Clinical Efficacy of Real-Time Artificial Intelligence-Assisted Colonoscopy in Colorectal Polyp Detection: A...

Area of Science:

• Gastroenterology
• Medical Imaging
• Oncology

Background:

• Early detection of colon polyps is crucial for preventing colorectal cancer.
• Computer-aided detection (CADe) systems aim to improve polyp detection rates (PDR) during colonoscopy.
• This study evaluates the efficacy of a CADe system in detecting colorectal neoplasms.

Purpose of the Study:

• To assess the impact of a CADe system on colorectal neoplasm detection rates.
• To compare the adenoma detection rate (ADR) between CADe-assisted and conventional colonoscopy.
• To identify factors influencing adenoma detection, with a focus on CADe system efficacy.

Main Methods:

• A prospective, randomized controlled trial involving 998 patients across two tertiary centers.
• Patients were randomized to either CADe-assisted or conventional colonoscopy.
• Primary endpoint was ADR; secondary endpoints included PDR, adenomas per colonoscopy (APC), and relative risk (RR) of polyp detection.

Main Results:

• The CADe group demonstrated significantly higher PDR (72.2% vs 54.5%) and ADR (52.3% vs 36.1%) compared to the conventional group.
• CADe use resulted in higher polyp counts (2.2 vs 1.4 per colonoscopy) and APC values (1.2 vs 0.8).
• CADe was the strongest independent predictor of adenoma detection (odds ratio, 1.914), surpassing factors like patient age and sex.

Conclusions:

• Real-time CADe-assisted colonoscopy significantly enhances both polyp and adenoma detection rates.
• The CADe system is a powerful independent predictor of adenoma detection.
• This technology holds significant promise for improving the effectiveness of colonoscopy in colorectal cancer prevention.