Old Dogs Can Learn New Tricks: Artificial Intelligence Improves Adenoma Detection Rates in Screening Colonoscopies in...

Area of Science:

• Gastroenterology and Endoscopy
• Artificial Intelligence in Medicine
• Colorectal Cancer Screening

Background:

• High adenoma detection rate (ADR) in screening colonoscopy is linked to reduced post-colonoscopy colorectal cancer (CRC) incidence.
• Artificial intelligence (AI) is emerging as a tool to enhance adenoma identification during colonoscopy.

Purpose of the Study:

• To evaluate if AI-powered computer-aided detection (CADe) systems improve ADR during colonoscopy.
• To determine if CADe yields the greatest ADR increase among low-volume and junior endoscopists.

Main Methods:

• Retrospective review of ADR before and after CADe implementation at a single, high-volume endoscopy center.
• Inclusion of board-certified colorectal surgeons and gastroenterologists with diverse experience levels.
• Utilized CADe software integrated with colonoscopy equipment.

Main Results:

• Overall mean ADR significantly increased from 53.9% to 61.3% (p=0.02) post-CADe implementation.
• Colorectal surgeons showed a significant ADR increase (49% to 62%, p=0.02), while gastroenterologists did not (56.9% to 60.8%, p=0.15).
• Senior endoscopists (51.7% to 59.6%, p=0.02) and high-volume endoscopists (51.3% to 59.4%, p=0.01) experienced significant ADR increases, unlike junior and low-volume groups.

Conclusions:

• AI-powered CADe in colonoscopy is associated with increased ADR, particularly benefiting experienced endoscopists.
• Further studies are needed to ascertain if this ADR improvement impacts overall CRC incidence.
• Limitations include a single institution and a limited study duration.