Histopathology DatasetGAN: Synthesizing Large-Resolution Histopathology Datasets
This work addresses a bottleneck for medical imaging researchers by enabling more efficient deep learning on label-scarce, high-resolution histopathology datasets, though it is incremental as an extension of an existing framework.
The paper tackles the challenge of scaling self-supervised learning to high-resolution medical images by proposing the Histopathology DatasetGAN (HDGAN) framework, which reduces memory consumption and demonstrates strong performance on a thrombotic microangiopathy dataset.
Self-supervised learning (SSL) methods are enabling an increasing number of deep learning models to be trained on image datasets in domains where labels are difficult to obtain. These methods, however, struggle to scale to the high resolution of medical imaging datasets, where they are critical for achieving good generalization on label-scarce medical image datasets. In this work, we propose the Histopathology DatasetGAN (HDGAN) framework, an extension of the DatasetGAN semi-supervised framework for image generation and segmentation that scales well to large-resolution histopathology images. We make several adaptations from the original framework, including updating the generative backbone, selectively extracting latent features from the generator, and switching to memory-mapped arrays. These changes reduce the memory consumption of the framework, improving its applicability to medical imaging domains. We evaluate HDGAN on a thrombotic microangiopathy high-resolution tile dataset, demonstrating strong performance on the high-resolution image-annotation generation task. We hope that this work enables more application of deep learning models to medical datasets, in addition to encouraging more exploration of self-supervised frameworks within the medical imaging domain.