DSA-CycleGAN: A Domain Shift Aware CycleGAN for Robust Multi-Stain Glomeruli Segmentation

TL;DR

This paper introduces DSA-CycleGAN, a novel model that reduces noise in stain translation for histopathology segmentation, improving robustness across different stains.

Contribution

The paper proposes DSA-CycleGAN, which addresses noise issues in stain translation and enhances multi-stain glomeruli segmentation performance.

Findings

DSA-CycleGAN outperforms existing methods in reducing translation noise.

The model improves segmentation accuracy across biologically distinct stains.

Code is publicly available at https://github.com/zeeshannisar/DSA-CycleGAN.

Abstract

A key challenge in segmentation in digital histopathology is inter- and intra-stain variations as it reduces model performance. Labelling each stain is expensive and time-consuming so methods using stain transfer via CycleGAN, have been developed for training multi-stain segmentation models using labels from a single stain. Nevertheless, CycleGAN tends to introduce noise during translation because of the one-to-many nature of some stain pairs, which conflicts with its cycle consistency loss. To address this, we propose the Domain Shift Aware CycleGAN, which reduces the presence of such noise. Furthermore, we evaluate several advances from the field of machine learning aimed at resolving similar problems and compare their effectiveness against DSA-CycleGAN in the context of multi-stain glomeruli segmentation. Experiments demonstrate that DSA-CycleGAN not only improves segmentation performance in glomeruli segmentation but also outperforms other methods in reducing noise. This is particularly evident when translating between biologically distinct stains. The code is publicly available at https://github.com/zeeshannisar/DSA-CycleGAN.