Can AI solve radiology's capacity challenge? With Dr. Tessa Cook

In this episode of Scanning the Market: CAIMI Conversations, I sit down with Dr. Tessa Cook, Associate Professor of Radiology at Penn Medicine, to ask a pretty simple question with a much harder answer: Can AI actually solve radiology’s capacity challenge? We get into where the real bottlenecks are today, why the biggest productivity gains may come from everything around image interpretation rather than the image itself, and what tools like generative EHR search, ambient reporting and foundation models could realistically change inside the reading room. We also cover the less glamorous but more important part of AI adoption: workflow friction, trust, monitoring, model drift, governance and whether efficiency gains actually improve life for radiologists or simply result in more work. A few of the areas we cover: Why rising imaging volumes and workforce pressure have created a capacity problem that recruitment alone probably won’t solve Where AI is already saving time in clinical workflows Why non-interpretive tasks may deliver more immediate value than another detection algorithm How generative AI and foundation models could reshape reporting and workflow What good AI governance and post-deployment monitoring should actually look like The balance between productivity, quality and radiologist wellbeing This episode is also part of our wider partnership with SIIM ahead of CAIMI 2026, taking place October 26–27 in Philadelphia. CAIMI is built around exactly these kinds of conversations: what is genuinely working in imaging AI, what still needs to be pressure-tested, and how we move from interesting research into something that actually survives inside a health system. We’ll be releasing more conversations with leaders across imaging AI in the run-up to the event. #ScanningTheMarket #CAIMI2026 #SIIM #RadiologyAI #ImagingAI #HealthcareAI