Description

This multi-part workshop introduces the concept of algorithmic fairness and its significance in healthcare. In Part 1 we introduce the concept of data equity and discuss how principles of data equity arise in every phase of the research process. We delve into algorithmic bias in Part 2, illustrating via several real-world examples how it can manifest. In Part 3 we formally define the notion of fairness and discuss tradeoffs between fairness and accuracy. Parts 1 through 3 constitute Session 1.

We describe tools that can be used to identify and address bias in Part 4. In Part 5 we conclude by introducing an open-source Python package, Fairlearn, that can be used to identify and address bias in statistical models. Parts 4 and 5 constitute Session 2.

Parts 1 through 4 assume some awareness of predictive models in the healthcare setting but are introductory. Part 5 assumes some experience with Python. This workshop is intended for clinicians, researchers, educators, and others who are interested in algorithmic fairness.

Register via Compass. Workshop will be recorded.

Questions? Contact Amy Laird: laird@ohsu.edu

If you have a disability and need an accommodation to attend or participate in this event, please contact Amy Laird (laird@ohsu.edu) at least five business days before the event.