Commissioning of the Iseult CEA 11.7 T whole-body MRI: current status, gradient–magnet interaction tests and first imaging experience

Abstract

Objectives The Iseult MRI is an actively shielded whole-body magnet providing a homogeneous and stable magnetic field of 11.7 T. After nearly 20 years of research and development, the magnet successfully reached its target field strength for the first time in 2019. This article reviews its commissioning status, the gradient–magnet interaction test results and first imaging experience. Materials and methods Vibration, acoustics, power deposition in the He bath, and field monitoring measurements were carried out. Magnet safety system was tested against outer magnetic perturbations, and calibrated to define a safe operation of the gradient coil. First measurements using parallel transmission were also performed on an ex-vivo brain to mitigate the RF field inhomogeneity effect. Results Acoustics measurements show promising results with sound pressure levels slightly above the enforced limits only at certain frequency intervals. Vibrations of the gradient coil revealed a linear trend with the B0 field only in the worst case. Field monitoring revealed some resonances at some frequencies that are still under investigation. Discussion Gradient-magnet interaction tests at up to 11.7 T are concluded. The scanner is now kept permanently at field and the final calibrations are on-going to pave the road towards the first acquisitions on volunteers.

Scaling the mountains: what lies above 7 Tesla magnetic resonance?

“…This is in stark contrast to scan durations of approximately 1-2 h available today for the same sub-millimetre spatial resolution [45][46][47]. MR of nuclei such as 35 Cl and 39 K will greatly benefit from 14 and 20 T. Arguably, 39 K MR remains quite challenging because the sensitivity is about six orders of magnitude less than that of 1 H MR [48]. Notwithstanding this constraint, it stands to reason that 39 K MR at 14 T or 20 T will enable, for the first time, quantitative in vivo assessment of myocardial potassium content on a cellular level, which would open an entirely new research field of MRI-based physio-metabolic fingerprinting as a link to personalised medicine, as pointed out in this Special Issue in [43].…”

“…As we move up the mountain face, we should pause for a moment to acknowledge and applaud the pioneers who climbed the very steep mountains to 9.4 T [33], 10.5 T [8,37,38] and 11.7 T [39]. These explorers were essentially free climbing, reaching the summit without vendor-provided RF coils and without push-button software, yet nevertheless navigating through the enormous challenges towards beautiful images of the brain and the body enabled by parallel transmission.…”

“…These explorers were essentially free climbing, reaching the summit without vendor-provided RF coils and without push-button software, yet nevertheless navigating through the enormous challenges towards beautiful images of the brain and the body enabled by parallel transmission. First MR images obtained at 11.7 T of an ex vivo brain are reported in this Special Issue, and the imaging community is excited to hear about the next achievements [39].…”

Power deposition in the He bath induced by the Z gradient coil in the 11.7 T Iseult magnet: theory versus measurements

“…But while the 7000 l volume of helium and the 1.8 K temperature in the Iseult magnet provide a large safety buffer for exploitation, on the other hand it requires time-consuming gradient frequency sweeps to cover a large frequency range and obtain the desired helium boil-off spectra, which can be demanding for the gradient coil and gradient cables as well. For this reason, the measurements presented here employed a sweep rate of 50 Hz h −1 and focused on the frequency interval 1300-2000 Hz over which the main cryogenic peaks could be detected in some previous work [14,15]. For the theoretical predictions, calculations were run at fixed frequencies every 1 Hz using optimized routines in Fortran.…”

“…One of the fundamental and key assumptions of our model is that the gradient coil emits an (imperfectly shielded) magnetic vector potential which is not affected by the gradient-magnet interactions, and constitutes the source of the perturbations. Its vibration, measured to be hundreds of µm at the most [14] and thus small compared to the wavelengths at a few kHz, is thus assumed to not perturb the fields. The gradient coil also is assumed to be perfectly decoupled mechanically from the surrounding structures.…”