Can I read Deep Learning Applied to Chest X-Rays: Exploiting and Preventing Shortcuts on EtoBox?

Deep Learning Applied to Chest X-Rays: Exploiting and Preventing Shortcuts by Jabbour, Sarah; Fouhey, David; Kazerooni, Ella; Sjoding, Michael W.; Wiens, Jenna is a scholarly article available to read on EtoBox.

What is Deep Learning Applied to Chest X-Rays: Exploiting and Preventing Shortcuts about?

While deep learning has shown promise in improving the automated diagnosis of disease based on chest X-rays, deep networks may exhibit undesirable behavior related to shortcuts. This paper studies the case of spurious class skew in which patients with a particular attribute are spuriously more likely to have the outcome of interest. For instance, clinical protocols might lead to a dataset in which patients with pacemakers are disproportionately likely to have congestive heart failure. This skew can lead to models that take shortcuts by heavily relying on the biased attribute. We explore this problem across a number of attributes in the context of diagnosing the cause of acute hypoxemic respiratory failure. Applied to chest X-rays, we show that i) deep nets can accurately identify many patient attributes including sex (AUROC = 0.96) and age (AUROC >= 0.90), ii) they tend to exploit correlations between such attributes and the outcome label when learning to predict a diagnosis, leading to poor performance when such correlations do not hold in the test population (e.g., everyone in the test set is male), and iii) a simple transfer learning approach is surprisingly effective at prevent

Author

Jabbour, Sarah; Fouhey, David; Kazerooni, Ella; Sjoding, Michael W.; Wiens, Jenna

Published

2020

Language

EN