Unsupervised Domain Adaptation with Semantic Consistency across Heterogeneous Modalities for MRI Prostate Lesion Segmentation

TL;DR

This paper proposes a novel unsupervised domain adaptation method for MRI prostate lesion segmentation that maintains semantic consistency across heterogeneous imaging modalities, improving target domain performance.

Contribution

It introduces a semantic cycle-consistency loss and pseudo-labeling strategy to adapt models across different MRI modalities, addressing heterogeneity in medical imaging domains.

Findings

Significant performance improvements over existing methods on VERDICT-MRI.

Effective transfer of labeled mp-MRI data to enhance VERDICT-MRI segmentation.

Method outperforms in semi-supervised and supervised settings.

Abstract

Any novel medical imaging modality that differs from previous protocols e.g. in the number of imaging channels, introduces a new domain that is heterogeneous from previous ones. This common medical imaging scenario is rarely considered in the domain adaptation literature, which handles shifts across domains of the same dimensionality. In our work we rely on stochastic generative modeling to translate across two heterogeneous domains at pixel space and introduce two new loss functions that promote semantic consistency. Firstly, we introduce a semantic cycle-consistency loss in the source domain to ensure that the translation preserves the semantics. Secondly, we introduce a pseudo-labelling loss, where we translate target data to source, label them by a source-domain network, and use the generated pseudo-labels to supervise the target-domain network. Our results show that this allows us to extract systematically better representations for the target domain. In particular, we address the challenge of enhancing performance on VERDICT-MRI, an advanced diffusion-weighted imaging technique, by exploiting labeled mp-MRI data. When compared to several unsupervised domain adaptation approaches, our approach yields substantial improvements, that consistently carry over to the semi-supervised and supervised learning settings.