Diagnosing AI

​Join Dr. Milan Toma, author of Diagnosing AI: Evaluation of AI in Clinical Practice, for a live session exploring one of the most pressing questions in modern medicine: how do we actually know if an AI system is ready for clinical use?

​As AI tools become more visible in healthcare settings, the confidence with which they present information can be misleading. Fluent, authoritative-sounding output does not equal diagnostic accuracy, and this session draws directly from the empirical studies and frameworks in the book to show why that distinction matters. Dr. Toma will walk through how large language models create the illusion of expertise, where multimodal models fall short in medical image interpretation, and what a rigorous, clinically grounded evaluation process should actually look like.

​Whether you work in clinical practice, health informatics, data science, or healthcare administration, this is a chance to engage directly with the ideas in the book and ask questions about applying them in real-world settings. The book itself (ISBN 979-8999832481, published by Dawning Research Press) covers everything from traditional machine learning algorithms to modern transformers, with practical guidance on evaluation protocols, training-validation curves, and the economic factors that shape AI deployment decisions. Come ready to think critically.

​This event is part of #AIWeekNY by Pulse NYC — a community-led festival celebrating innovation across the AI ecosystem. More at https://pulse.nyc/ai-week/

​Homepage: www.tomamil.com

​YouTube: www.youtube.com/@drmilantoma