Purpose

Recent work has shown that tradition coil design are not optimal at ultra-high field (UHF, >=7T) ^[1,2]. Studies on ideal current patterns (ICP) at UHF have shown that the ultimate intrinsic signal-to-noise (SNR) at the center of a cylindrical dielectric sample can be achieved using distributed z-oriented current patterns rotating at the Larmor frequency ^[1]. This configuration cannot be achieved with typical conductive strips ^[3], which limit the current patterns to the conductor’s surfaces, therefore, limiting the optimal SNR at the center, and by reciprocity, transmit efficiency. Last year, a slot antenna concept for UHF was introduced, demonstrating an antenna structure for wide field-of-view imaging using a single antenna element ^[4]. In this work, we compared performance of a single planar slot antenna with a 3-dipole array. We also juxtaposed the electric field vectors produced by these coil structures to ideal current patterns, which at UHF correspond to the projection of the electric field on a spin precessing at the voxel of interest ^[2].

Methods

Simulations were performed using the Comsol Multiphysics 5.2a finite element modeling (FEM) solver (Burlington, MA, USA). A 0.8cm x 0.4cm x 0.8cm rectangular phantom with relative permittivity of 45.33 and conductivity of 0.33 S/m, resembling average electrical tissue properties of the torso at 300 MHz was used in the simulations. Two simulation were performed: the first, with a three dipole array positioned 3 cm above the phantom. The distance between the dipoles in the x-direction was 20cm. Each dipole was 50 cm long and 1cm in diameter with a match <9.9 dB. In the second simulation, a single slot antenna modeled positioned 3 cm above the phantom. For the slot antenna, a conductor of length 46.3cm x 25cm was modeled with a 1cm by 43.8cm gap in the middle. Antennas were driven with a 1V source at the central port operating at 300 MHz. Transmit fields (B₁⁺) of both simulations normalized by the square root of total power deposition inside the phantom was plotted, central B₁⁺ was evaluated, and electric field vectors 1cm above the phantom were plotted (coronal). Similarly, ICP were computed for optimal SNR at the center of a cylinder with radius of 20cm and length of 80cm. ICP were defined on a cylindrical surface of radius=20.5cm, while the system shield was modeled at radius=34.25cm. The conductivity and relative permittivity of the cylindrical phantom were 0.33 S/m and 45.33, respectively. Simulations for an array with 8-dipoles and one with 4-slots were performed with the same cylindrical phantom, where transmit efficiency was evaluated.

Results

B₁⁺ efficiency of a planar slot and a three-dipole antenna setup are shown in figure 1. The B₁⁺ at the center of the dipole and slot setups were 6.38e-8 Tesla and 1.09e-7 Tesla, respectively. This corresponds to a central B₁⁺ efficiency improvement > 70% for the single slot setup. Power deposition for the dipole and slot setups was 141.3 and 395.3 Watts, respectively. Figure 2A shows that the ICP are dominated by distributed z-directed currents, contributing the optimal performance at the center. The electric field distribution for the dipoles and slot antennas are shown in figure 2B and C, respectively, demonstrating distributed z-directed vectors in the slot antenna simulations. Figure 3 illustrates transmit efficiency comparing a 4-slot antenna and 8-channel dipole arrays. The improvement is observed in transmit efficiency at the phantom center alongside a reduction in homogeneity.

Conclusions

As observed by the ICP simulation, distributed z-directed electric field vectors are responsible for optimizing transmit efficiency at the center of the cylinder. Distributed z-directed electric field vectors were observed when utilizing slot antennas and can explain improved efficiency over dipoles.

Acknowledgements

Funding from NIH through R01 EB011551, R01 EB002568, and P41 EB017183.

References

[1] Lattanzi R, Sodickson DK. Mag. Reson. Med. 2012; 68:286–304. [2] Sodickson DK, et al. Intl. Soc. Mag. Reson. Med. 24 (2016). p. 389. [3] Raaijmakers AJ. Magn. Reson. Med. 2011 Nov;66(5):1488-97. [4]. Alon et al. Proc. Intl. Soc. Mag. Reson. Med. 24 (2016). P. 3516.