VS-Net: Variable splitting network for accelerated parallel MRI reconstruction
This work addresses the problem of faster and more accurate MRI scans for medical imaging, though it appears incremental as it builds on existing deep learning and compressed sensing methods.
The paper tackles accelerated parallel MRI reconstruction by proposing VS-Net, a deep learning method that formulates reconstruction as an energy minimization problem and unrolls a variable splitting scheme into a neural network; it shows that VS-Net outperforms state-of-the-art algorithms for 4-fold and 6-fold acceleration on knee images, achieving higher reconstruction accuracy and perceptual quality.
In this work, we propose a deep learning approach for parallel magnetic resonance imaging (MRI) reconstruction, termed a variable splitting network (VS-Net), for an efficient, high-quality reconstruction of undersampled multi-coil MR data. We formulate the generalized parallel compressed sensing reconstruction as an energy minimization problem, for which a variable splitting optimization method is derived. Based on this formulation we propose a novel, end-to-end trainable deep neural network architecture by unrolling the resulting iterative process of such variable splitting scheme. VS-Net is evaluated on complex valued multi-coil knee images for 4-fold and 6-fold acceleration factors. We show that VS-Net outperforms state-of-the-art deep learning reconstruction algorithms, in terms of reconstruction accuracy and perceptual quality. Our code is publicly available at https://github.com/j-duan/VS-Net.