Deep learning models for COVID-19 chest x-ray classification: Preventing shortcut learning using feature disentanglement.

In response to the COVID-19 global pandemic, recent research has proposed creating deep learning based models that use chest radiographs (CXRs) in a variety of clinical tasks to help manage the crisis. However, the size of existing datasets of CXRs from COVID-19+ patients are relatively small, and researchers often pool CXR data from multiple sources, for example, using different x-ray machines in various patient populations under different clinical scenarios. Deep learning models trained on such datasets have been shown to overfit to erroneous features instead of learning pulmonary characteristics in a phenomenon known as shortcut learning. We propose adding feature disentanglement to the training process. This technique forces the models to identify pulmonary features from the images and penalizes them for learning features that can discriminate between the original datasets that the images come from. We find that models trained in this way indeed have better generalization performance on unseen data; in the best case we found that it improved AUC by 0.13 on held out data. We further find that this outperforms masking out non-lung parts of the CXRs and performing histogram equalization, both of which are recently proposed methods for removing biases in CXR datasets.
Competing Interests: The authors have read the journal’s policy and have the following competing interests to declare: AYL reports support from the US Food and Drug Administration, grants from Santen, Regeneron, Carl Zeiss Meditec, and Novartis, and personal fees from Genentech, Roche, Verana Health, and Johnson and Johnson, outside of the submitted work. This article does not reflect the opinions of the Food and Drug Administration. CR, AT, AO, RD, and JMLF are paid employees of the Microsoft Corporation. JMLF also receives personal fees from Singularity University as a speaker. JD and SG are paid employees of IRIS. In addition, SG is a consultant/advisor for Alcon Laboratories, Allergan, Inc., Andrews Institute, GENENTECH, Novartis, Alcon Pharmaceuticals, Regeneron Pharmaceuticals, Inc., Roche Diagnostics, and Spark Therapeutics, Inc. as well as an equity owner in Intelligent Retinal Imaging Systems (IRIS), Retina Specialty Institute, and USRetina. This does not alter the authors’ adherence to PLOS ONE policies on sharing data and materials. There are no patents, products in development or marketed products associated with this research to declare. MB, PKB, WCL, and JKC have no relevant financial or non-financial interests to disclose.