Reflections on CAIMI 2026

“The only way to predict the future is to invent it.” 

Dr. Pari Pandharipande, Chair of Radiology at the University of Pennsylvania, opened this year’s SIIM Conference on AI in Medical Imaging (CAIMI) with this quote. It was a fitting introduction to two days of discussions about the future of AI in medical imaging. 

My biggest takeaway was the growing emphasis on translating AI research into clinical practice. Foundation models and AI agents are accelerating the development of new applications, while raising important questions about validation, patient safety, and governance. 

The inaugural AI Builder Showcase was a welcome addition, giving participants an opportunity to demonstrate working applications and prototypes addressing clinical and operational challenges. It was encouraging to see practical development featured alongside traditional scientific research. 

We also held a SIIM Board meeting during CAIMI. One important decision was to establish a new SIIM Training Committee, reimagining the existing SIIM Training Advisory Council. The committee will oversee SIIM’s training programs, coordinate training materials, and support the development of competency-based validation. Over time, we hope to offer certificates that allow participants to demonstrate their knowledge and skills in specific areas of imaging informatics. 

Dr. Nabile Safdar, Chief AI Officer at Emory University and Emory Healthcare, closed CAIMI with a presentation about how the pace of change differs across technology, regulation, finances, and culture. Sometimes these timelines align, creating opportunities for progress. Other times, one area moves ahead while others struggle to keep up. His closing slide referenced A Wrinkle in Time: 

“Time folds. Be in position when it does.” 

Nabile’s message brought us back to where Pari started the meeting. We have an opportunity to shape the future of imaging AI, but we also need to prepare for it and advocate for the changes we want to see. 

How can SIIM help our members prepare for the future of imaging AI? Share your suggestions through this form.