Harkness, R., Frangi, A.F. orcid.org/0000-0002-2675-528X, Zucker, K. orcid.org/0000-0003-4385-3153 et al. (1 more author) (2023) Learning disentangled representations for explainable chest x-ray classification using Dirichlet VAEs. In: Išgum, I. and Colliot, O., (eds.) Medical Imaging 2023: Image Processing. SPIE Medical Imaging, 19-24 Feb 2023, San Diego, CA, USA. SPIE ISBN 978-1-5106-6033-5
Abstract
This study explores the use of the Dirichlet Variational Autoencoder (DirVAE) for learning disentangled latent representations of chest X-ray (CXR) images. Our working hypothesis is that distributional sparsity, as facilitated by the Dirichlet prior, will encourage disentangled feature learning for the complex task of multi-label classification of CXR images. The DirVAE is trained using CXR images from the CheXpert database, and the predictive capacity of multi-modal latent representations learned by DirVAE models is investigated through implementation of an auxiliary multi-label classification task, with a view to enforce separation of latent factors according to class-specific features. The predictive performance and explainability of the latent space learned using the DirVAE were quantitatively and qualitatively assessed, respectively, and compared with a standard Gaussian prior-VAE (GVAE). We introduce a new approach for explainable multi-label classification in which we conduct gradient-guided latent traversals for each class of interest. Study findings indicate that the DirVAE is able to disentangle latent factors into class-specific visual features, a property not afforded by the GVAE, and achieve a marginal increase in predictive performance relative to GVAE. We generate visual examples to show that our explainability method, when applied to the trained DirVAE, is able to highlight regions in CXR images that are clinically relevant to the class(es) of interest and additionally, can identify cases where classification relies on spurious feature correlations.
Metadata
Item Type: | Proceedings Paper |
---|---|
Authors/Creators: |
|
Editors: |
|
Keywords: | Chest imaging; Education and training; Lung; Image classification; Opacity; Visualization; Image restoration; Diagnostics; Diseases and disorders; Binary data |
Dates: |
|
Institution: | The University of Leeds |
Academic Units: | The University of Leeds > Faculty of Engineering & Physical Sciences (Leeds) > School of Computing (Leeds) The University of Leeds > Faculty of Medicine and Health (Leeds) > School of Medicine (Leeds) > Leeds Institute of Medical Research (LIMR) > Division of Oncology |
Depositing User: | Symplectic Publications |
Date Deposited: | 01 Sep 2023 12:11 |
Last Modified: | 01 Sep 2023 12:11 |
Status: | Published |
Publisher: | SPIE |
Identification Number: | 10.1117/12.2654345 |
Related URLs: | |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:202909 |