Scaling AI Monitoring in Clinical Care: Platforms, Processes and Problems

Presentation Time: 09:45 AM - 11:00 AM

Abstract Keywords: Artificial Intelligence, Fairness and elimination of bias, Evaluation, Workflow
Primary Track: Applications
Programmatic Theme: Clinical Informatics
Moderator: Sara Murray, MD, MAS
UCSF

Artificial intelligence (AI) has the potential to transform healthcare delivery, but as AI becomes more integrated into clinical care, ensuring these tools remain safe and effective is essential. While some organizations assess AI trustworthiness at initial implementation, performance and workflows often evolve over time. Few health systems have established robust processes for continuous, real-world monitoring. Without a standardized approach to AI oversight at scale, most are left to navigate these challenges independently.
This panel will explore key frameworks and emerging best practices for AI monitoring, addressing both technical and operational barriers. Panelists will share experiences with a range of solutions, including innovative homegrown platforms and commercial tools. The discussion will highlight the importance of both platforms and processes in building a successful AI monitoring strategy for diverse care settings. Finally, panelists will examine common challenges in AI monitoring and offer practical strategies for implementing scalable oversight solutions.

Speaker(s):
Mark Sendak, MD, MPP
Duke Institute for Health Innovation

Peter Embi, MD
VUMC

Marylyn Ritchie, PhD
University of Pennsylvania, Perelman School of Medicine

Jinoos Yazdany, MD MPH
UCSF

Author(s):
Sara Murray, MD, MAS - UCSF; Mark Sendak, MD, MPP - Duke Institute for Health Innovation; Marylyn Ritchie, PhD - University of Pennsylvania, Perelman School of Medicine; Peter Embi, MD - VUMC; Jinoos Yazdany, MD MPH - UCSF;