ASGE Annual Postgraduate Course: Clinical Challenges and Real-Time Decision Making in GI Endoscopy (On-demand) | May 2023-Endoscopic imaging

Video Transcription

Video Summary

In this video, Yuichi discusses the practical steps for building an AI healthcare application in a program. He highlights three options for implementation: clinical practice usage, clinical research, or AI development for commercialization. He explains that for clinical practice usage, regulatory approval is necessary to ensure the device's safety and efficacy. The current status of regulatory approval for AI devices in endoscopy is presented, with the G.I. Genius by Medtronic being the only AI device available in the States. Yuichi emphasizes the importance of constructing a research consortium and distributing work packages for clinical research. He suggests collaboration between gastroenterologists, computer vision researchers, and industries. The process of collecting learning material is discussed, including retrospective and prospective databases, as well as the use of open access databases. Informed consent processes and accurate labeling are important considerations. For clinical validation, Yuichi suggests benchmark tests before conducting prospective trials. He also references guidelines for reporting study results and constructing protocols. Finally, Yuichi shares his personal experience with commercialization, recommending collaboration with experts, pilot studies, securing funding, and potential partnerships with industrial partners or launching startups. He illustrates this with examples of AI devices that have obtained regulatory approval in Japan. The time and cost factors involved in regulatory clearance are mentioned, with the need for substantial funding and time investment.

Asset Subtitle

Yuichi Mori,MD, PhD, FASGE

Keywords

AI healthcare application

regulatory approval

clinical research

learning material

commercialization