Introduction

To address the local transmit field inhomogeneities caused by wave tissue interactions at high and ultrahigh field strengths high permittivity materials (HPM) have been used strategically ^1-8. These have included relatively bulky bags of water or gel (1-3), slurries of ceramic powder in water which can both raise permittivity and reduce bulk compared to pure water (4-6), and pure ceramic materials which can increase permittivity even further with relatively little volume and increasing stability through time (7, 8). While many of these application have been focused on the imaging of the human head (1, 2, 4, 7, 8), some have been focused on imaging of the torso (3, 6). Recently, a thin flexible metasurface (MS) was designed to mimic the electrical properties of an HPM pad for improving RF field performance in imaging of the human abdomen⁹. In this work we quantitatively evaluated the performance of two ultra-thin flexible metasurface structures at 3T.

Methods

Metasurface Design & Construction: Two different metasurfaces were evaluates in this work. The first was a slightly smaller version of one published previously⁸ and produced with the same methods but without the alterations removing portions on each side to avoid resonance of the structure. The second was designed to have equivalent performance as the previously published version⁸, but with a slightly thicker substrate (50 micron compared to 25 micron) requiring larger capacitive overlaps between the two sides. The first (smaller) metasurface had strip width 0.1cm and periodicity of 1.4 cm machined on a DuPont Pyralux 8515R substrate (ε_r -3.3) with polyimide thickness of 25um and copper thickness of 18 um, and the second (larger) metasurface had strip width 0.1cm and periodicity of 2 cm manufactured on a DuPont Pyralux 9121R substrate (ε_r -3.3) with polyimide thickness of 50 um and copper thickness of 35 um.

Imaging: To quantify the performane of the metasurface structures, imaging experiments were performed on a tissue equivalent body phantom with and without the two metasurfaces. Measurements were performed using a whole-body 3 T scanner (MAGNETOM, Siemens, Healthineers, Germany). Spin excitation and signal reception were evaluated in two configurations 1) System body coil in Tx/Rx mode 2) System body coil Tx and 12 channel general purpose flexible array for Rx. B₁⁺ and SNR maps were respectively acquired using TurboFLASH (TE-1.9ms, TR-10000ms) and GRE sequences (with and without RF). To evaluate in vivo performance abdomen scans were performed on a healthy subject in configuration 2 with and without the large metasurface.

Results

In for both Tx/Rx body coil and Tx body coil, Rx receive array, both the small and large metasurfaces provided approximately a 25% improvement in transmit efficiency at the phantom center over the setup with no metasurface (Fig.2).

Three-axis SNR maps (normalized to excitation flip angle) for the two imaging configurations are presented in Figure 3A, 3B. For Tx and Rx with the body coil, SNR maps with both the small and large metasurface show significant increase in SNR near the periphery and extending toward the coil center. The effects of the metasurfaces on SNR when receiving with the array are relatively small and are limited towards the phantom periphery.

As with the phantoms, in vivo B₁⁺ maps show that the transmit field is stronger, especially near the metasurface, when the metasurface is present (Fig.4).

Discussion and Conclusions

Here we present quantitative evaluations of how two different metasurfaces affect transmit efficiency and SNR for a Tx/Rx body coil and for excitation with the body coil and reception with a commercial surface array. In all cases, the metasurfaces improved transmit efficiency in the imaging region adjacent the coil, and when the body coil was used for reception, similar increases in SNR were found. The metasurfaces had relatively little affect on SNR of a 12-element commercial receive array. These results indicate that thin, flexible metasurfaces could likely be used to replace bulkier high-permittivity pads used to improve regional image SNR in imaging of the torso.

Acknowledgements

Ryan Brown

References

1. Alsop et al., MRM 1998; 40:49-54.

2. Yang et al., JMRI 2006; 24:197-202.

3. Franklin et al., JMRI 2008;27:1443-7

4. Haines at al., JMRI 2010; 203:323-327.

5. Webb, Concepts Magn Reson Part A 2011;38(4):148-84.

6. De Heer et al., MRM 2012;68:1317-24

7. Sica C et al., MRM. 2020 83(3):1123-34.

8. Lakshmanan et al., MRM. 2021;86:1167-74.

9. Vorobyev et al., MRM. 2022;87:496-508