KR-20260064458-A - METHOD AND SYSTEM FOR SETTING EXERCISE PROGRAM DIFFICULTY LEVEL USING A LARGE LANGUAGE MODEL

KR20260064458AKR 20260064458 AKR20260064458 AKR 20260064458AKR-20260064458-A

Abstract

A method for setting the difficulty level of an exercise program using a large language model according to the present invention may include the steps of: providing at least one survey related to a user’s indication to an electronic device; receiving response data for the survey from the electronic device; setting evaluation criteria for evaluating the user’s exercise performance ability based on at least one of the response data and the user’s history information; specifying at least one evaluation item among a plurality of evaluation items related to the evaluation of exercise performance ability based on the evaluation criteria; inputting the user’s exercise data into a pre-trained motion evaluation model to evaluate the user’s exercise performance ability for the at least one evaluation item; generating a prompt to be input into a large language model using the response data and the evaluation result for the exercise performance ability; inputting the prompt into the large language model to obtain an exercise difficulty level related to the user’s indication from the large language model; and generating an exercise program related to the user’s indication based on the obtained exercise difficulty level.

Inventors

윤찬
차하리

Assignees

에버엑스 주식회사

Dates

Publication Date: 20260507
Application Date: 20250430
Priority Date: 20241030

Claims (15)

A step of providing at least one questionnaire related to the user's indications to an electronic device; A step of receiving response data for the survey from the electronic device; A step of setting evaluation criteria for evaluating the user's exercise performance ability based on at least one of the above response data and the user's history information; A step of specifying at least one evaluation item among a plurality of evaluation items related to exercise performance evaluation based on the above evaluation criteria; A step of inputting the exercise data of the above user into a pre-trained motion evaluation model to evaluate the exercise performance ability of the above user for at least one evaluation item; A step of generating a prompt to be input into a large language model using the above response data and the evaluation results of the above exercise performance ability; A step of inputting the above prompt into the above large language model to obtain an exercise difficulty related to the user's indication from the above language model; and A method for setting the difficulty level of an exercise program using a large language model, characterized by including the step of generating an exercise program related to the user's indications based on the obtained exercise difficulty level.
In paragraph 1, The above evaluation criteria are, It is set differently based on at least one of the response data and history information for the above survey, and The step of specifying at least one evaluation item above is, A step of identifying evaluation restriction items that satisfy unsuitability conditions among a plurality of evaluation items related to the exercise performance evaluation based on the evaluation criteria set differently above; and A method for setting the difficulty level of an exercise program using a large language model, characterized by including the step of specifying at least one evaluation item among the plurality of evaluation items based on the specified evaluation limit item.
In paragraph 2, The above evaluation limitation items are, A method for setting the difficulty level of an exercise program using a large language model, characterized by being determined based on at least one of the pain level of the user's pain area and medical records included in at least one of the response data to the above survey and the above history information.
In paragraph 3, The above non-conforming conditions are, A method for setting the difficulty of an exercise program using a large language model, characterized by a condition specifying the evaluation restriction item that is excluded from the evaluation of the user's exercise performance ability based on the pain level and a restriction standard previously set in at least one of the medical records.
In paragraph 2, The step of evaluating the exercise performance ability of the above user is, A step of receiving the user's exercise data corresponding to at least one specified evaluation item from the electronic device; A step of analyzing at least one of the user's static posture, joint range of motion, balance ability, and muscle strength using the above-mentioned learned motion evaluation model; and A method for setting the difficulty level of an exercise program using a large language model, characterized by including the step of generating an exercise performance evaluation result for at least one evaluation item.
In paragraph 1, The above exercise program is configured to include a plurality of exercise modules, and Each of the above plurality of motion modules is, A method for setting the difficulty level of an exercise program using a large language model, characterized by being matched to different exercise types and including at least one exercise motion content related to the exercise type matched to each of the exercise modules.
In paragraph 6, The step of acquiring the above exercise difficulty level is, A method for setting the difficulty of an exercise program using a large language model, characterized by obtaining a specific exercise difficulty among a plurality of exercise difficulty levels through the large language model into which the prompt is input, based on a pre-set difficulty setting standard.
In Paragraph 7, The above-mentioned at least one exercise motion content includes motion difficulty information, and Each of the above plurality of motion modules is, A method for setting the difficulty of an exercise program using a large language model, characterized by including at least one specific exercise movement content that satisfies a preset exercise difficulty condition based on the above specific exercise difficulty.
In paragraph 8, The above-mentioned difficulty setting criteria include multiple different setting criteria, and A method for setting the difficulty level of an exercise program using a large language model, characterized in that the above-mentioned specific exercise difficulty level satisfies all of the above-mentioned different multiple setting criteria.
In Paragraph 7, The step of generating the above exercise program is, A step of extracting at least one specific exercise movement content satisfying the preset exercise difficulty condition using the specific exercise difficulty and the prompt; and A method for setting the difficulty level of an exercise program using a large language model, characterized by including the step of generating the exercise program composed of the plurality of exercise modules based on the exercise type matched to at least one specific exercise motion content extracted through the large language model.
In paragraph 1, The above at least one survey is, A basic questionnaire including questions related to the indications of the above-mentioned user and A method for setting the difficulty level of an exercise program using a large language model, characterized by including at least one additional survey provided differently depending on the user's response to the basic survey above.
In Paragraph 11, Each of the above basic survey and the above additional survey is, A multiple-choice questionnaire comprising at least one question item related to the indications of the user and a plurality of selection items corresponding to each of a plurality of different responses to the question item, and A method for setting the difficulty level of an exercise program using a large language model, characterized by including at least one interactive questionnaire capable of receiving natural language input from the electronic device in relation to the indications of the user.
In Paragraph 12, The step of receiving response data for the above survey is, A step of receiving response data for the selection survey from the electronic device, the response being matched to an item selected by user input among the plurality of selection items; and A method for providing an exercise program using a large language model, characterized by including the step of processing the natural language input entered into the electronic device as input to the large language model and receiving response data for the conversational survey from the large language model.
It includes a control unit that provides at least one survey related to a user's indication to an electronic device, and a communication unit that receives response data for the survey from the electronic device. The above control unit is, A system for providing an exercise program using a large language model, characterized by establishing an evaluation criterion for evaluating the exercise performance ability of the user based on at least one of the above response data and the user's history information, specifying at least one evaluation item among a plurality of evaluation items related to the evaluation of exercise performance ability based on the above evaluation criterion, inputting the user's exercise data into a pre-trained motion evaluation model to evaluate the user's exercise performance ability for the at least one evaluation item, generating a prompt to be input into a large language model using the above response data and the evaluation result for the exercise performance ability, inputting the prompt into the large language model to obtain an exercise difficulty related to the user's indication from the large language model, and generating an exercise program related to the user's indication based on the obtained exercise difficulty.
A program that is executed by one or more processes in an electronic device and stored on a computer-readable recording medium, The above program is, A step of providing at least one questionnaire related to the user's indications to an electronic device; A step of receiving response data for the survey from the electronic device; A step of setting evaluation criteria for evaluating the user's exercise performance ability based on at least one of the above response data and the user's history information; A step of specifying at least one evaluation item among a plurality of evaluation items related to exercise performance evaluation based on the above evaluation criteria; A step of inputting the exercise data of the above user into a pre-trained motion evaluation model to evaluate the exercise performance ability of the above user for at least one evaluation item; A step of generating a prompt to be input into a large language model using the above response data and the evaluation results of the above exercise performance ability; A step of inputting the above prompt into the above large language model to obtain an exercise difficulty related to the user's indication from the above language model; and A program stored on a computer-readable recording medium characterized by including instructions that perform the step of generating an exercise program related to the user's indications based on the acquired exercise difficulty.

Description

Method and System for Setting Exercise Program Difficulty Level Using a Large Language Model The present invention relates to a method and system for setting the difficulty level of an exercise program using a large language model. This research was conducted with funding from the Ministry of Health and Welfare and supported by the Health and Medical Technology Research and Development Project of the Korea Health Industry Development Institute. (Project Unique Number: 2460000121, Project Number: RS-2024-00401350, Ministry: Ministry of Health and Welfare, Project Management (Specialized) Agency: Korea Health Industry Development Institute, Research Project Name: Translational Research Linked to Clinical Field Demand, Research Project Title: Development and Clinical Validation of a Comprehensive Digital Healthcare Service for Personalized Self-Rehabilitation After Rotator Cuff Reconstruction and Shoulder Arthroplasty, Project Executing Agency: Sungkyunkwan University Industry-Academic Cooperation Foundation, Research Period: 2024-04-01~2026-12-31). With the rapid advancement of artificial intelligence (AI) technology in recent years, generative AI models capable of natural conversation with humans (e.g., ChatGPT) have emerged. In particular, Large Language Models (LLMs), unlike existing dictionary-rule-based chatbots, provide conversational quality similar to humans based on the ability to understand various contexts and generate natural language, and are being rapidly adopted in various industrial fields such as medical, education, and healthcare. Furthermore, alongside the advancement of artificial intelligence (AI) technology, its application in the medical industry is rapidly expanding. For example, there is growing interest in AI technology that utilizes large language models to provide personalized, non-face-to-face rehabilitation to patients requiring consistent management and rehabilitation during treatment. In this medical and healthcare field, large language models can be effectively utilized for the management and rehabilitation of musculoskeletal disorders. Musculoskeletal disorders refer to pain or injury occurring in the musculoskeletal system, including muscles, nerves, tendons, ligaments, bones, and surrounding tissues. As a general principle, the treatment of musculoskeletal disorders should begin with less invasive procedures; non-pharmacological conservative treatments (e.g., exercise therapy and education, cognitive therapy, or relaxation therapy) should be implemented first, followed by pharmacological treatment and surgical treatment in sequence. Treatment guidelines strongly recommend non-pharmacological conservative treatment for musculoskeletal disorders, and active research on methods for implementing such treatments is being conducted, primarily in the United States and Europe. However, since continuous treatment and rehabilitation are crucial for non-pharmacological conservative treatment, the requirement for patients to visit the hospital frequently poses a significant burden. Consequently, interest in services providing non-face-to-face rehabilitation is increasing. In non-face-to-face rehabilitation, it is essential to deliver an exercise program accompanied by setting the difficulty level of rehabilitation exercises while considering the user's age, gender, occupation, level of daily living activities, current primary diagnosis and history of related diseases, past medical history and surgical history, medication history, current physical functional status (muscle strength, range of motion, balance, presence and severity of pain, etc.), cardiorespiratory function, neurological status, mental/psychological state, lifestyle habits (smoking, drinking, dietary habits, exercise experience, etc.), residential and exercise environments (use of mobility aids, exercise equipment owned, etc.), specific details (skin condition, condition of surgical site, etc.), and user goals. If exercise difficulty is set uniformly without considering the user's functional level or medical history (e.g., arthritis, rotator cuff injury, surgical records, etc.), exercises of inappropriate difficulty may be delivered due to a failure to account for differences in diseases, functional levels, and age. This can increase the risk of cardiorespiratory strain, musculoskeletal damage, and re-injury, which may lead to the worsening of pain or functional decline. Furthermore, in non-face-to-face rehabilitation therapy, if the difficulty level is inappropriate, the effectiveness of the rehabilitation may be diminished. For example, if an exercise program of easy difficulty is conducted, the body may not receive appropriate stimulation (exercise threshold), failing to induce physical changes; consequently, necessary recovery may not occur, or achieving goals may take longer. Additionally, if the exercise feels boring or meaningless, user motivation may decrease, potentially leading to a higher dropout rate from the