Translation of Artificial Intelligence in Colonoscopy

Area of Science:

• Gastroenterology
• Medical Imaging
• Artificial Intelligence

Background:

• Artificial intelligence (AI) shows rapid advancement in gastroenterology, particularly colonoscopy.
• Colonoscopy's high procedure volume and quality variability make it ideal for AI benefits.
• This review summarizes AI updates, benefits, and challenges in clinical gastroenterology.

Purpose of the Study:

• To review the latest advancements in AI for gastroenterology.
• To assess the current benefits and challenges of AI implementation in clinical practice.
• To outline future directions for AI in gastroenterology.

Main Methods:

• Review of current literature on AI in gastroenterology.
• Analysis of randomized controlled trials for AI systems.
• Evaluation of implementation challenges and future research needs.

Main Results:

• Computer-aided detection (CAD) systems significantly improve adenoma detection rates in colonoscopy.
• CAD systems face real-world implementation hurdles, including reimbursement and workflow integration.
• AI shows promise in inflammatory bowel disease (IBD) for disease scoring and relapse prediction.
• Computer-aided diagnosis and quality systems are in earlier stages with variable results.

Conclusions:

• AI adoption hinges on workflow integration, human-AI interaction, cost-effectiveness, and clinician endorsement.
• Addressing fairness, accountability, and bias is crucial for successful AI implementation.
• Future AI applications may include multi-modal systems, personalized surveillance, and therapeutic interventions.