0475

A 48-channel multi-coil shim array for B0 inhomogeneities correction in the NHP brain at 7T.
Elias Djaballah1, Anojan Uthayakumar2, Martin Bouchet 1, André Kalouguine1, Alexis Amadon2, and Qi ZHU1
1Cognitive Neuroimaging Unit, INSERM, CEA, Université Paris-Saclay, NeuroSpin Center, GIF-SUR-YVETTE, France, 2BAOBAB, Université Paris-Saclay, CEA/Joliot/NeuroSpin, GIF-SUR-YVETTE, France

Synopsis

Keywords: Shims, High-Field MRI, non-human primate, B0 shimming, fMRI

Motivation: Functional MRI faces challenges due to B0 field inhomogeneities, which are amplified in non-human primate studies, especially at ultra-high fields.

Goal(s): Our aim was to develop a Multi-Coil Array (MCA) specifically designed for whole brain shimming of non-human primates.

Approach: A semi-heuristic approach was used to design an MCA, based on the Principal Components Analysis of subject-optimal stream functions. The MCA was then constructed and used to image an in-vivo NHP brain.

Results: The designed MCA effectively reduced B0 inhomogeneities in the primate brain, resulting in significant improvements in B0 homogenization and enhanced image quality in functional MRI images at 7T.

Impact: The development of a Multi-Coil Shim Array designed specifically for non-human primates significantly improves functional MRI image quality by effectively reducing B0 inhomogeneities. This advancement could pave the way for more accurate and detailed primate brain studies at ultra-high fields.

Introduction

Static field inhomogeneities pose a challenge for brain MRI at high fields as they become more pronounced with the increased main B0 field. They cause enhanced Echo-Planar image distortions and signal losses due to magnetic susceptibility gradients at air-tissue interfaces within the head and become a major issue in functional MRI studies. The main technology used to homogenize the magnetic field in the brain is active shimming with Multi-Coil Arrays (MCA), which involves generating a counteracting magnetic field using coils to reduce B0 inhomogeneities in the brain (1,2,3). However, with Ultra-High Fields (UHF) imaging, 2nd or 3rd order Spherical Harmonic (SH) shim systems typically offered by MRI manufacturers are insufficient to provide artifact-free high-resolution acquisitions. In addition, these systems are usually designed for humans and do not cater to the small size and specific position of non-human primates (NHPs) while they undergo imaging in an MRI scanner. As a result, brain activation maps obtained through EPI sequences in NHP functional studies are highly sensitive to B0 inhomogeneities in the primate brain.
Here we present a specifically designed MCA optimized for whole brain shimming of macaques, adapted to an awake acquisition setup chair. Based on a cylindrical structure, it features several layers of small coils whose shape, size and location are determined from Singular Value Decomposition (SVD) of ideal Stream Functions (SF) computed from multiple macaque brain field maps (4). The B0 shim performances were assessed in simulations and compared to in vivo acquisitions at 7T.

Methods

To design an MCA adapted to the inhomogeneous field patterns encountered in the macaque brain, a semi-heuristic approach was employed for optimization of channels’ geometry and position. It is based on the Principal Components Analysis (PCA) of subject-optimal stream functions (SO-SFs), which provide indications of electric current distribution over a macaque brain for B0 shimming (4,5,6).
11-subject 𝛿𝐵0 brain field maps were acquired, then cylindrical SO-SFs were computed for each subject, using the Dipole Boundary Method (DBM) (4). SVD was performed on the SO-SFs set. The cylinders used as coil formers consider the mechanical constraints imposed by the awake NHP setup, centered at the magnet center. From the first PCA components, indications of where electric current are most demanded for primate brain B0 shimming were obtained, and the MCA was designed based on these indications. For performance assessment and validation, shimming simulations with the obtained design were performed on field maps not used for the MCA B0 shim coil computations.
The 48-channels B0 shim array was then constructed using fiberglass cylinders. The 20-turns coils were assembled onto their specific position with epoxy resin. The B0 shim array was controlled via an Arduino board, by an opensource multi-channel feedback-controlled current driver (7), limited to 3A per channel. A control interface in MATLAB was used to communicate with the Arduino board and transfer ideal current setpoints to the current driver. Current calculation was performed through joint optimization with scanner’s 1st and 2nd order SH coefficients. The driving system, originally made for controlling a human brain B0 shim array (SCOTCH (4)), was adapted to control this NHP brain shim array.
Acquisitions were conducted in vivo on a macaque in a 7T siemens MAGNETOM scanner using a single loop transmitter and a 16-loop receive antenna. B0 field maps were acquired with a 1.5mm isotropic AFI sequence (TR=15ms, TEs=1.8,6.2,8.5ms, flip-angle=7°) and EPI images were acquired with a 0.6mm isotropic 2D SMS-EPI sequence (TR=3s, TE=20ms, IPAT=3, phase encoding direction=AP, multiband factor=2).

Results

Fig.2 shows simulated 𝛿B0 field maps with and without the NHP B0 shimming. It demonstrates a clear B0 homogenization and predicts an up to 40% improvement with the NHP B0 shimming.
Fig.3a shows acquired 𝛿B0 field maps with and without the NHP B0 shimming. Experimental results demonstrate that the NHP B0 shimming yielded a 38% improvement in addition to the scanner shim, reducing the 𝛿B0 standard deviation from 62.0Hz without the NHP B0 shimming to 36.8 Hz with it. The predicted 𝛿B0 map (Fig.3b) shows a standard deviation of 34.7Hz. The spherical harmonic rate(4) is 5, with 22.7W dissipated in MCA during shimming.
Fig.4 shows two volumes of acquired 0.6mm isotropic EPI on a macaque at 7T and depicts a qualitative image improvement, reducing B0 inhomogeneity-induced artifacts.

Conclusion

We developed a Multi-Coil Array (MCA) for whole brain shimming of non-human primates, optimized based on SO-SFs to effectively reduce B0 inhomogeneities in the primate brain. Simulations and in vivo acquisitions demonstrate significant improvements in B0 homogenization, resulting in enhanced image quality and reduced artifacts in fMRI images of non-human primates at UHF.

Acknowledgements

We acknowledge the financial support of the French national research agency (ANR) under the reference ANR-20-CE37-0005 and the exploratory program of CEA, the French Alternative Energies and Atomic Energy Commission

References

[1] Juchem, C., Nixon, T.W., McIntyre, S., Rothman, D.L., Graaf, R.A.d., 2010. Magnetic field homogenization of the human prefrontal cortex with a set of localized electrical coils. Magn. Reson. Med. 63 (1), 171–180. doi:10.1002/mrm.22164.

[2] Aghaeifar, A., Zhou, J., Heule, R., Tabibian, B., Schölkopf, B., Jia, F., Zaitsev, M., Scheffler, K., 2020. A 32-channel multi-coil setup optimized for human brain shimming at 9.4T. Magn. Reson. Med. 83 (2), 749–764. doi:10.1002/mrm.27929.

[3] Koch, K.M., Brown, P.B., Rothman, D.L., de Graaf, R.A., 2006. Sample-specific diamagnetic and paramagnetic passive shimming. J. Magn. Reson. 182 (1), 66–74. doi:10.1016/j.jmr.2006.06.013.

[4] Pinho Meneses, B. et al. Shim coils tailored for correcting B0 inhomogeneity in the human brain (SCOTCH): Design methodology and 48-channel prototype assessment in 7-Tesla MRI. NeuroImage 261, 119498 (2022).

[5] Cai, Y., Yang, H.-J., Li, X., Hu, T., Huang, Y., Shan, Y., Lu, M., Liu, W.S., Li, D., Han, H., 2020. Optimization of multi-coil array design for efficient human brain shimming at 3T. In: Proceedings of the 28th Annual Meeting of the International Society for Magnetic Resonance in Medicine. ISMRM, Virtual, p. 4229.

[6] F., Elshatlawy, H., Aghaeifar, A., Chu, Y., Hsu, Y., Littin, S., Kroboth, S., Yu, H., Amrein, P., Gao, X., Yang, W., LeVan, P., Scheffler, K., Zaitsev, M., 2020. Design of a shim coil array matched to the human brain anatomy. Magn. Reson. Med. 83 (4), 1442–1457. doi:10.1002/mrm.28016.

[7] Arango, N., Stockmann, J.P., Witzel, T., Wald, L.L., White, J., 2016. Open-source, lowcost, flexible, current feedback-controlled driver circuit for local B0 shim coils and other applications. In: Proc. Int. Soc. Magn. Reson. Med, Vol. 24, p. 1157.

Figures

Figure 1: a) 48ch NHP brain B0 shim system b) B0 MCA shim with an 1Tx/16Rx antenna c) 3D Model of computed MCA

Figure 2: Simulated 𝛿B0 field maps a) with only the scanner shimming and b) with NHP MCA + the scanner shimming.

Figure 3: a) Experimental 𝛿B0 field maps with and without additional NHP MCA B0 shimming b) Predicted 𝛿B0 field maps computed with applied MCA currents

Figure 4: EPI images of an NHP using a 1Tx/16Rx antenna at 7T with a 0.6mm isotropic resolution. The red outlines represent the brain edges in the B0 shimmed case.

Proc. Intl. Soc. Mag. Reson. Med. 32 (2024)
0475
DOI: https://doi.org/10.58530/2024/0475