Zero-DeepSub: Zero-shot deep subspace reconstruction for rapid multiparametric quantitative MRI using 3D-QALAS.

PURPOSE: To develop and evaluate methods for (1) reconstructing 3D-quantification using an interleaved Look-Locker acquisition sequence with T2 preparation pulse (3D-QALAS) time-series images using a low-rank subspace method, which enables accurate and rapid T1 and T2 mapping, and (2) improving the fidelity of subspace QALAS by combining scan-specific deep-learning-based reconstruction and subspace modeling. THEORY AND METHODS: A low-rank subspace method for 3D-QALAS (i.e., subspace QALAS) and zero-shot deep-learning subspace method (i.e., Zero-DeepSub) were proposed for rapid and high fidelity T1 and T2 mapping and time-resolved imaging using 3D-QALAS. Using an ISMRM/NIST system phantom, the accuracy and reproducibility of the T1 and T2 maps estimated using the proposed methods were evaluated by comparing them with reference techniques. The reconstruction performance of the proposed subspace QALAS using Zero-DeepSub was evaluated in vivo and compared with conventional QALAS at high reduction factors of up to nine-fold. RESULTS: Phantom experiments showed that subspace QALAS had good linearity with respect to the reference methods while reducing biases and improving precision compared to conventional QALAS, especially for T2 maps. Moreover, in vivo results demonstrated that subspace QALAS had better g-factor maps and could reduce voxel blurring, noise, and artifacts compared to conventional QALAS and showed robust performance at up to nine-fold acceleration with Zero-DeepSub, which enabled whole-brain T1, T2, and PD mapping at 1 mm isotropic resolution within 2 min of scan time. CONCLUSION: The proposed subspace QALAS along with Zero-DeepSub enabled high fidelity and rapid whole-brain multiparametric quantification and time-resolved imaging.

Stochastic-offset-enhanced restricted slice excitation and 180° refocusing designs with spatially non-linear ΔB0 shim array fields.

PURPOSE: Developing a general framework with a novel stochastic offset strategy for the design of optimized RF pulses and time-varying spatially non-linear ΔB0 shim array fields for restricted slice excitation and refocusing with refined magnetization profiles within the intervals of the fixed voxels. METHODS: Our framework uses the decomposition property of the Bloch equations to enable joint design of RF-pulses and shim array fields for restricted slice excitation and refocusing with auto-differentiation optimization. Bloch simulations are performed independently on orthogonal basis vectors, Mx, My, and Mz, which enables designs for arbitrary initial magnetizations. Requirements for refocusing pulse designs are derived from the extended phase graph formalism obviating time-consuming sub-voxel isochromatic simulations to model the effects of crusher gradients. To refine resultant slice-profiles because of voxelwise optimization functions, we propose an algorithm that stochastically offsets spatial points at which loss is computed during optimization. RESULTS: We first applied our proposed design framework to standard slice-selective excitation and refocusing pulses in the absence of non-linear ΔB0 shim array fields and compared them against pulses designed with Shinnar-Le Roux algorithm. Next, we demonstrated our technique in a simulated setup of fetal brain imaging in pregnancy for restricted-slice excitation and refocusing of the fetal brain. CONCLUSIONS: Our proposed framework for optimizing RF pulse and time-varying spatially non-linear ΔB0 shim array fields achieve high fidelity restricted-slice excitation and refocusing for fetal MRI, which could enable zoomed fast-spin-echo-MRI and other applications.

Latent signal models: Learning compact representations of signal evolution for improved time-resolved, multi-contrast MRI.

PURPOSE: To improve time-resolved reconstructions by training auto-encoders to learn compact representations of Bloch-simulated signal evolution and inserting the decoder into the forward model. METHODS: Building on model-based nonlinear and linear subspace techniques, we train auto-encoders on dictionaries of simulated signal evolution to learn compact, nonlinear, latent representations. The proposed latent signal model framework inserts the decoder portion of the auto-encoder into the forward model and directly reconstructs the latent representation. Latent signal models essentially serve as a proxy for fast and feasible differentiation through the Bloch equations used to simulate signal. This work performs experiments in the context of T2 -shuffling, gradient echo EPTI, and MPRAGE-shuffling. We compare how efficiently auto-encoders represent signal evolution in comparison to linear subspaces. Simulation and in vivo experiments then evaluate if reducing degrees of freedom by incorporating our proxy for the Bloch equations, the decoder portion of the auto-encoder, into the forward model improves reconstructions in comparison to subspace constraints. RESULTS: An auto-encoder with 1 real latent variable represents single-tissue fast spin echo, EPTI, and MPRAGE signal evolution to within 0.15% normalized RMS error, enabling reconstruction problems with 3 degrees of freedom per voxel (real latent variable + complex scaling) in comparison to linear models with 4-8 degrees of freedom per voxel. In simulated/in vivo T2 -shuffling and in vivo EPTI experiments, the proposed framework achieves consistent quantitative normalized RMS error improvement over linear approaches. From qualitative evaluation, the proposed approach yields images with reduced blurring and noise amplification in MPRAGE-shuffling experiments. CONCLUSION: Directly solving for nonlinear latent representations of signal evolution improves time-resolved MRI reconstructions.

Scan-specific artifact reduction in k-space (SPARK) neural networks synergize with physics-based reconstruction to accelerate MRI.

PURPOSE: To develop a scan-specific model that estimates and corrects k-space errors made when reconstructing accelerated MRI data. METHODS: Scan-specific artifact reduction in k-space (SPARK) trains a convolutional-neural-network to estimate and correct k-space errors made by an input reconstruction technique by back-propagating from the mean-squared-error loss between an auto-calibration signal (ACS) and the input technique's reconstructed ACS. First, SPARK is applied to generalized autocalibrating partially parallel acquisitions (GRAPPA) and demonstrates improved robustness over other scan-specific models, such as robust artificial-neural-networks for k-space interpolation (RAKI) and residual-RAKI. Subsequent experiments demonstrate that SPARK synergizes with residual-RAKI to improve reconstruction performance. SPARK also improves reconstruction quality when applied to advanced acquisition and reconstruction techniques like 2D virtual coil (VC-) GRAPPA, 2D LORAKS, 3D GRAPPA without an integrated ACS region, and 2D/3D wave-encoded imaging. RESULTS: SPARK yields SSIM improvement and 1.5 - 2× root mean squared error (RMSE) reduction when applied to GRAPPA and improves robustness to ACS size for various acceleration rates in comparison to other scan-specific techniques. When applied to advanced reconstruction techniques such as residual-RAKI, 2D VC-GRAPPA and LORAKS, SPARK achieves up to 20% RMSE improvement. SPARK with 3D GRAPPA also improves RMSE performance by ~2×, SSIM performance, and perceived image quality without a fully sampled ACS region. Finally, SPARK synergizes with non-Cartesian, 2D and 3D wave-encoding imaging by reducing RMSE between 20% and 25% and providing qualitative improvements. CONCLUSION: SPARK synergizes with physics-based acquisition and reconstruction techniques to improve accelerated MRI by training scan-specific models to estimate and correct reconstruction errors in k-space.