Join us for this presentation that will look at the challenges of implementing Artificial Intelligence in a hospital setting, concentrating on integrating AI into clinical workflows in a safe and effective manner.
Topics discussed:
-How to identify and address shortcuts or confounders which allow AI tools to perform well in test datasets while showing dangerous behavior in real-world applications
- How AI models generalize between different hospitals
Brudno will draw upon several examples of ML models deployed at UHN, including vision models for laparoscopic surgery, Machine Learning medical directives for making low-risk decisions in pediatric Emergency Rooms, and estimating cardiovascular risk for patients who have undergone a liver transplant.
Michael Brudno is a Professor in the Department of Computer Science at the University of Toronto, as well as the Chief Data Scientist at the University Health Network (UHN). He is also a CIFAR AI Chair at the Vector Institute. His main research interest focuses on the capture of structured phenotypic data from clinical encounters, using both refined User Interfaces, and Large Language Models to streamline clinical workflows and enable faster and better treatments. After receiving a BA in Computer Science and History from UC Berkeley, Brudno received his PhD from the Computer Science Department of Stanford University, working on genomics and bioinformatics. He then completed a postdoctoral fellowship at UC Berkeley and was a Visiting Scientist at MIT. He has been awarded a Sloan Research Fellowship, a Canada Research Chair, and a CIFAR AI Chair, among other honors.
Add to Calendar2025-12-10 18:30:002025-12-10 20:00:00BCHSI Seminar Series: “All Clinical AI Deployment is Local"
Guest Speaker:
Michael Brudno, PhD, professor of Computer Science, University of Toronto
Seminar title: “All Clinical AI Deployment is Local"
Wednesday, Dec. 10, 2025
-Networking and Refreshments: 10:30am
-Talk: 11 – 12 pm
Genentech Hall, Byers Auditorium
Join us for this presentation that will look at the challenges of implementing Artificial Intelligence in a hospital setting, concentrating on integrating AI into clinical workflows in a safe and effective manner.
Topics discussed:
- How to identify and address shortcuts or confounders which allow AI tools to perform well in test datasets while showing dangerous behavior in real-world applications
- How AI models generalize between different hospitals
Brudno will draw upon several examples of ML models deployed at UHN, including vision models for laparoscopic surgery, Machine Learning medical directives for making low-risk decisions in pediatric Emergency Rooms, and estimating cardiovascular risk for patients who have undergone a liver transplant.
Michael Brudno is a Professor in the Department of Computer Science at the University of Toronto, as well as the Chief Data Scientist at the University Health Network (UHN). He is also a CIFAR AI Chair at the Vector Institute. His main research interest focuses on the capture of structured phenotypic data from clinical encounters, using both refined User Interfaces, and Large Language Models to streamline clinical workflows and enable faster and better treatments. After receiving a BA in Computer Science and History from UC Berkeley, Brudno received his PhD from the Computer Science Department of Stanford University, working on genomics and bioinformatics. He then completed a postdoctoral fellowship at UC Berkeley and was a Visiting Scientist at MIT. He has been awarded a Sloan Research Fellowship, a Canada Research Chair, and a CIFAR AI Chair, among other honors.
Calendar Link | Questions? Contact [email protected]Division of Clinical Informatics and Digital TransformationAmerica/Los_Angelespublic