PT  - JOURNAL ARTICLE
AU  - Harry T. Mason
AU  - Nadine N. Graedel
AU  - Karla L. Miller
AU  - Mark Chiew
TI  - Subspace-constrained approaches to low-rank fMRI acceleration
AID  - 10.1101/2020.12.15.422908
DP  - 2020 Jan 01
TA  - bioRxiv
PG  - 2020.12.15.422908
4099  - http://biorxiv.org/content/early/2020/12/16/2020.12.15.422908.short
4100  - http://biorxiv.org/content/early/2020/12/16/2020.12.15.422908.full
AB  - Acceleration methods in fMRI aim to reconstruct high fidelity images from undersampled k-space, allowing fMRI datasets to achieve higher temporal resolution, reduced physiological noise aliasing, and increased statistical degrees of freedom. While low levels of acceleration are typically part of standard fMRI protocols through parallel imaging, there exists the potential for approaches that allow much greater acceleration. One such existing approach is k-t FASTER, which exploits the inherent low-rank nature of fMRI. In this paper, we present a reformulated version of k-t FASTER which includes additional L2 constraints within a low-rank framework.We evaluated the effect of three different constraints against existing low-rank approaches to fMRI reconstruction: Tikhonov constraints, low-resolution priors, and temporal subspace smoothness. The different approaches are separately tested for robustness to undersampling and thermal noise levels, in both retrospectively and prospectively-undersampled finger-tapping task fMRI data. Reconstruction quality is evaluated by accurate reconstruction of low-rank subspaces and activation maps.The use of L2 constraints were found to achieve consistently improved results, producing high fidelity reconstructions of statistical parameter maps at higher acceleration factors and lower SNR values than existing methods, but at a cost of longer computation time. In particular, the Tikhonov constraint proved very robust across all tested datasets, and the temporal subspace smoothness constraint provided the best reconstruction scores in the prospectively-undersampled dataset. These results demonstrate that regularized low-rank reconstruction of fMRI data can recover functional information at high acceleration factors without the use of any model-based spatial constraints.Highlights We introduce an alternate implementation of low-rank fMRI reconstruction by using alternating minimization, which allows for easy integration of the subspace-specific L2 constraintsWe use the alternating minimization approach to accelerate FMRI by exploiting coil sensitivity, low-rank structures, and additional L2 constraintsWe found Tikhonov and Temporal Subspace Smoothness constraints show improved performance over other methods for R=15-30Tikhonov Constraints were the most robust of the constrained-subspace methods, with the shortest reconstruction timeTemporal Subspace Smoothness produced the highest reconstruction scores in the prospectively under-sampled dataCompeting Interest StatementThe authors have declared no competing interest.