Jenna Wiens Abstract and download – Stanford – Department of Biomedical Data Science

Download Abstract and readings here: Seminar Series, Flyer- Jenna Wiens

Title: White Coat, Black Box – Navigating Pitfalls in Using AI to Augment Diagnostic Decision Making

Jenna Wiens
Associate Professor of Computer Science & Engineering, University of Michigan

October 19, 2023 1:30PM-2:50PM MSOB X303

Abstract:

AI tools designed to aid clinicians in complex diagnostic decisions have the potential to enhance treatment selection and consequently improve patient outcomes. Despite their promise, simply applying existing AI approaches carries a significant risk of inadvertently perpetuating or even exacerbating biases present in clinical care. In this talk, I will describe our work in developing AI systems to diagnose common causes of acute respiratory failure (i.e., pneumonia, heart failure and/or chronic obstructive pulmonary disease), highlighting challenges and presenting potential strategies to mitigate the risk of these systems replicating harmful biases.

Reading list:

If the students aren’t yet familiar with Grad-CAM this would be a good paper for them to read: https://arxiv.org/abs/1610.02391

This paper from ICLR 2021 is also relevant: https://openreview.net/forum?id=mNtmhaDkAr

I’m going to use diagnosing the underlying etiology of acute respiratory failure as the running example throughout the talk. Students can learn more about the task here: https://arxiv.org/abs/2108.12530

For more information on the BMDS 280 Seminars please visit: