2696

A Simple Head-sized Phantom for Realistic System Characterization at 7T

Wyger M. Brink¹, Zhiyi Wu¹, and Andrew G. Webb¹

¹Leiden University Medical Center, Leiden, Netherlands

Synopsis

A simple head-sized phantom has been developed to produce realistic B₁ and B₀ features and electrical loading conditions, as a tool for the evaluation of MR techniques and RF validation in high field MR systems.

Introduction

Tissue-mimicking magnetic resonance (MR) phantoms with appropriate dielectric properties are useful for many MR applications including RF coil validation and the development of acquisition protocols. In this work, a simple head-sized phantom is implemented and evaluated to realistically reproduce B₀ and B₁ field interactions of the human head on a 7T MR system. The utility of the phantom is demonstrated in a RF validation study using MR thermometry in a 7T volume transmit coil.

Methods

A spheroidal phantom of size 18×22×27 cm was constructed from polymethylmethacrylate (PMMA), in which a 7-cm diameter spherical cavity was incorporated to mimic the B₁ and B₀ field perturbations induced by the nasal cavity.1 The shell was filled using an aqueous solution of 76g Polyvinylpyrrolidone (PVP10, Sigma Aldrich, The Netherlands) together with 1.78g Sodium Chloride (NaCl) in 100g of demineralized water, in order to obtain appropriate dielectric properties. The solution was then gelled using 1.5% agarose (w/w) to reduce thermal convection. The dielectric properties of the phantom material were characterized using a dielectric probe kit (DAK-12, SPEAG, Zürich, Switzerland) at a relative permittivity of ε_r = 50.1 and electrical conductivity of σ = 0.58 S/m. The thermodynamic properties were measured using a thermal probe (KD2-PRO, Decagon Devices, Pullman, WA, USA) and recorded at a thermal conductivity of κ = 0.405 W/(m∙K) and volumetric heat capacity of 3.460 MJ/(m3∙K).

All experimental data were acquired on a 7T MR system (Achieva, Philips Healthcare, Best, the Netherlands) using a 16-rung birdcage transmit coil for RF transmission and a 32-channel receive array for signal reception (Nova Medical, Wilmington, MA, United States). B₁⁺ maps were acquired using a DREAM sequence (2.5 mm² resolution, 5 mm slice thickness, STEAM/imaging tip angle = 50°/10°). MR thermometry was performed using the proton reference frequency method by acquiring a dynamic series of coronal 3D gradient echo acquisitions (2.5×2.5×5 mm³ resolution, FOV = 240×240×180 mm³, TR/TE = 8.9/4 ms, acquisition time = 30 sec per dynamic, 120 dynamics). A preparation pulse module was added using a block-pulse modulated 100 kHz off-resonance to increase the RF power deposition of the sequence without interfering with the image acquisition.

Numerical simulations of the heating profile were performed using CST Microwave Studio (2016, Darmstadt, Germany) to model the RF power deposition and temperature increase from a 16-rung highpass birdcage transmit coil driven in quadrature mode at 300 MHz. The RMS B₁⁺ level was used to align the simulated field amplitudes with experimental conditions, which resulted in a simulated input power of 50W.

Results

Figure 1 shows simulated B₁⁺ profiles in a heterogeneous and homogeneous head model, emphasizing the relevance of the dielectric heterogeneity on the observed B₁⁺ inhomogeneity. The phantom approximates the outline of the human head in a simple ellipsoidal geometry, and the most prominent B₁⁺ heterogeneities are reproduced by including a spherical cavity in this design.

Figure 2 shows an experimental comparison of the B₁⁺ fields measured in the phantom as well as in vivo. In addition to the perturbations in the B₁⁺ profile, the spherical cavity also induces the appropriate B₀ perturbations corresponding to those observed in the frontal cortex, shown in Figure 3. Finally, as the phantom reproduces the electrical loading conditions of the human head the phantom can be used for RF validation purposes by means of MR thermometry, the results of which are illustrated in Figure 4.

Discussion/Conclusion

A simple phantom design is presented and evaluated for system characterization at 7T. The phantom reproduces B₁⁺ and B₀ heterogeneities similar to those encountered in the human head, allowing for realistic RF characterization and method evaluation.

Acknowledgements

No acknowledgement found.

References

[1] Brink W.M., Webb A.G. Apparant B1+ Asymmetry in Symmetric Objects at High Fields. Proceedings of the 22nd ISMRM in Milan, Italy, 2014; p. 4815.

Figures

Figure 1. Motivation for the phantom design. Simulations in a heterogeneous body model (left) shows several B₁⁺ features (arrows) which are not reproduced in a homogeneous body model (center), but require the inclusion of a spherical cavity (white) to account for RF field perturbation induced by the nasal cavity. Simulated data were normalized to 1W of dissipated power.

Figure 2. Experimental characterization of B₁⁺ in the transverse (top) and coronal plane (bottom). The in vivo profiles in the B₁⁺ pattern (left) are reproduced in the phantom (right), with a distinct left-right asymmetry and central brightening profile. Measured B₁⁺ data were normalized by transmitted RF power recorded at the amplifier output.

Figure 3. Experimental characterization of B₀ in the transverse (top) and sagittal plane (bottom). The B₀ inhomogeneities around the nasal cavity (left) are reproduced by the inclusion of a spherical cavity in the phantom.

Figure 4. Example MR thermometry result. The temperature increase is shown in the sagittal (left) and coronal plane (right) after one hour of heating at an average power of 50W at the coil input.

Proc. Intl. Soc. Mag. Reson. Med. 25 (2017)

2696