Introduction

Parallel transmission (pTx) has been proposed for flip-angle (FA) uniformity excitations at high fields while maintaining or decreasing the local specific absorption rate (SAR) [1,2]. PTx has been suggested as a possible strategy for SAR control at the electrode tip of deep brain stimulation (DBS) patients [3,4]. However, to our knowledge, no study has been performed exploring the SAR-reduction potential of pTx in realistic DBS patient models using multi-row pTx coil designs. In this work, we use a realistic DBD patient model in conjunction with a co-simulation strategy and constrained pTx pulse design to assess the potential of pTx to solve the RF-induced heating problem in DBS patients at 3 Tesla.

Methods

DBS patient model: The DBS patient model used in this study was created as described in [5]. The model is a realistic representation of both the anatomy (bone, internal air and “average tissue” compartments) and the DBS implant (including extension cables, individual electrode cabling and the Implantable Pulse Generator). The model is based on head and neck CT scans of an actual DBS patient. FEM simulation: We simulated 5 coils, each of which were loaded with the DBS patient model: A 32-rungs high-pass body birdcage (BC) coil and four pTx coils with 8/16 channels distributed in 1/2 rows (Fig. 1). The BC coil was driven in two ways: As a quadrature coil or as a pTx coil with each quadrature port driven independently. The body pTx coils were designed on the same cylindrical former as the BC (diameter=711 mm, height=450 mm). The head pTx coils were designed on a former with a cylindrical base (diameter=320 mm) to which we added a partial circular arc curving toward the head (total height of coil=288 mm). We used a co-simulation strategy based on ADS (Agilent, Santa Clara CA) and HFSS (Ansys, Canonsburg PA) to compute the electromagnetic fields in the body model [2,6]. This allowed fast tuning and matching of the coil using a single multi-port simulation of each coil (simulation time was <24 hours per coil and was dominated by meshing of the DSB model). Pulse design: We arranged the electric field E in matrix form (E is Npixels×Nchannels), from which we computed the following Nchannels×Nchannels matrices: QP=dV×EH×diag(s)×E and QGS=EH×diag(s/(2r))×E), where dV is the voxel volume, s is the conductivity and r is the density vectors. QP is a matrix allowing fast computation of the absorbed power around the electrode tips (APAET) in Watts. The QP matrix was evaluated using high-resolution (0.1 mm) E-field maps within a small cylindrical volume (diam.=6 mm, length=15 mm) enclosing each electrode tip. The global SAR matrix QGS was computed using medium resolution (1.8 mm) E-field maps including the entire head but excluding a small spherical volume (rad.=15 mm) around each electrode tip. Using these matrices, we designed pTx pulses while constraining both the APAET and global SAR. We designed magnitude least-squares single-spoke slice-select pulses (transverse slice through the frontal lobe) with a target flip-angle (FA) of 10°. For each coil, L-curves where obtained by varying the APAET hard constraint.

Results

The L-curves in Fig. 2 show the optimal tradeoff between the pulses FA error and associated APAET for each coil modeled. The pulses in Fig. 2 were designed by only constraining APAET (global SAR was not constrained). Although pTx coils are clearly able to dramatically decrease APAET while maintaining or improving the FA quality compared to BC-quad, this comes at the cost of unacceptably high global SAR if global SAR is not explicitly constrained in the pulse design process (Fig. 3). Figure 4 shows the more useful approach of constraining both the APAET and global SAR. In this case, the pTx coils were able to reduce the APAET by 35-91% compared to BC-quad while maintaining global SAR and improving the FA quality. The simplest pTx strategy consisting of driving the 2 channels of the BC coil independently reduced APAET by 17% at constant global SAR and FA error.

Results

PTx coils with 8/16 channels arranged in 1/2 rows are able to greatly reduce the APAET in DBS patients imaged at 3 Tesla without increasing global SAR nor the FA excitation error. At constant number of channels, local pTx coils (=head only) greatly outperform body pTx coils in the APAET and FA error metric at constant global SAR.

Acknowledgements

NIH grants K99/R00 EB019482, R01EB006847.

References

[1] Lee, J., et al. (2012). "Local SAR in parallel transmission pulse design." Magnetic Resonance in Medicine 67(6): 1566-1578. [2] Guérin, B., et al. (2015). "Comparison of simulated parallel transmit body arrays at 3 T using excitation uniformity, global SAR, local SAR and power efficiency metrics." Magnetic resonance imaging 73(3): 1137-1150. [3] Eryaman, Y., et al. (2014). "Parallel transmit pulse design for patients with deep brain stimulation implants." Magnetic Resonance in Medicine 73(5): 1896-1903. [4] Gudino, M., et al. (2013). Parallel transmit excitation at 1.5 T based on the minimization of a driving function for device heating. Proceedings of the 21st Annual Meeting of ISMRM, Salt Lake City, Utah, USA. [5] Guerin et al (2018). “Generation of a “virtual population” of deep brain stimulation patient models for MRI safety studies”. Proceedings of the ISMRM, Paris France. [6] Kozlov, M. and R. Turner (2009). "Fast MRI coil analysis based on 3-D electromagnetic and RF circuit co-simulation." Journal of Magnetic Resonance 200(1): 147-152.