Introduction

A prominent solution to B1+ inhomogeneity at ultra-high field strengths is the implementation of localised RF shimming with the help of high dielectric constant (HDC) pads which have been shown to improve transmit B1 fields over regions of interest (ROI) [1,2]. In this abstract, we aim to validate the use of conformal dielectric pads for localised RF shimming of the visual cortex at 7T with 3 different body voxel models across 2 transmit elements - a single dipole and a single loop – using electromagnetic simulations.

Methods

The head-sized dipole and single loop were modelled in CST (CST AG, Darmstadt, DE). The dipole has a length of 20cm and a diameter of 2mm with the feed port and tuning/matching network placed at the middle of the dipole. The rectangular loop has an area of 10 x 12 cm2­, with a trace width of 5mm and a thickness of 50μm. Distributed capacitors were modelled as lumped elements along the length of the loop and 2 discrete ports added to the loop. The discrete ports allowed us to place variable tuning / matching capacitors and 50 Ohm terminations in the simulated coil schematic along with predefined capacitors (RLC, series) placed at the lumped element ports along the coil. Both coil elements were placed 20mm away from the posterior ends of the voxel models. The dielectric pads were designed to be 12 x 12 cm2­ with a thickness of 5mm, and placed posterior to the occipital pole, conformal to the voxel models’ scalp in the simulation space. The dielectric pads were simulated with permittivity εr = 288 and conductivity σ = 0.36 S/m, similar to that of a BaTiO3 slurry [3]. EM simulations were undertaken for 3 different body models: Gustav, Laura and Child, both with and without the dielectric pads. Both coils were tuned and matched to 300 MHz at 50 Ω, with a S11 of -40dB or better. B1+ fields and Specific Absorption Rates (SAR10g) were calculated for all coil and voxel model configurations.

Results

Axial views of the B1+ profiles for all simulations and models are shown in Fig 1. For single transmit elements, the conformal HDC pads show changes in B1+ and SAR distributions across all models, with the areas under the pads showing increased B1+ and deeper penetration. Axial views of the SAR profiles for all simulations and models are shown in Fig2. The maximum SAR was slightly lower when using the dielectric pads than without, but within 10% of each other [Fig2].

Discussion

The aim of the study was to simulate high dielectric constant materials for use in-vivo at 7T for 2 different transmitter geometries. The simulation data shows that a conformal placement of the dielectric pads with respect to the voxel model results in an increased B1+ close to the pads for both coils, while the maximum SAR is decreased by upto 10% when using the pads. The pads can be effectively used to enhance local RF field performance, depending upon the imaged ROI. This can also allow for direct integration of dielectric pads in RF coil designs, where the pads can be integrated into the coil former. Further work will involve simulation and characterisation of B1+ profiles and SAR for different multi-transmit coil geometries while utilising dielectric pads - namely phased arrays, dipole arrays and microstrip arrays.

Acknowledgements

No acknowledgement found.

References

[1] Teeuwisse et al., MRM 2012, 67:1285-93; [2] Yang, et al., JMRI 2006, 24: 197–202 [3] Brink., et al., Proc. ISMRM 2015, 3108