Extensive research on accelerating Magnetic Resonance Imaging (MRI) has been done on two fronts: (i) hardware acceleration, and (ii) image post processing. We present the results of our work on image post processing, where the input is a sparsely sampled volumetric images, and our deep-learning based models seek to output the original, densely sampled images. Specifically, we propose two different methods at accelerating MRI, in two different aspects of the MR acquisition process.

First, We propose a marginal super-resolution (MSR) approach based on 2D convolutional neural networks (CNNs) for interpolating an anisotropic brain magnetic resonance scan along the highly under-sampled direction. Previous methods for slice interpolation only consider data from pairs of adjacent 2D slices. The possibility of fusing information from the direction orthogonal to the 2D slices remains unexplored. Our approach performs MSR in both sagittal and coronal directions, which provides an initial estimate for slice interpolation. The interpolated slices are then fused and refined in the axial direction for improved consistency. Since MSR consists of only 2D operations, it is more feasible in terms of GPU memory consumption and requires fewer training samples compared to 3D CNNs. Our experiments demonstrate that the proposed method outperforms traditional linear interpolation and baseline 2D/3D CNN-based approaches. We conclude by showcasing the method's practical utility in estimating brain volumes from under-sampled brain MR scans through semantic segmentation.

Secondly, although undersampled MR image recovery has been widely studied for accelerated MR acquisition, it has been mostly studied under a single sequence scenario, despite the fact that multi-sequence MR scan is common in practice. We aim to optimize multi-sequence MR image recovery from undersampled k-space data under an overall time constraint while considering the difference in acquisition time for various sequences. We first formulate it as a constrained optimization problem and then show that finding the optimal sampling strategy for all sequences and the best recovery model at the same time is combinatorial and hence computationally prohibitive. To solve this problem, we propose a blind recovery model that simultaneously recovers multiple sequences, and an efficient approach to find the near-optimal combination of sampling strategy and recovery model. Our experiments demonstrate that the proposed method not only outperforms sequence-wise recovery, but also sheds light on how to optimally undersample the k-space for each sequence within an overall time budget.