Synopsis

Keywords: High-Field MRI, Simulations, Head position, B1+ field, 7T MRI

In this work, we computationally and experimentally evaluated the B₁⁺ field and SAR generated by a 60-channel Tic-Tac-Toc RF coil at 7T MRI scanner on a variety of participants, head models and head positions. On three anatomically detailed human head models, we found slight changes in the intensity and the distribution of B₁⁺ field and SAR over the whole brain when the head was moved out from the top of the coil. This was confirmed in both simulation and experimental B₁⁺ maps. Our results highlighted the impact of imaging setup on imaging quality.

Introduction

To improve the imaging quality under ultra-high field, the Tic-Tac-Toe (TTT) radiofrequency (RF) head coil design for 7T MRI have been devised to increase B₁⁺ homogeneity and have demonstrated load insensitivity ^1-3. This has been further improved with the new 60-channel double-row TTT coil system⁴. However, the impact of imaging setup, such as the spatial position of the head, on imaging quality is still under investigation. Thus, in this study, we assessed the impact of the head position on the B₁⁺ field distribution/intensity, attempting to provide better understanding of imaging strategy and guidance of examination setup.

Methods

To computationally simulate the influence of head position on the B₁⁺ field, electromagnetic fields produced on the geometry of the 60-channel octagonal Tic-Tac-Toc RF head coil were simulated with an in-house Finite-Difference Time-Domain (FDTD) algorithm. The simulations were performed by using three different anatomically detailed head models including the Duke (34 years old male, 155 lbs), Ella (26 years old female, 126 lbs), and Billie (11 years old female, 75 lbs) models from the Virtual Family⁵. We adopted two different head positions (Figure 1). For the Duke model, we moved out 8 mm from the reference position (i.e. the position very close to the coil). For the Ella and Billie models, we moved out 16mm from the reference position. To analyze and compare B₁⁺ field and SAR distributions, we calculated mean B₁⁺ value, coefficients of variation(CV) of B₁⁺ field, specific absorption rate (SAR) efficiency and peak/average SAR over the whole brain including the cerebellum. Experimentally, a 12-minute Turbo-FLASH sequence (TR/TE = 2000/1.16ms, flip angle from 0° to 90° in 18° increments, 3.2mm isotropic resolution) was used to obtain in vivo B₁⁺ maps, which were acquired from three participants with the similar experimental position to the simulation to evaluate the real-world scenario.

Results

The simulated B₁⁺ fields of three head models were visualized in Figure 2. We observed that all three models had a higher average B₁⁺ field over the whole brain when the head was moved out from the top of the coil (position B in Figure 1). Regionally speaking, there existed a higher B₁⁺ field at the superior region of the brain and lower B₁⁺ field in the cerebellum when moving out the head. Also, it showed that the B₁⁺ field became slightly less homogeneous when the position was moved out, which was indicated by higher coefficients of variation of B₁⁺ field (Figure 2). Besides, the SAR efficiency and peak/average SAR of simulations were assessed (Table 1); the outer position provided better SAR efficiency compared to the reference position. Moreover, the experimental in vivo B₁⁺ maps demonstrated consistent findings with the simulated outcomes (Figure 3 & Table 1), confirming the effect of head position under the scanning of ultra-high fields.

Discussion and Conclusion

In this study, we found that the head position would slightly affect the behavior of B₁⁺ field at 7T MRI; the head at an outer position would have a relatively higher average B₁⁺ value, but it would be less homogeneous. Also, the regional difference of B₁⁺ field would be also affected by the position. Our result suggested that the position of scanning objects had a certain impact on imaging quality at 7T MRI. The examination setup should consider the regions of interest to optimize imaging strategies under ultra-high field, especially for repeated measures in a longitudinal study.

Acknowledgements

This work was supported by NIH R01MH111265, R01AG063525, R56AG074467, and T32MH119168. This research was also supported in part by the University of Pittsburgh Center for Research Computing (CRC) through the resources provided.