Clinical significance of computer-aided quality assessment systems in colonoscopy: a comprehensive review
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Colonoscopy is the primary tool for colorectal cancer screening. High-quality colonoscopy is crucial for the detection of precancerousadenomas; however, the adenoma detection rate varies depending on the skill and experience of the endoscopist. Computer-aidedquality assessment (CAQ) uses artificial intelligence (AI) technology to evaluate the quality of colonoscopy examinations. It plays animportant role in reducing variations in examination quality and obtaining high-quality colonoscopic images. In this review, we focusspecifically on the speedometer, effective withdrawal time, fold examination quality, bowel preparation quality assessment, and cecalintubation with CAQ systems and discuss the role and effectiveness of these systems. CAQ systems are expected to contribute to increasein adenoma detection rates, improvement in endoscopist skills, and standardization of examination quality. However, challengessuch as variability in AI performance across different clinical settings and potential overreliance on automated prompts remain keylimitations.
Abstract
Colonoscopy is the primary tool for colorectal cancer screening. High-quality colonoscopy is crucial for the detection of precancerousadenomas; however, the adenoma detection rate varies depending on the skill and experience of the endoscopist. Computer-aidedquality assessment (CAQ) uses artificial intelligence (AI) technology to evaluate the quality of colonoscopy examinations. It plays animportant role in reducing variations in examination quality and obtaining high-quality colonoscopic images. In this review, we focusspecifically on the speedometer, effective withdrawal time, fold examination quality, bowel preparation quality assessment, and cecalintubation with CAQ systems and discuss the role and effectiveness of these systems. CAQ systems are expected to contribute to increasein adenoma detection rates, improvement in endoscopist skills, and standardization of examination quality. However, challengessuch as variability in AI performance across different clinical settings and potential overreliance on automated prompts remain keylimitations.