Unsupervised Cross-Modality Domain Adaptation of ConvNets for Biomedical Image Segmentations with Adversarial Loss

TL;DR

This paper introduces an unsupervised adversarial domain adaptation framework for biomedical image segmentation, effectively bridging cross-modality differences without requiring target domain labels.

Contribution

It proposes a novel plug-and-play domain adaptation module and a domain critic module to align features across modalities using adversarial learning.

Findings

Successfully adapted MRI-trained ConvNets to CT data for cardiac segmentation

Achieved promising results without using target domain labels

Demonstrated effectiveness in cross-modality biomedical image segmentation

Abstract

Convolutional networks (ConvNets) have achieved great successes in various challenging vision tasks. However, the performance of ConvNets would degrade when encountering the domain shift. The domain adaptation is more significant while challenging in the field of biomedical image analysis, where cross-modality data have largely different distributions. Given that annotating the medical data is especially expensive, the supervised transfer learning approaches are not quite optimal. In this paper, we propose an unsupervised domain adaptation framework with adversarial learning for cross-modality biomedical image segmentations. Specifically, our model is based on a dilated fully convolutional network for pixel-wise prediction. Moreover, we build a plug-and-play domain adaptation module (DAM) to map the target input to features which are aligned with source domain feature space. A domain critic module (DCM) is set up for discriminating the feature space of both domains. We optimize the DAM and DCM via an adversarial loss without using any target domain label. Our proposed method is validated by adapting a ConvNet trained with MRI images to unpaired CT data for cardiac structures segmentations, and achieved very promising results.