Reduction of B₁ sensitivity in selective single‐slab 3d turbo spin echo imaging with very long echo trains

Abstract

Single-slab 3D turbo/fast spin echo (SE) imaging with very long echo trains was recently introduced with slab selection using a highly selective excitation pulse and short, nonselective refocusing pulses with variable flip angles for high imaging efficiency. This technique, however, is vulnerable to image degradation in the presence of spatially varying B 1 amplitudes. In this work we develop a B 1 inhomogeneity-reduced version of single-slab 3D turbo/fast SE imaging based on the hypothesis that it is critical to achieve spatially uniform excitation. Slab selection was performed using composite adiabatic selective excitation wherein magnetization is tipped into the transverse plane by a nonselective adiabatic-half-passage pulse and then slab is selected by a pair of selective adiabatic-full-passage pulses. Single-slab 3D turbo/fast spin echo (SE) imaging (1) was recently introduced using short, nonspatially selective refocusing RF pulses to achieve a short echo spacing (ESP) and thus enhance imaging efficiency. To reduce power deposition and permit a longer echo train compared to that achieved with conventional high-flip-angle (Ϸ180°) refocusing RF pulses, variable, low-flip-angle pulses, calculated based on prescribed signal evolutions that consider tissue-specific T 1 and T 2 values, can be used in the refocusing pulse train (2-4). Despite the enhanced imaging efficiency, this technique suffers from nonuniform signal intensity or variable contrast over the field of view (FOV) due to spatially varying B 1 amplitude, particularly in clinical body imaging at 3.0 T.Adiabatic RF pulses (5-7), which employ simultaneous modulation of amplitude and frequency, have been widely used for MRI and spectroscopy to achieve uniform rotation in the presence of inhomogeneous B 1 amplitude. However, it has been challenging to achieve slice selection with adiabatic pulses due to nonlinear phase variation across the slice. Since this nonlinear dephasing is not refocused by linear gradients, signal loss results. To overcome this problem, self-refocused gradient-modulated adiabatic selective pulses (8 -10) were developed, but they required high RF power, long pulse duration, and abrupt gradient switching at its maximum strength. Additionally, the slice profile degrades rapidly if the desired B 1 amplitude or gradient slew rates are not achieved. To mitigate this high demand on the performance of the imaging hardware system, composite adiabatic selective pulses (11,12), adiabatic-half-passage (AHP) pulse followed by a pair of adiabatic-full-passage (AFP) pulses for phase compensation, can be used. For interleaved multislice SE imaging (11) the AHP pulse was replaced by a conventional selective 90°pulse, but with this approach the 90°pulse is not immune to inhomogeneous B 1 amplitude. For sequential multislice SE imaging (12), all the RF pulses in selection and refocusing can be fully adiabatic; however, for clinical imaging with very long echo trains the fully adiabatic approach may not be feasible due to the constraints of hardware ...

The Technical and Clinical Features of 3D-FLAIR in Neuroimaging

“…Slab-selective excitation can be performed using an excitation RF pulse that is spatially selective in one or more dimensions. 15,17 In a VFA-TSE sequence, a relatively long echo spacing is used for the first echo to accommodate the long slab-selective excitation pulse, while a short echo spacing is used for the second and subsequent echoes. 15 To suppress the out-of-slab FID artifacts, partial averaging of more than 1.4 is usually employed.…”

Whole‐brain 3DFLAIR at 7T using direct signal control

“…To solve these problems, refocusing pulse trains with variable and lower refocusing angles have been introduced. However, as shown in reference , 3D FSE imaging with variable refocusing angles is sensitive to dropouts.…”

Design of the Magnetic Resonance Imaging Evaluation of Mineralocorticoid Receptor Antagonism in Diabetic Atherosclerosis (MAGMA) Trial

“…A novel aspect of the protocol is the implementation of new MRI pulse sequence design that enables motion‐free high‐resolution coverage of the entire aorta, including the arch, to evaluate longitudinal changes in PAV in patients on spironolactone vs placebo. Coverage of the entire aorta allows detection of smaller changes and inclusion of smaller numbers of patients and provides an improvement over invasive surrogate measures acquired in limited portions of the arterial bed …”