Magnetic resonance (MR) imaging is a valuable tool to examine lesions (e.g. lymph node lesions) and visualize anatomical structures (e.g. brainstem and cervical vertebrae) in the head and neck¹. While a standard head coil is sufficient for visualizing anatomical regions above the base of the skull, inferior regions require a separate neck coil¹. Water bags in a head coil were previously used to improve the transmit field (|B1+| ) for pulsed Arterial Spin Labeling². Additionally, high permittivity materials (HPM) can be used to extend the field of view (FOV) of a surface coil³. This work aims to extend the FOV of a commercial head coil into the neck using strategically placed HPM bags.

Simulations: Numerical simulations were carried out to determine whether the HPM pads should be constructed from powder/water slurries using Barium titanate (BaTiO₃, εr = 298, σ = 0.39 S/m) or Calcium titanate (CaTiO₃, εr = 110, σ = 0.08 S/m)^4,5. A 4-port head sized quadrature-driven birdcage coil, similar to the experimental transmit head coil (Nova Medical, Wilmington, MA), was modeled using Microwave Studio (CST 2015,Germany), and loaded with a human model (Duke). Each port was tuned and matched at 297.2 MHz in the absence of HPMs (S-parameters, Fig. 1). Two HPM pads (~12x14x2cm³) constructed from sections of spherical and cylindrical shells were positioned at the base of the skull to extend from the posterior of the ears around the back of the head and neck on both sides, and extending 7cm outside the coil (Fig 1). The birdcage coil was not re-tuned and matched after placing HPM pads in simulation, as re-tuning and matching the commercial coil was not possible. S-parameters did not significantly change in the presence of HPMs (Fig 1). An accuracy of -30 dB was used to ensure convergence, and 5-8 million mesh cells were used with same global and local mesh settings for all cases. Transmit efficiency (|B1+|/sqrt(dissipated power)) was evaluated for three cases: 1) No HPM, 2) BaTiO₃ pads, and 3) CaTiO₃ pads.

Experiments: Simulation results suggested that CaTiO₃ pads improved FOV and transmit efficiency, so HPM pads were constructed from CaTiO₃ powder (Alfa Aesar, 99 %+, -325 mesh powder) and deuterium oxide (Sigma Aldrich, 99% atom %D). The saturated slurry was measured (85070E Dielectric Probe Kit, Agilent Technologies), and heat-sealed between plastic sheets to create two pads: 12x14x2cm³ and 11x14x2cm³ (εr = 110, σ = 0.08 S/m). A healthy volunteer was scanned on a Siemens whole-body 7T scanner (MAGNETOM, Erlangen, Germany) using a one channel transmit, 24 channel receive head coil (Nova Medical, Wilmington, MA). A transmit reference voltage (to achieve 90^o flip angle in the center with a 500 μs hard pulse) of 295 V was used for all cases. Flip angle maps (TR/TE = 5.000/2.42 ms, BW = 650 Hz/pixel, FOV = 285x285 mm², matrix size = 128x128) were obtained used a turbo-FLASH based technique (6). SNR was calculated using the Kellman method from the raw data (7), obtained using a GRE sequence (TR/TE = 200/4.1 ms, nominal FA = 20°, BW = 300 Hz/pixel, FOV = 285x285 mm², matrix size = 256x256). In order to assess image quality, T1-weighted MPRAGE (TR/TE = 2.250/3.48 ms, nominal FA = 9^o, BW =200 Hz/pixel, FOV = 256x216x158 mm³, matrix size = 348x348x144) images were also acquired. In-vivo sagittal and coronal slices through the brainstem were obtained for each case. All studies involving human subjects were performed in accordance with the institution’s IRB.

The simulation and experimental data demonstrate that extending the FOV of a head coil and improving the SNR in the neck is feasible using HPM pads. All experiments were carried out using HPM pads containing CaTiO₃. Experimental results show that HPM improved the SNR in the neck, especially the brainstem and neck muscles with percentage improvement ranging from 45% (spinal cord) to 478% (muscle) (Fig 3). Frontal and superior regions of the brain showed SNR loss with the HPM pads (Fig 3). However, as SNR is generally very high in those regions due to coil proximity, we anticipate that this decrease will not affect image quality.T1-weighted MPRAGE images showed improved quality in the posterior of the neck, especially in the muscles, brainstem, and the superior spinal cord and cervical vertebrae. HPM pads positioned in a head coil are an inexpensive method for extending SNR further into the brainstem and spinal cord, and this method may be applied to other existing commercial coils.