Diffusion-Weighted Magnetic Resonance Brain Images Generation with Generative Adversarial Networks and Variational Autoencoders: A Comparison Study
This addresses data scarcity and inhomogeneity in medical imaging, enabling better data augmentation for restricted access scenarios, though it is incremental as it compares existing methods.
The study tackled the problem of generating high-quality, diverse diffusion-weighted MRI brain images using deep generative models, specifically Introspective Variational Autoencoder and Style-Based GAN, and found they qualified for data augmentation in medical fields based on neuroradiologist evaluations and metrics.
We show that high quality, diverse and realistic-looking diffusion-weighted magnetic resonance images can be synthesized using deep generative models. Based on professional neuroradiologists' evaluations and diverse metrics with respect to quality and diversity of the generated synthetic brain images, we present two networks, the Introspective Variational Autoencoder and the Style-Based GAN, that qualify for data augmentation in the medical field, where information is saved in a dispatched and inhomogeneous way and access to it is in many aspects restricted.