0268

RF transparent ultrafast gradient insert

Dennis Klomp^1,2, Edwin Versteeg¹, Riccardo Metere³, Andrew Webb⁴, Erik Van den Boogert³, David G Norris³, Matrino Borgo⁵, and Jeroen Siero^1,6

¹Radiology, UMC Utrecht, Utrecht, Netherlands, ²MRCoils, Zaltbommel, Netherlands, ³Donders institute for Brain Cognition and Behaviour, Radboud University, Nijmegen, Netherlands, ⁴Gorter center, Leiden UMC, Leiden, Netherlands, ⁵Futura Composites, Heerhugowaard, Netherlands, ⁶Spinoza Center for Neuroimaging, Amsterdam, Netherlands

Synopsis

Ultrafast gradient insert coils can boost EPI performance when designed as an independent 4^th gradient chain. However, at high operating frequencies, substantial eddy currents will be generated in the shields of RF coils. Here we show that RF shields can be removed without compromising MRI performance yet enabling two orders of magnitude increased gradient efficiency at high frequency.

Introduction

Gradient slew rates limit spatiotemporal resolution of MRI. Recently, it was demonstrated that with dedicated head gradient coil inserts interfaced to state-of-the-art gradient amplifiers, the slew rate can be increased to 900 ¹ or 1200 T/m/s ². In fact, when making the gradient coil resonant, even higher slew rates can be obtained at a fixed frequency of operation, particularly when designed for the z-axis ^3-5. However, at very high operating frequencies (i.e. >10kHz) the efficiency of the gradient coil is severely compromised by eddy currents generated in the RF shield, generally in close proximity to the gradient coil. Here we report for two gradient inserts the gradient efficiency loss due to different types of RF shields, and the efficiency gain in the absence of RF shielding, obtained at 3T, 7T, wide bore, Philips and Siemens systems. Moreover, the RF transparency of the gradient insert is demonstrated for phantoms and the human head.

Methods

Two gradient insert coils were designed, one fully embedded in glass fiber reinforced epoxy including an embedded RF shield (Futura, NL), and one custom-built at lower inductance wound on a Plexiglas former with insertable RF shields. Inductance, resistance and eddy current losses were measured with an LRC device (Agilent, Santa Clara, USA) operated at 100, 500, 1k, 2k, 5k, 10k, and 20kHz. In the final measurement, the RF shield was removed by milling sleeves at every cm through the embedded RF shield. After the bench measurements, the low inductance gradient was made resonant using a parallel capacitance and matched via a series capacitance to 4ohm when positioned outside of the MRI. Its change in impedance was measured when positioned in two 7T systems (Philips, Best, Netherlands) and one wide bore 3T system (Skyra, Siemens, Erlangen, Germany). The RF transparency was measured using the 3T body coil as transceiver and obtaining B1+ maps from a phantom and from the human head at fixed RF power in the absence versus presence of the gradient insert. The final gradient performance of the non-shielded 20kHz resonant Futura coil driven with a commercially available 18kW audio amplifier (Powersoft, Scandicci, Italy) was calculated and measured using field cameras (Skope, Zurich, Switzerland).

Results

The gradient power losses due to eddy currents in the RF shield with the Futura gradient insert (109µH) are 227-fold when comparing 20kHz versus 100Hz of operation (Figure 1a), while the inductance reduces by 30% (self-eddy currents, Figure 1b). For the low inductance gradient coil (39µH) this power loss is 48-fold and inductance reduced by 30%. When removing the shield, the power loss reduces to 15-fold and 2.5-fold respectively and no change in self-inductance. When inserting the non-shielded gradient inside the MRI bore, the 20kHz losses of the Futura coil increased by 2.1-fold when inserting in the 60cm-bore 7T MRI system, likewise for the low inductance gradient coil the 10kHz losses increased by 1.9-fold in the 60cm-bore 7T MRI systems and 1.5-fold in the 70cm-bore 3T MRI. The B1+ level at 3T driven with the same RF power remained similar (difference 3% in volunteer) with or without the gradient insert (figure 2). When driving the Futura insert gradient without RF shield tuned at 20kHz with 18kW, a gradient field of 41mT/m and slew rate of 5200T/m/s was measured (figure 3), matching to the calculated performance.

Discussion and conclusion

A more than 100-fold efficiency gain of high frequency (20kHz) gradient coils can be obtained with practically no self-eddy currents when removing a traditional RF screen from the proximity of the gradient coil. Moreover, when excluding the RF shield from the (low inductance) gradient insert coil, no need for an insert RF transmit coil is needed, as demonstrated by the uncompromised B1+ performance of the RF bodycoil in the presence of the non-shielded gradient insert. The non-shielded highly efficient resonant gradient coil can outperform any commercially available gradient system in terms of fast switching capabilities, even when driven with a (low-cost) audio amplifier as reflected by the slew rate of 5200T/m/s driven by a 18kW amplifier.

Acknowledgements

No acknowledgement found.

References

1. Winkler SA, Schmitt F, Landes H, DeBever J, Wade T, Alejski A, Rutt BK. Gradient and shim technologies for ultra high field MRI. Neuroimage 2016:0–1. doi: 10.1016/j.neuroimage.2016.11.033.

2. Weiger M, Overweg J, Rösler MB, et al. A high-performance gradient insert for rapid and short-T 2 imaging at full duty cycle. Magn. Reson. Med. 2017;00:1–11. doi: 10.1002/mrm.26954.

3. Turner R. Gradient coil design: A review of methods. Magn. Reson. Imaging 1993;11:903–920. doi: 10.1016/0730-725X(93)90209-V.

4. Van der Velden TA, Van Leeuwen CC, Huijing ER, Borgo M, Luijten PR, Klomp DWJ, Siero JCW. (2017), A lightweight gradient insert coil for high resolution brain imaging ISMRM , #4329

5. Van der Velden T, Rivera D, Hendrise A, Siero J, Klomp D. (2018), Method and apparatus for ultrasonic gradients in magnetic resonance imaging, European Patent Application 3364205

Figures

Figure 1: Resistance (a, in mΩ) and total self-inductance (b, in µH) of Futura and low-inductance gradient coils obtained at different frequencies with and without the presence of an RF shield obtained outside the MRI. Note the orders of magnitude increased losses at high frequency compared to low frequency, and the order of magnitude reduction of loss when removing the RF shield.

Figure 2: B1+ maps (in % of nominal B1+) from a phantom (a and b) and from the human head (c and d) obtained either without (a and c) and with (b and d) the presence of the gradient insert obtained at identical RF power levels. Note the similarity in the B1+ maps with a slight increase in B1+ efficiency per unit of RF power when adding the insert gradient.

Figure 3: Demonstration of the efficiency of the resonant Futura gradient insert without a closely positioned RF shield driven with 18kW and measured at 7T using field cameras. Note the high slew rate of 5200T/m/s that can be obtained.

Proc. Intl. Soc. Mag. Reson. Med. 27 (2019)

0268