem countermeasure dataset from the third memory unit corresponding to the type of target problem classified by the target problem classification unit, and uses generating AI to create a plan in which text relating to the problem of the creator and countermeasures for that problem is written in natural language, using the problem countermeasure dataset corresponding to the type of target problem and the numerical data. An output unit that outputs the plan created by the aforementioned planning unit and A system for planning health, medical care, and nursing care, equipped with the necessary components.

11. Using a computer, A voice or conversation input receiving step that accepts voice or conversation input, The system further performs a prompt creation step which involves analyzing the voice or conversation received in the voice or conversation receiving step and creating a prompt that includes questions and / or instructions for the generating AI. A method for formulating a health, medical, and nursing care plan according to claim 1, wherein, in the prompt creation step, a prompt is created that instructs the reproduction of the plan output in the output step, the generating AI is used to execute the issue setting step, the second reception step, the statistical calculation step, the target issue classification step, the third reception step, the extraction step, and the plan creation step, using the plan as the issue countermeasures dataset.

12. In the first reception step of the health, medical, and nursing care planning method described in any one of claims 1 to 9, the current situation description data received or the prompt including the instructions given to the generating AI during the processing, and the numerical data received in the second reception step, the attributes of the entity that created the current situation description data and the numerical data. A data management method that manages data according to the requirements.

13. A data management method for managing current situation description data received in the first reception step and numerical data received in the second reception step or prompts including instructions given to the generated AI during the processing, according to a thesaurus relating to health, medical care, and nursing care, in accordance with any one of claims 1 to 9.

Images