This virtual, three-session, hands-on course introduces the practical foundations of AI/ML for biomedical research, moving from core concepts to clinical data pipelines, model training, evaluation, interpretation, and deployment.
Course workbooks
Session 1 workbookSession 2 workbookSession 3 workbookTo use the workbooks, you will need a Google account. Any email associated with Google, a gmail account not required. Please confirm the workbooks open for you before the course opens. If you are unfamiliar with a Jupyter notebook, take a quick look here.
Session recordings
- build a shared vocabulary around modern AI,
- work through Python notebooks using clinical datasets including chest x-rays of pneumonia,
- train a model end-to-end on a clinical task, and
- learn how to assess whether model performance is helpful, robust, and aligned with real-world clinical use.
No prior machine learning experience is required; the course is designed for investigators who want to understand how AI systems are built, evaluated, and responsibly translated into clinical research settings.
Virtual sessions
Wednesdays, 10am-11am ET, July 22, July 29, August 5
Taught by
Samuel Friedman, PhD, Broad Institute
Additional Faculty
Jamie Fairclough, PhD, Dartmouth
Joseph Frassica, MD, Hood Pediatric Innovation Hub at MIT
Rafael Fricks, PhD, US Department of Veterans Affairs
Mahnaz Maddah, PhD, Broad Institute