Conditional WGANs with Adaptive Gradient Balancing for Sparse MRI Reconstruction

TL;DR

This paper introduces a novel Conditional WGAN with Adaptive Gradient Balancing to improve the quality of sparse MRI reconstructions, producing sharper images with fewer artifacts compared to existing methods.

Contribution

The paper proposes a new combination of Conditional WGANs and Adaptive Gradient Balancing to enhance MRI image reconstruction quality and training stability.

Findings

Sharper MRI images with fewer artifacts

Improved training stability of GANs in MRI reconstruction

Superior image quality compared to existing techniques

Abstract

Recent sparse MRI reconstruction models have used Deep Neural Networks (DNNs) to reconstruct relatively high-quality images from highly undersampled k-space data, enabling much faster MRI scanning. However, these techniques sometimes struggle to reconstruct sharp images that preserve fine detail while maintaining a natural appearance. In this work, we enhance the image quality by using a Conditional Wasserstein Generative Adversarial Network combined with a novel Adaptive Gradient Balancing technique that stabilizes the training and minimizes the degree of artifacts, while maintaining a high-quality reconstruction that produces sharper images than other techniques.