Fast model‐based T₂ mapping using SAR‐reduced simultaneous multislice excitation

Abstract

Purpose To obtain whole‐brain high‐resolution T2 maps in 2 minutes by combining simultaneous multislice excitation and low‐power PINS (power independent of number of slices) refocusing pulses with undersampling and a model‐based reconstruction. Methods A multi‐echo spin‐echo sequence was modified to acquire multiple slices simultaneously, ensuring low specific absorption rate requirements. In addition, the acquisition was undersampled to achieve further acceleration. Data were reconstructed by subsequently applying parallel imaging to separate signals from different slices, and a model‐based reconstruction to estimate quantitative T2 from the undersampled data. The signal model used is based on extended phase graph simulations that also account for nonideal slice profiles and B1 inhomogeneity. In vivo experiments with 3 healthy subjects were performed to compare accelerated T2 maps to fully sampled single‐slice acquisitions. The accuracy of the T2 values was assessed with phantom experiments by comparing the T2 values to single‐echo spin‐echo measurements. Results In vivo results showed that conventional multi‐echo spin‐echo, simultaneous multislice, and undersampling result in similar mean T2 values within regions of interest. However, combining simultaneous multislice and undersampling results in higher SDs (about 7 ms) in comparison to a conventional sequence (about 3 ms). The T2 values were reproducible between scan and rescan (SD < 1.2 ms) within subjects and were in similar ranges across subjects (SD < 4.5 ms). Conclusion The proposed method is a fast T2 mapping technique that enables whole‐brain acquisitions at 0.7‐mm in‐plane resolution, 3‐mm slice thickness, and low specific absorption rate in 2 minutes.

Fast and high‐resolution myelin water imaging: Accelerating multi‐echo GRASE with CAIPIRINHA

“…In vivo experiments showed that the proposed sequence yields MWF maps comparable with those derived from a reference MESE sequence (r = 0.83, mean bias = −0.2%), while providing increased brain coverage (multi-echo GRASE: 1.6 × 1.6 × 1.6 mm 3 , 84 slices; MESE: 1 × 1 × 3 mm 3 , single slice) in a TA of 10:30 min. The MESE sequence could be further accelerated with parallel imaging and the brain coverage extended using simultaneous multi-slice schemes, however, at the expense of introducing magnetization transfer contamination of the T 2 decays across slices, 38,50 imperfect slice profiles, 39 and increased SAR. Conversely, the 3D nature of the proposed multi-echo GRASE sequence allows to reduce the slice thickness to achieve isotropic resolution while simplifying F I G U R E 4 A, Axial brain slice acquired at the first echo time for different undersampling CAIPIRINHA schemes along with the RMSE between the T 2 decays acquired with the R y × R z (Δkz) = 2 × 2 (1) scheme and every other undersampled acquisition.…”

Improvement of Fast Model-Based Acceleration of Parameter Look-Locker T1 Mapping

“…In order to get better time of computational we decided to choose a simple and efficient method of model fitting. Other solutions, in some cases are more efficient but are also numerically advanced, and require more time to reconstruct the final image [16][17][18][19][27][28][29]33]. In the literature we did not find any benchmark that enables us to compare methods of T 1 mapping reconstruction and quality of results of different works for similar data.…”

“…Acceleration using GPU and implementation of some methods in C/CUDA (Compute Unified Device Architecture) generates a time complexity of 10-20 min [19,33] and in the range from minutes to hours depending on data size [18]. Fast multi-slice method for T 2 mapping [29] calculates 50 slices on an office computer within 7 h, which gives approximately 9 min per slice or alternatively rapid T 1 quantification [28] reports reconstruction time of approximately 10 min per slice. We showed that a single iteration of the FIR-MAP for all projections could take 30 s. The data preparation of initial model for all projections took an additional 2 min for IFM and 1 min for MFM.…”

“…Zibetti et al provides comparison and evaluation of 12 different types of CS sparsity for acceleration of T 1 mapping [27]. More recent methods use improvements of previous algorithms by means of total-generalized-variations (TGV) based regularization and further adapted to a multiparametric setting [28] or split-slice GRAPPA and a model-based iterative algorithm for T 2 -mapping [29]. Different approach bases on the method of magnetic resonance fingerprinting [30,31] in which the benefit comes from simultaneous computation of T 1 and T 2 maps but in slightly longer acquisition time.…”

“…Most of the current works operates on a single slice, which from a clinical point of view is not applicable. The limited methods of multi-slice parameter mapping have been used in few works [28,29,33] resulting in calculation time from 10 min [28] up to 7 h [29] in multi-slice analysis. In multi-channel systems the number of coils may be limited in the preprocessing step.…”