Wave-encoding and Shuffling Enables Rapid Time Resolved Structural Imaging

Abstract

Purpose. T 2 -Shuffling reconstructs multiple sharp T 2 -weighted images from a single volumetric fast spin-echo (3D-FSE) scan. Wave-CAIPI is a parallel imaging technique that achieves good reconstruction at high accelerations through additional sinusoidal gradients that induce a voxel spreading effect in the readout direction to better take advantage of coilsensitivity information. In this work, the Shuffling model in T 2 -Shuffling is augmented with wave-encoding to achieve higher acceleration capability. The resulting "Wave-Shuffling" approach is applied to 3D-FSE and Magnetization-Prepared Rapid Gradient-Echo (MPRAGE) to achieve rapid, 1 mm-isotropic resolution, time-resolved structural imaging.Theory and Methods. Wave-encoding is incorporated into the Shuffling model and the wave gradient encoding parameters are optimized for the target time-resolved imaging application for higher acceleration capability. 3D-FSE and MPRAGE sequences are modified to enable sinusoidal wave-gradients and random sub-sampling of k-space. In-vivo Shuffling and Wave-Shuffling under-sampling MPRAGE experiments are performed to demonstrate the better reconstruction enabled by the wave-encoding. In-vivo Wave-Shuffling 3D-FSE data are also acquired and reconstructed. A wave gradient calibration method to account for system hardware imperfections suitable for Wave-Shuffling acquisitions with random undersampling and without a densely-sampled, low-resolution calibration region is proposed.Results. For brain imaging application with 32-channel coil at 3T, Wave-Shuffling MPRAGE provides comparable reconstruction at ≈ 3× higher acceleration compared to Shuffling.Wave-Shuffling is successfully applied to MPRAGE and 3D-FSE to recover high-quality time-series of images. Conclusion.Wave-Shuffling enables rapid, time-resolved structural imaging.

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

“…Low-rank subspace and shuffling methods have emerged as powerful methods for reconstructing time-resolved MR images and qMRI since they incorporate low-rank subspace bases that are calculated from Bloch equations. 20,21,[32][33][34][35][36] T 2 -shuffling, for instance, showed multi-contrast and sharp T 2 -weighted images by leveraging the T 2 relaxation during the fast spin echo (FSE) readout using a low-rank subspace method and shuffled k-space data acquisition. 33 3D-EPTI acquires highly undersampled k-t data using an inversion-recovery gradient-echo (IR-GE) and a variable flip angle gradient and spin-echo (VFA-GRASE) and also exploits a low-rank subspace approach to reconstruct time-series data efficiently.…”

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

“…Efficient time-resolved MRI enables a wide range of applications such as MRS, 1 quantitative parameter mapping, 2,3 motion-resolved imaging, 4,5 and blur-free, multi-contrast imaging from fast-spin-echo 6 (FSE) and MPRAGE 7,8 acquisitions. In this paper, we focus on applications that reconstruct individual echo images with differing contrasts from acquisitions with evolving signal evolution over an echo train.…”

“…In this paper, we focus on applications that reconstruct individual echo images with differing contrasts from acquisitions with evolving signal evolution over an echo train. 7 Time-resolved MRI requires lengthy scan times using traditional methods. Compressed sensing 9,10 and parallel imaging 11,12 reduce acquisition times in structural MRI, with some time-resolved applications.…”