No abstract

Background Artificial Intelligence (AI) is set to impact several fields within gastroenterology. In gastrointestinal (GI) endoscopy, AI-based tools have translated into clinical practice faster than expected. We aimed to evaluate the status of research for AI in Gastroenterology while predicting its future applications. Methods All studies registered on Clinicaltrials.gov right up to November 2021 were analyzed. Studies using AI in GI-endoscopy, Inflammatory Bowel Diseases (IBD), hepatology and pancreaticobiliary disease were selected. Data regarding study field, methodology, endpoints, and publication status were retrieved, pooled, and analyzed to observe underlying temporal and geographical trends. Results Of the 103 study entries retrieved according to our inclusion/exclusion criteria, 76/103 (74%) were based on AI application to GI-endoscopy, mainly on the detection and characterization of colorectal neoplasia (52/103, 50%). Image analysis was also more frequently reported than data analysis for pancreatico-biliary 6/10 (60%), liver diseases 8/9 (89%) and IBD 6/8 (75%). Overall, 48/103 (47%) study entries were interventional and 55 (53%) observational. In 2018, 1/8 (12.5%) studies were interventional, while in 2021 21/34 (61.8%) were interventional with an inversed ratio between observational and interventional studies in the study time. Majority of the studies were planned as single-centre in 74/103 (72%) geographically located more in Asia 45/103 (44%) and Europe 44/103 (43%). Conclusion AI implementation in gastroenterology is dominated by Computer Aided detection and characterization of colorectal neoplasia. The timeframe of translational research is characterized by a rapid conversion of observational into interventional studies.

No abstract