Wasserstein GANs for MR Imaging: from Paired to Unpaired Training

Abstract

Lack of ground-truth MR images impedes the common supervised training of neural networks for image reconstruction. To cope with this challenge, this paper leverages unpaired adversarial training for reconstruction networks, where the inputs are undersampled k-space and naively reconstructed images from one dataset, and the labels are high-quality images from another dataset. The reconstruction networks consist of a generator which suppresses the input image artifacts, and a discriminator using a pool of (unpaired) labels to adjust the reconstruction quality. The generator is an unrolled neural network -- a cascade of convolutional and data consistency layers. The discriminator is also a multilayer CNN that plays the role of a critic scoring the quality of reconstructed images based on the Wasserstein distance. Our experiments with knee MRI datasets demonstrate that the proposed unpaired training enables diagnostic-quality reconstruction when high-quality image labels are not available for the input types of interest, or when the amount of labels is small. In addition, our adversarial training scheme can achieve better image quality (as rated by expert radiologists) compared with the paired training schemes with pixel-wise loss.

References in corpus (17)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Spectral Normalization for Generative Adversarial Networks
• Image reconstruction by domain transform manifold learning
• Noise2Noise: Learning Image Restoration without Clean Data
• Compressed Sensing MRI Reconstruction using a Generative Adversarial Network with a Cyclic Loss
• Deep learning for undersampled MRI reconstruction
• Amortised MAP Inference for Image Super-resolution
• Banach Wasserstein GAN
• Neural Proximal Gradient Descent for Compressive Imaging
• Deep BCD-Net Using Identical Encoding-Decoding CNN Structures for Iterative Image Recovery
• Self-Supervised Deep Active Accelerated MRI
• Efficient Sum of Outer Products Dictionary Learning (SOUP-DIL) and Its Application to Inverse Problems
• Brain MRI super-resolution using 3D generative adversarial networks
• Deep Learning Methods for Parallel Magnetic Resonance Image Reconstruction
• Highly Scalable Image Reconstruction using Deep Neural Networks with Bandpass Filtering
• How Well Do WGANs Estimate the Wasserstein Metric?
• Compressed Sensing: From Research to Clinical Practice with Data-Driven Learning

Cited by in corpus (10)

• Self-Supervised Learning of Physics-Guided Reconstruction Neural Networks without Fully-Sampled Reference Data
• Adaptive Diffusion Priors for Accelerated MRI Reconstruction
• Multi-Mask Self-Supervised Learning for Physics-Guided Neural Networks in Highly Accelerated MRI
• Artifact- and content-specific quality assessment for MRI with image rulers
• Data and Physics driven Deep Learning Models for Fast MRI Reconstruction: Fundamentals and Methodologies
• Analysis of Deep Complex-Valued Convolutional Neural Networks for MRI Reconstruction
• Plug-and-Play Latent Feature Editing for Orientation-Adaptive Quantitative Susceptibility Mapping Neural Networks
• A Tutorial on MRI Reconstruction: From Modern Methods to Clinical Implications
• Semi-Supervised Learning of Mutually Accelerated MRI Synthesis without Fully-Sampled Ground Truths
• Adaptive Local Neighborhood-based Neural Networks for MR Image Reconstruction from Undersampled Data