From Vision to Precision: The Rise of Artificial Intelligence in Gastrointestinal Endoscopy

Abstract Artificial intelligence (AI) in gastrointestinal endoscopy in recent years has been considered a disruptor in technological innovations. From providing a level playing field irrespective of the experience of the individual endoscopist to providing precision diagnosis, AI has paved the way for a new revolution with the advent of convolutional neural networks and temporal convolutional networks that simplify the analysis of complex data. In this narrative review, we attempt to unwind the interlacing relationship between medicine and technology in endoscopy. AI has played a significant role in the identification and classification of colonic polyps, gastric dysplasia, periampullary neoplasms and cholangiocarcinoma, thus making it an important tool in the diagnostic armamentarium for the prevention of malignancies. AI has consistently performed well in ensuring quality adherence, completion, and monitoring of endoscopic procedures. In therapeutically challenging procedures such as endoscopic mucosal resection and submucosal dissection, AI acts as a mentor to facilitate precision and reduce complications, especially among novices. Diagnosis, monitoring, and assessment of the severity of inflammatory bowel diseases have been made more objective and reproducible by AI. However, drawing the line in attributing responsibilities to AI-based technologies is an ethical dilemma that is being faced today. Understanding the limitations of entirely AI-based technology is important to ensure that there is no overdependence. As of today, these innovations are becoming an indispensable part of endoscopy; however, the role of informed decision-making remains with the astute clinician.