Major neuroscience research centers are now equipped with MRI scanners featuring increasingly powerful magnetic fields. As the undisputed leader in this international race, NeuroSpin has housed the ISEULT MRI scanner, with its unmatched 11.7 Tesla magnetic field, for several years.

This technological leap aims, on the one hand, to increase the signal-to-noise ratio of images, thereby enabling the observation of details invisible at lower field strengths, as demonstrated by the first brain images acquired with ISEULT and published in 2024.

However, the expected gains must also benefitfunctional imaging (fMRI) by enhancing sensitivity to the BOLD contrast, which reflects local blood flow variations and, indirectly, brain activity. Yet, this requires overcoming the specific challenges associated with such a high magnetic field. Notably, since BOLD variations represent only asmall percentage of the signal, temporal signal stability is critical. At 11.7 T, this stability is compromised by Radiofrequency field inhomogeneity, B0 field inhomogeneity, Field variations induced by motion and vibrations.​

The team led by Nicolas Boulant (METRIC / BAOBAB / NeuroSpin), in collaboration with Florent Meyniel's team (Computational brain / UNICOG / NeuroSpin) the German Center for Neurodegenerative Diseases (DZNE Bonn), ETH Zurich, and the University of Glasgow, has published the first resting-state and task-based fMRI imagesacquired with ISEULT.

To conduct this study, the researchers optimized the acquisition sequence using:

• Parallel transmission,
• Reshaping of the signals sent to the gradient coil to avoid vibrations and, consequently, magnetic field disturbances,
• Effective correction of residual head movements and improvement of the signal-to-noise ratio, thanks to a real-time motion correction technique (servo-nav), and a post-processing technique (PEERs) aimed at stabilizing the signal and compensating for certain limitations of prospective correction.​

Overall, these results demonstrate the feasibility and reliability of fMRI at 11.7 T and represent an essential first step in quality control, paving the way for further increasing resolution at this unprecedented field strength.​

​

Contact at ​Frédéric-Joliot Institute for Life sciences:

Nico​las Boulant (nicolas.boulant@cea.fr) 

This text was translated with the assistance of Mistral AI.

European funding
This work was carried out within the framework of the​ "AROMA" coordinated by​ NICOLAS BOULANT.


French government PIA3 funding
The project received financial support from PRESENCE (under the "Structuring Equipment for Research: EquipEx+" initiative​").