StainGAN: Stain Style Transfer for Digital Histological Images
This addresses the issue of color variations in histological diagnosis, which can hinder accuracy, but it is incremental as it builds on existing CycleGAN methods for a specific domain.
The paper tackled the problem of stain color variations in digital histological images by proposing a deep-learning solution based on CycleGANs, which eliminated the need for a reference template slide and showed a 10% improvement in SSIM and a 12% increase in AUC for breast cancer tumor classification.
Digitized Histological diagnosis is in increasing demand. However, color variations due to various factors are imposing obstacles to the diagnosis process. The problem of stain color variations is a well-defined problem with many proposed solutions. Most of these solutions are highly dependent on a reference template slide. We propose a deep-learning solution inspired by CycleGANs that is trained end-to-end, eliminating the need for an expert to pick a representative reference slide. Our approach showed superior results quantitatively and qualitatively against the state of the art methods (10% improvement visually using SSIM). We further validated our method on a clinical use-case, namely Breast Cancer tumor classification, showing 12% increase in AUC. The code will be made publicly available.