Integration of Endocuff-Assisted and Computer-Aided Colonoscopy: A Meta-Analysis of Randomized Controlled Trials

Area of Science:

• Gastroenterology
• Oncology
• Medical Imaging

Background:

• Colorectal cancer (CRC) remains a leading cause of cancer mortality, with missed lesions during colonoscopy contributing to poor outcomes.
• AI-assisted computer-aided detection (CADe) systems show promise in reducing adenoma miss rates.
• Integrating CADe with advanced visualization tools like EndoCuff vision may further improve adenoma detection.

Purpose of the Study:

• To evaluate the effectiveness of combining CADe with EndoCuff vision-assisted colonoscopy (EAC).
• To compare the adenoma detection rate (ADR) of CADe+EAC against standard colonoscopy and CADe alone.
• To assess the impact on advanced adenoma and sessile serrated lesion detection.

Main Methods:

• A meta-analysis of randomized controlled trials (RCTs) was conducted.
• Literature search included MEDLINE, Embase, and clinicaltrials.gov up to November 2024.
• Four RCTs involving 3333 participants were analyzed for adenoma detection rate (ADR) and other metrics.

Main Results:

• CADe+EAC significantly improved ADR (RR: 1.36) and advanced ADR (RR: 1.62) compared to standard colonoscopy.
• The combination also showed higher detection rates for sessile serrated lesions (RR: 1.95) and mean adenomas per colonoscopy (MD: 0.50).
• Compared to CADe alone, CADe+EAC further enhanced ADR (RR: 1.13) and mean adenomas per colonoscopy (MD: 0.15).

Conclusions:

• The combination of CADe and EAC significantly enhances adenoma detection compared to standard colonoscopy and CADe alone.
• This synergistic approach improves the detection of precancerous lesions, potentially reducing CRC mortality.
• Further multi-center trials are recommended to validate these findings with diverse AI models.