Editable Stain Transformation Of Histological Images Using Unpaired GANs
This addresses the need for cost-effective and non-damaging alternatives in histopathology for metaplastic breast cancer diagnosis, though it is incremental as it builds on existing GAN methods.
The study tackled the problem of generating P63-like histological images from H&E stains to avoid tissue damage and high costs, achieving results where generated images were often comparable to actual ones in realism according to a histopathologist survey.
Double staining in histopathology, particularly for metaplastic breast cancer, typically employs H&E and P63 dyes. However, P63's tissue damage and high cost necessitate alternative methods. This study introduces xAI-CycleGAN, an advanced architecture combining Mask CycleGAN with explainability features and structure-preserving capabilities for transforming H&E stained breast tissue images into P63-like images. The architecture allows for output editing, enhancing resemblance to actual images and enabling further model refinement. We showcase xAI-CycleGAN's efficacy in maintaining structural integrity and generating high-quality images. Additionally, a histopathologist survey indicates the generated images' realism is often comparable to actual images, validating our model's high-quality output.