Deep, Deep Learning with BART
This work provides a general framework for reproducible deep-learning-based reconstruction in MRI, but it is incremental as it builds upon existing tools and methods.
The authors tackled the problem of developing a deep-learning-based image reconstruction framework for MRI by extending the BART toolbox with a non-linear operator framework for automatic differentiation, enabling the implementation and training of state-of-the-art networks like the Variational Network and MoDL with similar performance to TensorFlow-based implementations.
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.