Event Speaker

Bradley Erickson, MD, PhD, CIIP, FSIIM

Professor

Mayo Clinic

Dr. Erickson is a neuroradiologist who splits his time between clinical practice and directing the Mayo Clinic AI Lab. The lab has a number of post-doctoral fellows and graduate students, and proejcts include a number of tools for radiological images, but also histopathology images and skin images. We also develop core technologies for AI and also develop large language models for research purposes. He has been president of SIIM, chair of the ABII, Associate chair of research at Mayo Clinic. He is involved with 3 start companies, including Enquanta, FlowSIGMA, and Yunu.

Sessions

Bytes and Rights: A Debate on the Ethics of AI Data Use

Presented by Judy Gichoya, <br> MD, MS, FSIIM, Eliot Siegel, MD, FSIIM, FACR, Bradley Erickson, MD, PhD, CIIP, FSIIM May 23, 2025 | 8:15 AMOregon Ballroom 201

In this session, there will be a debate to explore the complexities and consequences of leveraging health data within academia and industry for research, product development, and services. Attendees will gain a nuanced understanding of the balance between economic benefits and privacy concerns, and explore innovative approaches to ethical data monetization. This debate promises to…

Closing General Session + Annual Meeting: Awards AI vs. Human: The Ultimate Showdown

Presented by Ali S. Tejani, MD, Christopher Roth, MD, MMCI, CIIP, FACR, FSIIM, Eliot Siegel, MD, FSIIM, FACR, Tessa S. Cook, <br>MD, PhD, CIIP, FSIIM, Richard Wiggins, MD, FSIIM, John Groth, MD, Veronica Rotemberg, MD, PhD, Bradley Erickson, MD, PhD, CIIP, FSIIM, Adam Flanders, MD, CIIP, FSIIM, Katherine P. Andriole, <br>PhD, FSIIM, Nikki Fennell, CIIP May 23, 2025 | 3:00 PM

From Chaos to Clarity – An LLM Learning Lab

Presented by Ali Ganjizadeh, MD, Bradley Erickson, MD, PhD, CIIP, FSIIM May 21, 2025 | 4:00 PMRoom B117

This interactive learning lab focuses on improving the consistency and reliability of AI-generated responses in radiology and clinical workflows. Participants will explore how to structure AI outputs using Pydantic models and error feedback loops, ensuring accurate and well-organized data. Additionally, the session introduces DSPy, a user-friendly tool that helps optimize how AI models process and…