Implementation of AI in Barrett’s surveillance endoscopy: current status, challenges and the road ahead

Endoscopic surveillance for Barrett’s neoplasia is performed to detect neoplasia at an early stage, when minimally invasive therapy can still achieve curative outcomes. Endoscopic detection of subtle, early neoplasia is, however, challenging. Artificial intelligence (AI) offers a potential solution by providing real-time detection and characterisation of neoplastic Barrett’s lesions, independent of operator experience. This review provides a comprehensive overview of the current status, challenges and future directions for AI implementation in Barrett’s surveillance endoscopy.