Top-Down Meets Bottom-Up: Enterprise AI Governance Lessons Learned From Radiology

Deploying artificial intelligence (AI) in healthcare requires more than technology, it demands coordinated governance that bridges the department and enterprise. This session explores how radiology, often at the forefront of clinical AI adoption, can serve as a model for integrating top-down and bottom-up governance approaches.

At the department level, radiologists, as the frontline domain experts, are uniquely positioned to identify meaningful real-world use cases, assess model performance in daily practice, and understand when AI fails. However, these bottom-up insights must align with enterprise-level strategies and processes, including infrastructure, integration, safety, compliance, privacy, cybersecurity, and financial sustainability. The session will highlight practical strategies for aligning departmental initiatives with organizational oversight, ensuring that innovation can prosper while supporting organizational priorities.

A special focus will be placed on human factors engineering and human-AI interaction principles, recognizing that successful deployments hinge on how clinicians engage with technology. Designing interfaces and workflows that engender trust, increase efficiency, and create awareness of risks such as automation bias and algorithm neglect, can maximize adoption and return on investment.

Attendees will gain a framework for building governance pathways that combine clinical expertise with enterprise accountability, fostering teamwork that accelerates safe, effective, and sustainable AI integration across healthcare systems.