Artificial intelligence in capsule endoscopy for detection of ulcers and erosions

Capsule endoscopy (CE) has revolutionized the diagnosis of small bowel (SB) disorders by allowing noninvasive visualization of the entire SB and colonic mucosa. However, several limitations remain in the diagnosis and monitoring of Crohn’s disease (CD) due to the low specificity of CE and need for manual review of the entire study. Artificial intelligence (AI) has many potential roles, including automated and improved detection of inflammatory lesions, diagnosis and monitoring of CD, evaluation for dysplasia and malignancy, prediction of fibrostenotic complications, and auto-documentation of reports. There is emerging data with proof-of-concept studies on the utility of AI in CD. These concepts and algorithms will require robust testing and validation before they can be reliably applied in clinical practice.