MR Imaging of deeply located targets at ultrahigh field strengths is hindered by reduced penetration depth and interferences of the B₁⁺-signal To overcome these problems, body imaging at high fields strengths is predominantly done with on-body transmit arrays^{1, 2}. Dipole antennas have shown advantages as transmit elements when imaging deeply situated targets ^3-5. It has been shown in a simulation study that segmenting the legs of dipole antennas by inductances can provide a significant decrease in local SAR while maintaining sufficient B₁⁺ level. This finding resulted in the design of the fractionated dipole antenna (figure 1) that is now frequently used for various clinical applications ^6-9. It has been demonstrated that by changing the dipole leg geometry, SAR characteristics can be made even more beneficial ¹⁰. In this study, the snake antenna is presented as a new type of transmit element for body imaging at ultrahigh field strengths. Its design consists of a dipole antenna where the legs have a continuously distributed inductance in the shape of a sinusoidal with slightly increasing amplitude (figure 1).

Finite-difference time domain simulations were carried out in Sim4Life (ZMT, Zurich, CH). Two different antennas were simulated on a 0.5*0.5*0.5 m³ phantom with electrical properties comparable to the human trunk (σ = 34, ρ = 0.4 S/m). The antenna models were compared in terms of B₁⁺ field strength and maximum 10 g averaged local peak SAR (SAR_max) An array of eight antennas was simulated on the human model Duke for the two antenna types. Results were evaluated in terms of B₁⁺ in the prostate and SAR_max. An array of eight snake antennas was constructed in the lab and tested on a healthy male volunteer (BMI 21.7). Performance in terms of B₁⁺ field strength in the prostate and image quality was compared to the fractionated dipole antenna array.

Figure 2 shows simulation results for the two investigated antenna types on a phantom. The SARmax is 1.3 W/kg for the fractionated dipole and 1.1 W/kg for the snake antenna. The snake antenna reaches less B₁⁺ signal strength than the fractionated dipole antenna, however the B₁⁺/√(SAR_max) ratio is better. The B₁⁺ field patterns in human model Duke are shown in figures 3b and 3e. The average B₁⁺ in the prostate is lower for the snake antenna array. Figures 3c and 3f show the SAR distributions in the human model. The SAR_max is 77% lower for the snake antenna array. The B₁⁺/√(SAR_max) ratio is 7.0 uT/√(W/kg) for the snake antenna array against 5.5 uT/sqrt(W/kg) for the fractionated dipole antenna array. Figure 4 shows two B₁⁺-maps that were acquired on the same volunteer with the different setups. The average B₁⁺ in the prostate region is 16% higher for the snake antenna array. T2w prostate images were acquired with both setups and are presented in figure 5.Simulations with a single element on a phantom show that the snake antenna has a beneficial B₁⁺/ √(SAR_max) ratio with respect to the fractionated dipole antenna. Simulations of an 8-element array setup on Duke confirm the single-element phantom simulation results; the B₁⁺/√(SAR_max) ratio is even more beneficial in this case. The asymmetric form of the load and the small variation in load-antenna distance may explain differences between the phantom and the human model simulation results. Figure 4 shows that a higher B₁⁺ is reached in the prostate with the snake antenna array, this finding is not confirmed by simulations. The small difference might be caused by a difference in matching performance for this specific volunteer or difference in RF shimming outcome. Nevertheless, these results provides confidence that the B₁⁺ levels of the snake antenna array are not significantly lower than for the fractionated dipole array. The lower SAR levels of the snake antennas make them overall more favorable. In figure 5, no discernable differences in image quality of the T2w images can be observed.The snake antenna is presented as a new type of transmit array element for body imaging at ultrahigh field strengths. Simulations show that the B₁⁺/√(SAR_max) ratio is 27% more beneficial for the snake antenna array than for the fractionated dipole antenna array. A prostate imaging comparison between both arrays shows that the snake antenna is able to reach at least equally high B₁⁺ values in the prostate region, and that no discernable differences in image quality can be found. The lower SAR levels of the snake antenna will reduce scanning constraints for ultrahigh field body imaging, for example enabling the acquisition of 62% more slices in multislice TSE prostate imaging.