Disentanglement enables cross-domain Hippocampus Segmentation

TL;DR

This paper introduces a novel GAN-based method for disentangling MRI images into content and domain features, enabling effective domain transfer and improved hippocampus segmentation across different data sources.

Contribution

The paper proposes the Content Domain Disentanglement GAN, a new approach for separating content and domain in MRI images to enhance cross-domain hippocampus segmentation.

Findings

Improved segmentation performance by 6-13% on unseen domains.

Outperforms existing domain transfer methods.

Enables better generalization in medical image segmentation.

Abstract

Limited amount of labelled training data are a common problem in medical imaging. This makes it difficult to train a well-generalised model and therefore often leads to failure in unknown domains. Hippocampus segmentation from magnetic resonance imaging (MRI) scans is critical for the diagnosis and treatment of neuropsychatric disorders. Domain differences in contrast or shape can significantly affect segmentation. We address this issue by disentangling a T1-weighted MRI image into its content and domain. This separation enables us to perform a domain transfer and thus convert data from new sources into the training domain. This step thus simplifies the segmentation problem, resulting in higher quality segmentations. We achieve the disentanglement with the proposed novel methodology 'Content Domain Disentanglement GAN', and we propose to retrain the UNet on the transformed outputs to deal with GAN-specific artefacts. With these changes, we are able to improve performance on unseen domains by 6-13% and outperform state-of-the-art domain transfer methods.