본문 바로가기 주메뉴 바로가기
국회도서관 홈으로 정보검색 소장정보 검색

결과 내 검색

동의어 포함

초록보기

The regulatory approval process for the manufacture, export, and import of medical devices mandates a series of procedures determined by device classification. Among these, the clinical trial phase is pivotal for demon- strating the safety and efficacy of the medical device. In particular, the thorough preparation of a clinical trial protocol at the initial stage is essential for facilitating the overall procedures and plays a key role in minimizing unnecessary time and cost expenditures during subsequent regulatory processes. Consequently, effective protocol preparation is a critical component in medical device development. However, practitioners in the field often face significant chal- lenges due to the need to comply with evolving regulations and the limited accessibility of up-to-date clinical trial protocol examples. While various support services have been introduced to address these issues, they frequently fall short in providing immediate and practical assistance. This study developed and evaluated MEDIVA (MEDical Inves- tigation Validation Assistant), a Retrieval-Augmented Generation (RAG)-based Large Language Model (LLM) tool designed to support the drafting of clinical trial protocols. The comparative analysis of completions generated by MEDIVA and a standard LLM lacking RAG capabilities based on pre-selected 4 representative questions likely to be encountered during protocol development demonstrated that MEDIVA provides more precise, actionable guidance through factually accurate and hallucination-free responses. It also delivers specific and consistent responses grounded in the latest regulatory information. The developed system is expected to extend its utility beyond protocol drafting, ultimately strengthening regulatory compliance capabilities, which are essential for the advancement of the medical device industry.

권호기사

권호기사 목록 테이블로 기사명, 저자명, 페이지, 원문, 기사목차 순으로 되어있습니다.
기사명 저자명 페이지 원문 목차
근전도 신호를 이용한 방향키 입력 기반 천지인 스펠러 시스템 개발 = Development of an EMG-based Chunjiin speller system using directional input 유기연, 안지원, 유도현, 최승호 p. 319-333
머신러닝 기반 맨발 및 운동화 착용 시 하지 근육 활성도 비교 분석을 통한 보행 상태 분류 연구 = A machine learning-based study on gait classification through comparative analysis of lower limb muscle activity in barefoot and shod walking 이미주, 김대희, 권영재, 최상일, 김정훈 p. 334-344
경골부 구성요소 내구성 평가를 위한 CoCrMo 합금 물성 분석 및 유한요소 해석 기반 유효성 평가 = Mechanical property evaluation of CoCrMo alloy and finite element-based validation for durability assessment of tibial components 박광민, 조민영, 강관수, 정태곤 p. 345-351
MEDIVA = MEDIVA : a support tool for medical device clinical trial protocol development using RAG-based LLMs : RAG기반의 LLM을 활용한 의료기기 임상시험계획 지원 도구 김혜은, 류영채, 문유빈, 남기창, 양윤석 p. 352-365
대동맥 인조혈관 원형 스테이플 문합에서의 누출 위험 분석 = Leakage risk analysis in aortic graft circular stapler anastomosis : a finite element simulation study : 유한요소 시뮬레이션 기반 연구 서명재, 권지연, 김성민 p. 366-375