Domain-adversarial neural networks to address the appearance variability of histopathology images
This addresses the issue of domain shift in histopathology for pathologists and researchers, but it is incremental as it builds on existing domain-adversarial methods.
The paper tackled the problem of appearance variability in histopathology images, which hampers generalization of automatic analysis, by proposing domain-adversarial neural networks combined with color augmentation, showing it improves generalization for mitosis detection in breast cancer images compared to standard approaches.
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.