Session Description

This presentation and discussion focused on enhancing the transparency and accountability of AI models using model cards. Model cards serve as standardized documentation capturing essential details about AI systems, including their purpose, performance metrics, ethical considerations, and limitations. By providing a structured inventory with comprehensive information, the initiative aims to foster trust among stakeholders, including developers, users, and regulators. This talk highlighted the importance of transparency in AI development to mitigate biases, ensure fairness, and improve decision-making. It aimed to discuss the challenges of creating and maintaining model cards, such as balancing transparency with proprietary concerns and ensuring accuracy. By promoting best practices, the talk advocated for widespread adoption of model cards as a step toward more responsible and ethical AI deployment.

Video

view slides

Speakers

Michael Burns, MD, PhD | Assistant Professor, Anesthesiology & Assistant Director, Informatics and Data Analytics, University of Michigan

Michael Burns, MD, PhD is an Assistant Professor of Anesthesiology and the Director of Informatics and Artificial Intelligence at the University of Michigan. He serves as the Director of the Michigan Medicine Clinical Intelligence Committee and Associate Chief Medical Informatics Officer for Artificial Intelligence. He is the Co-chair the Clinical Intelligence Committee at Michigan Medicine and the Assistant Director for Implementation Operations within Precision Health, leading research, and governance of artificial intelligence (AI) projects. Dr. Burns’ scholarly activities foster advancement of clinical AI tools and supports the development of a clinical AI infrastructure with emphasis on medical analytics, clinical operations, and perioperative healthcare. He utilizes machine learning, natural language processing, and novel data science methods to understand perioperative clinical patterns, studying methods to identify and improve healthcare.

Michael J. Sheppard | Data Foundations Director, Michigan Medicine & Co-Chair, Michigan Medicine Clinical Intelligence Committee

Michael Sheppard, MSIS, is the Data Foundations Director at Michigan Medicine, University of Michigan. He is the founder of the Data Innovation Community of Practice and is the Co-chair of the Michigan Medicine Clinical Intelligence Committee at Michigan.  The Clinical Intelligence Committee is responsible for AI governance, education, consultation and AI enablement across the health system.  Michael leads strategic efforts that focus on building robust IT systems at the intersection of AI and enterprise data management.

other health AI hubs