Discriminative Hamiltonian Variational Autoencoder for Accurate Tumor Segmentation in Data-Scarce Regimes
This addresses data scarcity in medical imaging segmentation, which is crucial for clinical applications, though it appears incremental as it builds on existing VAE and data augmentation approaches.
The paper tackles the problem of medical image segmentation with limited annotated training data by proposing a discriminative Hamiltonian Variational Autoencoder that generates realistic images and masks to augment datasets, achieving improved segmentation accuracy on BRATS and HECKTOR datasets.
Deep learning has gained significant attention in medical image segmentation. However, the limited availability of annotated training data presents a challenge to achieving accurate results. In efforts to overcome this challenge, data augmentation techniques have been proposed. However, the majority of these approaches primarily focus on image generation. For segmentation tasks, providing both images and their corresponding target masks is crucial, and the generation of diverse and realistic samples remains a complex task, especially when working with limited training datasets. To this end, we propose a new end-to-end hybrid architecture based on Hamiltonian Variational Autoencoders (HVAE) and a discriminative regularization to improve the quality of generated images. Our method provides an accuracte estimation of the joint distribution of the images and masks, resulting in the generation of realistic medical images with reduced artifacts and off-distribution instances. As generating 3D volumes requires substantial time and memory, our architecture operates on a slice-by-slice basis to segment 3D volumes, capitilizing on the richly augmented dataset. Experiments conducted on two public datasets, BRATS (MRI modality) and HECKTOR (PET modality), demonstrate the efficacy of our proposed method on different medical imaging modalities with limited data.