ASGE at DDW AI in GI Workshop | May 2022-AI and Endoscopy Training: Good or Bad?

Pdf Summary

The document discusses the applications and potential benefits of artificial intelligence (AI) in the field of endoscopy. It highlights the challenges faced in modern endoscopy, such as cost, difficulty in interpretation, and lack of expertise, and suggests that AI can help address these challenges. The document also mentions that AI has made progress in five key areas of gastroenterology, including computer-aided detection, computer-aided diagnosis, new endoscopic quality indicators, natural language processing/analytics, and precision medicine.<br /><br />There are several examples given of AI applications in endoscopy, such as the Colonoscopy Coverage Deficiency via Depth algorithm (C2D2) that can improve colonoscopy coverage by computing a depth image from a single RGB image. The document also mentions several AI-supported tools and technologies developed by companies like Olympus, Pentax, Medtronic, and Fuji.<br /><br />The document discusses the potential for AI to be embedded in the training and accreditation processes for endoscopists, as well as its role in improving performance auditing. It suggests that AI can be used to support and automate the audit of endoscopist performance. The document also highlights the need for scalable solutions that can have a rapid and widespread impact on gastrointestinal (GI) practice.<br /><br />The document includes the results of a study that found that endoscopists' diagnostic accuracy improved when they were supported by AI. It suggests that the use of AI can result in increased accuracy and safety in the diagnosis and decision-making process.<br /><br />Overall, the document presents AI as a promising technology in the field of endoscopy and highlights the potential benefits it can bring in terms of improving diagnostics, training, and performance auditing.

Asset Subtitle

Michael Byrne, MD

Keywords

artificial intelligence

endoscopy

challenges

gastroenterology

computer-aided detection

computer-aided diagnosis

endoscopic quality indicators

precision medicine

diagnostic accuracy

performance auditing