KR-102964648-B1 - SYSTEM DEVICE FOR PROVIDING PERSONALIZED HEALTH MANAGEMENT INFORMATION AND METHOD OF OPERATION OF THE SYSTEM DEVICE

KR102964648B1KR 102964648 B1KR102964648 B1KR 102964648B1KR-102964648-B1

Abstract

The method of operation of a system device may include the steps of: acquiring personal health information of a user; providing the acquired personal health information and an information checklist to a server; acquiring evaluation information from the server that evaluates the personal health information based on the information checklist generated through a Large Language Model (LLM); providing a request signal requesting at least one necessary health information not included in the personal health information among the information checklist based on the evaluation information; acquiring additional personal health information of the user corresponding to the request signal; retrieving expertise from a database containing multiple pieces of expertise; providing the retrieved expertise, personal health information, and additional personal health information to a server; and acquiring and providing user-customized health management information generated through a Large Language Model (LLM) from the server.

Inventors

황희정
이기원
서봉원
김정환
김규식
허동석
김나라
유정연
윤정한

Assignees

서울대학교산학협력단

Dates

Publication Date: 20260512
Application Date: 20251001

Claims (10)

In a method of operation of a system device that provides alternative user-customized health management information, A step of acquiring a user's personal health information including multiple health information items; A step of providing an information checklist to a server that includes the acquired personal health information and a plurality of pre-set necessary health information items, and obtaining from the server evaluation information that evaluates the personal health information based on the information checklist, which is generated through a Large Language Model (LLM); A step of providing a request signal to request at least one necessary health information item not included in the personal health information among the information checklist based on the above evaluation information; A step of obtaining additional personal health information of the user, including at least one additional health information item corresponding to the above request signal; A step of retrieving at least one piece of expertise corresponding to the personal health information and the additional personal health information from a database containing multiple pieces of expertise; A step of obtaining the ratio of the plurality of necessary health information items to the plurality of health information items included in the personal health information and the at least one additional health information item included in the additional personal health information as information completeness, selecting one personalization level among a plurality of personalization levels preset with different thresholds by comparing the threshold and the information completeness, and not requesting user-customized health management information from the large language model if the information completeness is smaller than a preset minimum value; A step of providing at least one searched expert knowledge, the personal health information, the additional personal health information, and the selected personalization level to the server, and obtaining personalized user-customized health management information from the server, which is generated to correspond to the selected personalization level through the Large Language Model (LLM), wherein the level of explanation, specificity of content, and sentence structure of the information included in the user-customized health management information are determined according to the selected personalization level; A step of obtaining alternative information indicating the degree to which nutrients included in at least one alternative food menu, which replaces at least one food menu among the plurality of food menus included in the user-customized health management information through a hill climbing algorithm, deviate from constraints considering the user's health condition based on the personal health information and the additional personal health information; A step of obtaining similarity information by calculating the similarity between the at least one food menu and the at least one alternative food menu through the EMD (Earth Mover's Distance) algorithm; A step of obtaining a cost function based on the value obtained by multiplying the replacement information by the first weight and the value obtained by multiplying the similarity information by the second weight through the LSAP (Linear Sum Assignment Problem) algorithm, wherein the sum of the first weight and the second weight is 1, and the magnitude of the first weight increases as the compliance with the constraint condition is considered more important, and the magnitude of the second weight increases as the high similarity is considered more important; A step of obtaining optimized alternative user-customized health management information by using the above cost function, taking into account the user's constraints and the similarity with at least one food menu from the user-customized health management information; and A method of operation of a system device comprising the step of providing alternative user-customized health management information obtained above.
In Article 1, The method of operation of the above-mentioned system device is, It further includes a step of determining whether the additional personal health information corresponding to the above request signal has been obtained, and As it is determined that the above additional personal health information has been obtained, the method of operation of the system device performs the step of searching for the at least one piece of expertise from the database, and A method of operation of a system device that, as it is determined that the above additional personal health information has not been obtained, performs the step of providing the request signal again.
In Article 2, The above database is, A method of operation of a system device including a medical knowledge database containing medical literature, medical guidelines, and public data.
In Paragraph 3, The above database is, A method of operation of a system device including a medical record database containing questions and answers between a doctor and a patient or conventional medical records.
In Paragraph 4, The above database is, A method of operation of a system device including an electronic medical record database including an electronic medical record (EMR).
In Article 2, The method of operation of the above-mentioned system device is, A step of obtaining feedback information regarding the user-customized health management information obtained from the server; and A method of operation of a system device further comprising the step of providing the above-mentioned acquired feedback information to the above-mentioned large language model in real time through the above-mentioned server.
In Article 2, The method of operation of the above-mentioned system device is, A step of identifying whether there exists an emergency situation corresponding to the user's health condition among a preset emergency situation judgment list based on the user-customized health management information obtained from the server; and A method of operation of a system device comprising further a step of performing a preset emergency response action as it is identified that the above emergency situation exists.
delete
delete
A system device that provides alternative user-customized health management information, Memory for storing at least one instruction; and It includes at least one processor that executes at least one instruction stored in the memory, By the above at least one processor executing the above at least one instruction, the system device, Acquires a user's personal health information including multiple health information items, and An information checklist including the aforementioned acquired personal health information and a plurality of pre-set necessary health information items is provided to a server, and evaluation information that evaluates the personal health information based on the information checklist, generated through a Large Language Model (LLM), is obtained from the server. Based on the above evaluation information, a request signal is provided to request at least one necessary health information item not included in the above personal health information among the above information checklist, and Acquiring additional personal health information of the user, including at least one additional health information item corresponding to the above request signal, and Retrieve at least one piece of expertise corresponding to the personal health information and the additional personal health information from a database containing multiple pieces of expertise, and The ratio of the plurality of necessary health information items to the plurality of health information items included in the personal health information and the at least one additional health information item included in the additional personal health information is obtained as the information completeness, and among a plurality of personalization levels preset with different thresholds, one personalization level is selected by comparing the threshold and the information completeness, and if the information completeness is smaller than a preset minimum value, user-customized health management information is not requested from the large language model, The above-mentioned at least one piece of specialized knowledge, the above-mentioned personal health information, the above-mentioned additional personal health information, and the above-mentioned selected personalization level are provided to the server, and the above-mentioned personalized health management information is obtained from the server, which is generated to correspond to the above-mentioned selected personalization level through the above-mentioned Large Language Model (LLM), wherein the level of information, the specificity of the content, and the sentence structure included in the above-mentioned personalized health management information are determined according to the above-mentioned selected personalization level. Through a hill climbing algorithm, alternative information is obtained indicating the degree to which the nutrients included in at least one alternative food menu, which replaces at least one food menu among the plurality of food menus included in the user-customized health management information, deviate from constraints that consider the user's health condition based on the personal health information and the additional personal health information. Similarity information is obtained by calculating the similarity between the at least one food menu and the at least one alternative food menu through the EMD (Earth Mover's Distance) algorithm, and A cost function is obtained through the LSAP (Linear Sum Assignment Problem) algorithm based on the value obtained by multiplying the replacement information by a first weight and the value obtained by multiplying the similarity information by a second weight, wherein the sum of the first weight and the second weight is 1, the magnitude of the first weight increases as the compliance with the constraint condition is considered more important, and the magnitude of the second weight increases as the high similarity is considered more important. Using the above cost function, optimized alternative user-customized health management information is obtained by considering the user's constraints and the similarity with at least one food menu from the user-customized health management information, and A system device that provides alternative user-customized health management information obtained above.

Description

System device for providing personalized health management information and method of operation of the system device The present invention relates to a system device and a method of operation of the system device for providing user-customized health management information. Specifically, the invention relates to a system device and a method of operation of the system device for providing user-customized health management information obtained based on personal health information provided by a user and corresponding expertise. With the recent advancement of artificial intelligence technology, particularly natural language processing technology based on Large Language Models (LLMs), automated response systems for user queries are being actively adopted in the medical and healthcare fields. Along with this, methods are being developed to retrieve highly reliable external knowledge using Retrieval Augmented Generation (RAG) technology and combine it with generative responses. While these technologies are increasing accessibility to health information, they still have limitations in that they are limited to standardized responses or answers based on general knowledge. In particular, there is a growing demand for technology to provide precise responses that take into account an individual's disease status, medication information, and dietary restrictions. FIG. 1 is a drawing for explaining the operation of a system device according to one embodiment of the present invention. FIG. 2 is a drawing for explaining the configuration of a system device according to one embodiment of the present invention. FIG. 3 is a flowchart for explaining the operation method of a system device according to an embodiment of the present invention. FIG. 4 is a drawing for explaining a database according to an embodiment of the present invention. FIG. 5 is a flowchart for explaining the operation of a system device according to whether additional personal health information corresponding to a request signal is obtained according to an embodiment of the present invention. FIG. 6 is a flowchart illustrating a method of operation for a system device that provides personalized user-customized health management information according to a selected personalization level, according to one embodiment of the present disclosure. FIG. 7 is a flowchart illustrating the operation method of a system device that provides feedback information regarding user-customized health management information to a large language model in real time, according to one embodiment of the present disclosure. FIG. 8 is a flowchart for explaining a method of operation of a system device according to whether there is an emergency situation corresponding to the user's health condition according to one embodiment of the present disclosure. FIG. 9 is a flowchart illustrating a method of operation for a system device that obtains alternative user-customized health management information that can replace the obtained user-customized health management information according to one embodiment of the present disclosure. In the following, embodiments are described in detail with reference to the attached drawings. However, various modifications may be made to the embodiments, and thus the scope of the patent application is not limited or restricted by these embodiments. It should be understood that all modifications, equivalents, and substitutions to the embodiments are included within the scope of the rights. Specific structural or functional descriptions of the embodiments are disclosed merely for illustrative purposes and may be modified and implemented in various forms. Accordingly, the embodiments are not limited to specific disclosed forms, and the scope of this specification includes modifications, equivalents, or substitutions that fall within the technical concept. Terms such as "first" or "second" may be used to describe various components, but these terms should be interpreted solely for the purpose of distinguishing one component from another. For example, the first component may be named the second component, and similarly, the second component may be named the first component. Furthermore, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless explicitly and specifically defined otherwise. In certain cases, terms have been selected at the applicant's discretion, and in such cases, their meanings will be described in detail in the relevant explanatory sections. Accordingly, terms used in this disclosure should be defined not merely by their names, but based on their meanings and the content throughout this disclosure. When it is stated that a component is "connected" to another component, it should be understood that it may be directly connected to or coupled with that other component, or that there may be other components in between. The terms used in the embodiments are for illustrative purposes only and should not be interpreted as