Purpose:

To improve the transmit (Tx) performance of a human head array at 9.4 T (400 MHz), a novel method of decoupling adjacent dipole antenna elements was developed and evaluated.

Introduction:

Dipole antennas have been recently introduced to the MRI field (1) and successfully utilized mostly as elements of ultra-high field (UHF, > 7 T) human body arrays (1-3). Usage of dipole antennas for UHF human head arrays is still in the development stages. Due to the substantially smaller size of the sample, dipoles must be significantly shorter than λ/2. In addition, head arrays are commonly placed on the surface of rigid helmets made sufficiently large to accommodate various heads. As a result, dipoles are not well loaded and are often poorly decoupled (4,5), which compromises the Tx-performance. Commonly adjacent elements of a Tx-array are decoupled by circuits electrically connected to both elements (6-8). Placement of such circuits between distantly located dipole antennas is difficult unless they are bent to bring their ends closer to each other (5). The latter restricts the choice of array geometries and may decrease the performance. Alternatively, decoupling is done by placing additional decoupling antennas between adjacent dipole elements (4). This method only works when decoupling components are made sufficiently small (compared to the distance to the sample) not to compromise the B₁ profile of the Tx-array itself (9), which is not the case of the reported design (4). In this work, we developed a novel decoupling method of adjacent dipole elements of a transceiver (TxRx) human head array at 9.4 T.

Methods:

In comparison to surface loops, which are coupled inductively (positive reactance), a pair of loaded dipole antennas is coupled capacitively (negative reactance). Therefore, a mutual inductance has to be created to decouple dipole elements. Fig.1A demonstrates how this inductive coupling is produced by folding the dipoles and placing an RF shield near their folded portions. The presence of the shield is equivalent to placing a mirror image of the dipole carrying an opposite current at a distance double of the distance to the shield. Folded portions of both dipoles (a real dipole and its mirror image) produce a capacitive bridge, which depends on the distance to the shield and the folded length. Thus, by adjusting the distance to the shield and folded length, we can compensate the intrinsic capacitive coupling. In addition to folded straight dipoles (Fig.1A), we also evaluated other types of dipoles shown in Figs.1B and 1C. Electromagnetic (EM) simulations of the coupling between dipoles (S₁₂), transmit B₁⁺, and local specific absorption rate (SAR) were performed using CST Studio Suite 2017 (CST, Darmstadt, Germany) and the time-domain solver based on the finite-integration technique. Two voxel models were used, i.e. a head/shoulder (HS) phantom (ε = 58.6, σ = 0.64 S/m), and the virtual family multi-tissue model “Duke” (Fig.1C). All data were acquired on a Siemens Magnetom 9.4 T human imaging system. After numerical optimization of decoupling and evaluation of array performance and safety, we constructed and tested an 8-element single-row (1x8) array consisting of 30-mm folded bent dipoles (30 mm – the length of the folded portion).

Results and Discussion:

As an example, Fig.2 shows the dependence of the S₁₂ value on the length of the folded portion for different heights of the folded bent antennas and the size of the phantom. As seen from this figure, for each height of the dipole, a corresponding length can be chosen, which provides the lowest S₁₂ value. It is noteworthy that creating capacitive coupling to the shield at one side of the dipole element (Fig.1B) is sufficient to generate a proper mutual inductance. Table 1 shows results of numerical evaluation of the average S₁₂ value between adjacent dipoles for different 8-element dipole arrays. The 30-mm folded bent dipoles provide the best decoupling, which is about 7 dB better than that of unfolded dipoles. Figs.3A and 3B show a photo of the dipole array and the schematic of a single dipole element. Fig.3C shows S₁₂ matrices measured using the new array for different heads. Average S₁₂ value between all adjacent elements measured -16.6 dB and -18.4 dB for the smaller and larger head, respectively The constructed array delivers good decoupling even for a small head. As an example, Fig.4 shows a set of sagittal MP2RAGE images obtained using the dipole array.

Conclusion:

We developed a novel method of decoupling of adjacent dipole antennas, and used this technique while constructing a 9.4 T human head 8-elelment TxRx dipole array coil. The array demonstrates good decoupling and full-brain coverage.

Acknowledgements

SG acknowledges a support by Government of Russian Federation (Grant 08-08) and Ministry of Education and Science of the Russian Federation (No. 3.2465.2017/4.6). Funding by the European Union (ERC Starting Grant, SYNAPLAST MR, Grant Number: 679927) is gratefully acknowledged by AN and RL.

References

1) Raaijmakers AJE, Ipek O, Klomp DWJ et al. Magn Reson Med 2011;66:1488-1497. 2) Oezerdem C., Winter L., Graessl A. et al. Magn Reson Med 2016;75:2553–2565. 3) Raaijmakers AJE, Italiaander M, Voogt IJ et al. Magn Reson Med 2016;75:1366–1374. 4) Clément J, Gruetter R, and Ipek Ö. Magn Reson Med 2019;81:1447–1458. 5) Connell IRO, Mennon RS IEEE Trans Med Imag 2019;38: 2177-2187. 6) Avdievich NI. Appl Magn Res 2011;41(2):483-506. 7) Jevtic J. Proc. ISMRM 2001, p. 17. 8) Shajan G, Kozlov M, Hoffmann J et al. Magn Reson Med 2014;71:870–879. 9) Avdievich NI, Pan JW, and Hetherington HP NMR in Biomed 2013;26(11):1547-1554.