Computer-aided quality control in colonoscopy: clinical applications and limitations.

Area of Science:

• Gastroenterology and Medical Technology

Background:

• Colonoscopy quality varies, impacting colorectal cancer detection and patient outcomes.
• Objective evaluation and real-time feedback are needed to standardize colonoscopy procedures.

Purpose of the Study:

• To review the clinical utility, implementation barriers, and future of computer-aided quality control (CAQ) systems in colonoscopy.
• To assess the impact of CAQ on quality metrics and patient outcomes.

Main Methods:

• Systematic literature search of PubMed (inception to January 2025).
• Included 66 publications: systematic reviews, meta-analyses, RCTs, cohort, validation, and observational studies.
• Focused on CAQ systems, including artificial intelligence-assisted colonoscopy.

Main Results:

• CAQ systems improve traditional metrics (withdrawal time, bowel prep, cecal intubation rates).
• Artificial intelligence-assisted colonoscopy increased adenoma detection rates (38.5% to 47.9%) and withdrawal time (5.68 to 7.03 min).
• Automated systems show high accuracy in bowel prep scoring (93.3%), cecal intubation recognition (95.5%), and surveillance interval assignment (92.0%).

Conclusions:

• CAQ systems offer transformative potential for standardizing colonoscopy quality and enhancing patient outcomes.
• Addressing implementation challenges (cost, adoption, regulation) is crucial for integration.
• Future research should focus on comparative effectiveness and large-scale clinical integration to reduce colorectal cancer burden.