Learning a Variational Network for Reconstruction of Accelerated MRI Data

TL;DR

This paper introduces a variational network that combines mathematical models with deep learning for fast, high-quality MRI reconstruction, outperforming standard methods and suitable for clinical use.

Contribution

The paper presents a novel variational network that learns all parameters of a variational model for accelerated MRI reconstruction, enabling fast and high-quality images without parameter tuning.

Findings

Outperforms standard algorithms in image quality and artifact reduction

Reconstruction time of approximately 193 ms on a single GPU

Preserves natural image appearance and unseen pathologies

Abstract

Purpose: To allow fast and high-quality reconstruction of clinical accelerated multi-coil MR data by learning a variational network that combines the mathematical structure of variational models with deep learning. Theory and Methods: Generalized compressed sensing reconstruction formulated as a variational model is embedded in an unrolled gradient descent scheme. All parameters of this formulation, including the prior model defined by filter kernels and activation functions as well as the data term weights, are learned during an offline training procedure. The learned model can then be applied online to previously unseen data. Results: The variational network approach is evaluated on a clinical knee imaging protocol. The variational network reconstructions outperform standard reconstruction algorithms in terms of image quality and residual artifacts for all tested acceleration factors and sampling patterns. Conclusion: Variational network reconstructions preserve the natural appearance of MR images as well as pathologies that were not included in the training data set. Due to its high computational performance, i.e., reconstruction time of 193 ms on a single graphics card, and the omission of parameter tuning once the network is trained, this new approach to image reconstruction can easily be integrated into clinical workflow.