End-to-end autoencoding architecture for the simultaneous generation of medical images and corresponding segmentation masks
This addresses the problem of insufficient training data for medical image segmentation, offering a solution for researchers and practitioners in medical imaging, though it appears incremental as it builds on existing variational autoencoder methods.
The paper tackles the challenge of generating diverse and realistic medical images and segmentation masks under data-scarce conditions by proposing an end-to-end architecture based on the Hamiltonian Variational Autoencoder, which outperforms generative adversarial architectures in image quality and precise tumor mask synthesis on datasets like BRATS and HECKTOR.
Despite the increasing use of deep learning in medical image segmentation, acquiring sufficient training data remains a challenge in the medical field. In response, data augmentation techniques have been proposed; however, the generation of diverse and realistic medical images and their corresponding masks remains a difficult task, especially when working with insufficient training sets. To address these limitations, we present an end-to-end architecture based on the Hamiltonian Variational Autoencoder (HVAE). This approach yields an improved posterior distribution approximation compared to traditional Variational Autoencoders (VAE), resulting in higher image generation quality. Our method outperforms generative adversarial architectures under data-scarce conditions, showcasing enhancements in image quality and precise tumor mask synthesis. We conduct experiments on two publicly available datasets, MICCAI's Brain Tumor Segmentation Challenge (BRATS), and Head and Neck Tumor Segmentation Challenge (HECKTOR), demonstrating the effectiveness of our method on different medical imaging modalities.