Fast design of local N‐gram‐specific absorption rate–optimized radiofrequency pulses for parallel transmit systems

Abstract

Designing multidimensional radiofrequency pulses for clinical application must take into account the local specific absorption rate (SAR) as controlling the global SAR does not guarantee suppression of hot spots. The maximum peak SAR, averaged over an N grams cube (local NgSAR), must be kept under certain safety limits. Computing the SAR over a three-dimensional domain can require several minutes and implementing this computation in a radiofrequency pulse design algorithm could slow down prohibitively the numerical process. In this article, a fast optimization algorithm is designed acting on a limited number of control points, which are strategically selected locations from the entire domain. The selection is performed by comparing the largest eigenvalues and the corresponding eigenvectors of the matrices which locally describe the tissue's amount of heating. The computation complexity is dramatically reduced. An additional critical step to accelerate the computations is to apply a multi shift conjugate gradient algorithm. Two transmit array setups are studied: a two channel 3 T birdcage body coil and a 12-channel 7 T transverse electromagnetic (TEM) head coil. In comparison with minimum power radiofrequency pulses, it is shown that a reduction of 36.5% and 35%, respectively, in the local NgSAR can be achieved within short, clinically feasible, computation times. Magn Reson Med 67:824-834, 2012. © 2011 Wiley Periodicals, Inc.Key words: parallel transmission; multidimensional RF pulse design; transmit array; ultra-high-field MRI; local SAR reduction High-field MRI faces problems with respect to B + 1 field inhomogeneity (1). These problems can be addressed by using parallel transmit systems. In radiofrequency (RF) shimming, the interference of the B + 1 fields of the individual channels is optimized for excitation homogeneity (2-5). Parallel transmit systems using spatially tailored RF pulses facilitate even more control of the local spin excitation within sufficiently short RF pulse durations and over a sufficient spectral range (6-9). Experimental demonstrations of these techniques have been shown in recent years (10-13). Their potential goes beyond the compensation for B + 1 inhomogeneity. Other applications include compensation for B 0 distortions (12) and reduced field-of-view imaging by spatially reduced excited region (inner volume excitation) (14,15). A major concern in spatially tailored parallel excitation (pTx) is the heating of the tissue being scanned. Especially for highly accelerated RF pulses, the specific absorption rate (SAR) tends to increase rapidly and can easily exceed the SAR guidelines (16-19). On the other hand, when integrated in the RF pulse design, pTx offers additional degrees of freedom to manipulate the SAR deposition for a given target excitation pattern (7).In vivo, local SAR guidelines are more likely to be exceeded than global SAR when using pTx excitation. Zelinski et al. (16) showed for 7 T head imaging how intricate the local SAR behavior can be for various target patterns an...

A generalized slab‐wise framework for parallel transmit multiband RF pulse design

“…In each case, Eq. [1] can be cast into an equivalent regularized MLS problem and solved using the algorithms described by Lee et al (28) and Sbrizzi et al (29). …”

Surface coil with reduced specific absorption rate for rat MRI at 7 T

“…Because the rat brain weighs approximately 2 g [40], we then assumed 100 mg for all numerical simulations. There seems to be no restrictions on the weight of the N-gram cube assumed to compute the SAR 100 mg according to previous studies [41][42][43]. To validate our numerical predictions, we used the fractional deposited power method reported by Prock et al [44] for circular-shaped coils of various radii.…”

Optimal control design of turbo spin‐echo sequences with applications to parallel‐transmit systems

“…The same formalism can be used to include constraints on global specific absorption rate and local specific absorption rate 35, 36. We expect that the computation time will increase by adding a large amount of local specific absorption rate constraints.…”