GCN Based Unsupervised Domain Adaptation With Feature Disentanglement For Medical Image Classification

Dwarikanath Mahapatra, steven Korevaar, Ruwan Tennakoon

Keywords: unsupervised domain adaptation, Graph convolution networks, Camelyon17, CheXpert, NIH Xray

TL;DR: Unsupervised domain adaptation using graph adversarial networks and feature disentanglement

Abstract: The success of deep learning has set new benchmarks for many medical image tasks. However, deep models often fail to generalize in the presence of distribution shifts between training (source) data and test (target) data. One method commonly employed to counter distribution shifts is domain adaptation: using samples from the target domain to learn to account for shifted distributions. In this work we propose an unsupervised domain adaptation approach that uses graph neural networks to learn semantic and structural features that are invariant across domains allowing for better performance across distribution shifts. We test the proposed method for classification on two challenging medical image datsets with distribution shifts - multi center chest xray images and histopathology images. Experiments show our method achieves state-of-the-art results on those data sets.

Registration: I acknowledge that publication of this at MIDL and in the proceedings requires at least one of the authors to register and present the work during the conference.

Authorship: I confirm that I am the author of this work and that it has not been submitted to another publication before.

Paper Type: validation/application paper

Primary Subject Area: Transfer Learning and Domain Adaptation

Secondary Subject Area: Detection and Diagnosis

Confidentiality And Author Instructions: I read the call for papers and author instructions. I acknowledge that exceeding the page limit and/or altering the latex template can result in desk rejection.