Practical improvements in the design of high permittivity pads for dielectric shimming in 7T neuroimaging
Thomas O'Reilly1, Wyger Brink1, and Andrew Webb1

1Leiden University Medical Centre, Leiden, Netherlands

Synopsis

Improvements are proposed for practical use of high permittivity materials in high field neuroimaging. These result in a simple formula to design materials with specified permittivity, formulation to improve the short term rigidity and long term stability of the material, and a method to incorporate devices such as headphones into the dielectric pad design.

Introduction

Regions of constructive and destructive interference of the B1+ field occur in high-field MRI as the RF wavelength is of the same order as the imaging region. Several techniques exist for B1+ shimming including the use of transmit arrays and k-space RF pulse design. “Dielectric pads” using high permittivity materials have been shown to be a simple and widely applicable alternative, particularly in neuroimaging at 7T 1-3. The aims of this current work are to improve the practical application of such pads in three different areas: (i) to provide a simple method to design pads with a specific permittivity, (ii) to improve the properties of the dielectric pad in terms of rigidity and long-term stability, and (iii) to test whether integration of devices such as headphones can be achieved while maintaining the advantages of the improved B1+ homogeneity.

Methods

Mixtures with different barium titanate (εr ≈ 2000) and calcium titanate (εr ≈ 160) ratios in deionized water were prepared . Complex permittivity measurements were performed using a SPEAG Dielectric Assessment Kit DAK-12 . Data were fitted to different mixing rules to determine the one which best fits the tri-component properties. The experimental measurements were repeated with different gelling agents (agar, xantham gum, phytagel and hydroxyethyl-cellulose (HEC)) added to increase the viscosity of the dielectric pad. To determine the effect of integrating headphones into the dielectric pad (by introducing a hole at the level of the ears), electromagnetic simulations were performed using xFDTD (7.4.0.2, Remcom, PA, United States) and the Duke body model . Based on the simulation results, dielectric pads were constructed using a mixture of calcium and barium titanates to give a relative permittivity of 150 with hole dimensions of 65 x 55mm. B1+ maps of the brain were obtained using the DREAM sequence4, as well as anatomical turbo spin echo images, on a 7T Achieva MRI system (Philips Achieva, Philips Healthcare, Best, the Netherlands) in healthy volunteers.

Results

The most appropriate mixing rule for the permittivity of aqueous suspensions of mixtures of barium titanate and calcium titanate was determined to be the Lichtenecker logarithmic law5

εeffCafCa εBafBa εwfw

where f represents volume fractions of calcium titanate, barium titanate and water. Figure 1 shows the fitted plot to individual data points, allowing the design a pad with any permittivity between 110 and 300. Of the different gelling agents used, the stability (both short-term and long-term) of the aqueous dielectric suspensions was improved most using 0.5 g of HEC per 100 ml of water. The addition of HEC to the dielectric material has no impact on the permittivity of the suspension but increases the conductivity from 0.1 S/m to 0.15 S/m. EM simulations showed that the impact of this increased conductivity on the performance of the pad in vivo is minimal. The results of the EM simulations to study the effect of a hole in the dielectric pads are shown in Figures 2 and 3, in terms of the B1+ and displacement currents in the pads, respectively. For a fixed permittivity of 110 a drop in the secondary B1+ created by the pad is seen at the location of the hole (Figure 2), but there is still significant secondary field enhancement elsewhere. Increasing the permittivity of the dielectric material in the pad to 150 largely compensates for the drop in the B1+ caused by the hole. Figure 3 shows that the displacement currents within the pad are not severely affected (except at the location of the hole) explaining the improved performance (compared to no pad) even in the presence of the hole. Figure 4 shows in vivo results using a turbo spin echo imaging sequence. In the absence of dielectric pads there is a well-characterized severe loss in image contrast and signal intensity in the temporal lobe and cerebellum. Using the dielectric pads with a hole for the earpads/headphones (shown in Figure 5) and a relative permittivity of 150 gives substantial improvements in image quality, similar to the effects of a full pad with relative permittivity of 110.

Conclusions

Using the Lichteneker logarithmic mixing rule allows a range of permittivities to be used for the construction of dielectric pads. Hydroxyethyl cellulose can be used to improve the stability of the dielectric suspensions with minimal impact on the efficacy of the dielectric pad. Introduction of a hole in the pad to accommodate headphones (or other devices) leads to a drop in the secondary B1+ produced by the pad at the location of the hole, but increasing the permittivity of the pad to ~150 produces very similar increases in image uniformity as the conventional full pad with relative permittivity 110.

Acknowledgements

This work was supported by NWO-STW Technology Grant 13783.

References

1. Haines,K., Smith N.B., and Webb A.G. New high dielectric constant materials for tailoring the B1+ distribution at high magnetic fields, J.Magn.Reson, 203, 323-327, 2010.

2. O'Brien KR, Magill AW, Delacoste J, Marques JP, Kober T, Fautz HP, Lazeyras F, Krueger G. Dielectric pads and low- B1+ adiabatic pulses: complementary techniques to optimize structural T1 w whole-brain MP2RAGE scans at 7 tesla. J Magn Reson Imaging. 2014 Oct;40(4):804-12.

3. Vu AT, Auerbach E, Lenglet C, Moeller 2, Sotiropoulos SN, Jbabdi S, Andersson J, Yacoub E, Ugurbil K. High resolution whole brain diffusion imaging at 7T for the Human Connectome Project. Neuroimage. 2015 Nov 15;122:318-31.

4. Nehrke, K. and Börnert, P. “DREAM – a novel approach for robust, ultrafast, multislice B1 mapping.” Magnetic Resonance in Medicine 68. 5 (2012).

5. Goncharenko, A.V., Lozovski, V.Z., Venger,E.F. “Lichtenecker’s Equation: applicability and limitations.” Optics Communications 174. 1 (2000).

Figures

Figure 1. The measured relative permittivity of an aqueous suspension of barium titanate and calcium titanate with a constant volume fraction of 0.6 of water. The volume fraction of calcium titanate is (0.4 – volume fraction barium titanate). The line is the predicted permittivity using the Lichtenecker logarithmic mixing rule.

Figure 2. Top) The B1+ distribution of a standard calcium titanate pad and two configurations with a 65 x 50 mm hole. Bottom) The change in B1+ distribution after the placement of dielectric pads. The pads with a hole show a decrease in B1+ at the location of the hole, marked with a circle.

Figure 3. Top) Locations of the dielectric pads (white), with and without holes, on the side of the head. Bottom) Simulated current densities inside the two pads: both pads have a relative permittivity of 110.

Figure 4. Coronal TSE images obtained from a male volunteer. Left) No pads used. Center) A standard dielectric pad (180 x 180 x 8 mm, εr=110) on each side. Right) A dielectric pad (180 x 180 x 8 mm, εr=150) with a 65x50mm hole above the ear placed on each side.

Figure 5. Dielectric pads (180x180x8mm, εr = 150) incorporated into the headphone system of the MRI scanner.



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
0394