ASGE Postgraduate Course at ACG: Innovative Practices for the Endoscopist of Today, Tomorrow and Beyond (On-demand) | October 2023-13

Pdf Summary

Dr. Tyler Berzin’s lecture at the Center for Advanced Endoscopy, Beth Israel Deaconess Medical Center, delves into the evolving role of Artificial Intelligence (AI) in gastrointestinal endoscopy. He opens by demystifying AI concepts and explaining their relevance to the field, addressing common AI myths and how AI differs from traditional programming. Dr. Berzin highlights that machine learning (ML) and deep learning (DL) are pivotal in medical advancements but acknowledges their current limitations, such as lack of continuous learning and common sense.<br /><br />The lecture underlines the initial applications of AI in GI endoscopy, notably Computer Aided Polyp Detection (CADe), which assists with identifying lesions during procedures. CADe aims to mitigate human variability in polyp detection, improving adenoma detection rates (ADR) and polypectomy rates, thereby enhancing patient outcomes. Despite these benefits, AI adoption in clinical settings remains slow due to economic considerations, lack of quality mandates, and the tech adoption curve.<br /><br />Dr. Berzin argues that while AI technologies like CADe provide point solutions, the future of AI in endoscopy lies in integrated platforms and generalist foundation models. These systems will transcend single-modality use, engaging with multimodal data (e.g., video, imaging, and electronic health records) to offer comprehensive support during procedures. He envisions AI evolving into a medical co-pilot, offering real-time assistance and step-by-step reasoning based on extensive medical knowledge.<br /><br />In his roadmap for AI in GI endoscopy, Dr. Berzin suggests several key areas of development: lesion measurement, computer-aided diagnosis, intraoperative navigation, exposure/prep assessment, and semi-automated reporting. The ultimate goal is to shift from point solutions to AI-enabled system changes, increasing efficiency and reducing variability in clinical practice.<br /><br />The lecture concludes by urging practitioners to consider AI adoption through the lenses of safety, economics, and innovation, framing it as a significant step toward enhancing GI practice's quality and positioning it at the forefront of medical innovation.

Meta Tag

Concept Artificial Intelligence

Concept Machine Learning

Concept Computer-Aided Detection

Concept Colorectal Polyp Detection

Concept Deep learning

Keywords

Artificial Intelligence

Gastrointestinal Endoscopy

Machine Learning

Deep Learning

Computer Aided Polyp Detection

Adenoma Detection Rates

Clinical Practice

Medical Co-pilot

Intraoperative Navigation

Medical Innovation

Artificial Intelligence

Machine Learning

Computer-Aided Detection

Colorectal Polyp Detection

Deep learning