Domain-adversarial neural networks to address the appearance variability of histopathology images

TL;DR

This paper introduces a domain-adversarial neural network approach to improve the generalization of histopathology image analysis across different labs and conditions, outperforming traditional normalization methods.

Contribution

The paper presents a systematic domain-adversarial training method that effectively reduces appearance variability in histopathology images for better model generalization.

Findings

Domain-adversarial training improves generalization in histopathology analysis.

Combining color augmentation with domain-adversarial training outperforms standard normalization.

Method shows superior results in mitosis detection across different datasets.

Abstract

Preparing and scanning histopathology slides consists of several steps, each with a multitude of parameters. The parameters can vary between pathology labs and within the same lab over time, resulting in significant variability of the tissue appearance that hampers the generalization of automatic image analysis methods. Typically, this is addressed with ad-hoc approaches such as staining normalization that aim to reduce the appearance variability. In this paper, we propose a systematic solution based on domain-adversarial neural networks. We hypothesize that removing the domain information from the model representation leads to better generalization. We tested our hypothesis for the problem of mitosis detection in breast cancer histopathology images and made a comparative analysis with two other approaches. We show that combining color augmentation with domain-adversarial training is a better alternative than standard approaches to improve the generalization of deep learning methods.