Artificial Intelligence in Colonoscopy: Where Are We Now in 2024?

Area of Science:

• Gastroenterology
• Medical Imaging
• Artificial Intelligence

Background:

• Colonoscopy is vital for colorectal cancer (CRC) prevention.
• Artificial intelligence (AI) systems are being developed to enhance colonoscopy quality assurance (computer-aided quality improvement [CAQ]), lesion detection (computer-aided detection [CADe]), and lesion characterization (computer-aided characterization [CADx]).

Purpose of the Study:

• To review the current evidence on AI applications in colonoscopy.
• To discuss existing limitations and explore future directions of AI in colonoscopy.

Main Methods:

• State-of-the-art narrative review of existing studies on AI in colonoscopy.

Main Results:

• CAQ systems show inconsistent results due to small datasets and varied methods.
• CADe systems improve adenoma detection but raise concerns about false positives and cost.
• CADx systems have mixed accuracy and are limited by data representation and trial rigor.

Conclusions:

• AI-assisted colonoscopy shows potential but faces challenges for real-world implementation.
• Further research is needed on clinical efficacy, cost-effectiveness, liability, and data sharing.