Deep, Deep Learning with BART

TL;DR

This paper introduces an extension of the BART toolbox with a non-linear operator framework that enables deep learning-based MRI reconstruction, matching the performance of existing neural network implementations.

Contribution

The authors extend BART with automatic differentiation and neural network components, facilitating deep learning-based MRI reconstruction within a unified framework.

Findings

BART can implement and train state-of-the-art deep reconstruction networks.

Performance is comparable to TensorFlow-based implementations.

The framework supports advanced MRI reconstruction algorithms.

Abstract

Purpose: To develop a deep-learning-based image reconstruction framework for reproducible research in MRI. Methods: The BART toolbox offers a rich set of implementations of calibration and reconstruction algorithms for parallel imaging and compressed sensing. In this work, BART was extended by a non-linear operator framework that provides automatic differentiation to allow computation of gradients. Existing MRI-specific operators of BART, such as the non-uniform fast Fourier transform, are directly integrated into this framework and are complemented by common building blocks used in neural networks. To evaluate the use of the framework for advanced deep-learning-based reconstruction, two state-of-the-art unrolled reconstruction networks, namely the Variational Network [1] and MoDL [2], were implemented. Results: State-of-the-art deep image-reconstruction networks can be constructed and trained using BART's gradient based optimization algorithms. The BART implementation achieves a similar performance in terms of training time and reconstruction quality compared to the original implementations based on TensorFlow. Conclusion: By integrating non-linear operators and neural networks into BART, we provide a general framework for deep-learning-based reconstruction in MRI.