Exploring in vivo human brain metabolism at 10.5 T: Initial insights from MR spectroscopic imaging
Lukas Hingerl,
Bernhard Strasser,
Simon Schmidt,
Korbinian Eckstein,
Guglielmo Genovese,
Edward J. Auerbach,
Andrea Grant,
Matt Waks,
Andrew Wright,
Philipp Lazen,
Alireza Sadeghi-Tarakameh,
Gilbert Hangel,
Fabian Niess,
Yigitcan Eryaman,
Gregor Adriany,
Gregory Metzger,
Wolfgang Bogner,
Małgorzata Marjańska
Affiliations
Lukas Hingerl
High-field MR Center HFMR, Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria
Bernhard Strasser
High-field MR Center HFMR, Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria
Simon Schmidt
Center for Magnetic Resonance Research, Department of Radiology, University of Minnesota, Minneapolis, USA
Korbinian Eckstein
The University of Queensland, School of Information Technology and Electrical Engineering, St Lucia, Australia
Guglielmo Genovese
Center for Magnetic Resonance Research, Department of Radiology, University of Minnesota, Minneapolis, USA
Edward J. Auerbach
Center for Magnetic Resonance Research, Department of Radiology, University of Minnesota, Minneapolis, USA
Andrea Grant
Center for Magnetic Resonance Research, Department of Radiology, University of Minnesota, Minneapolis, USA
Matt Waks
Center for Magnetic Resonance Research, Department of Radiology, University of Minnesota, Minneapolis, USA
Andrew Wright
Advanced Imaging Research Center, University of Texas Southwestern Medical Center, Dallas, USA
Philipp Lazen
High-field MR Center HFMR, Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria; Department of Neurosurgery, Medical University of Vienna, Vienna, Austria
Alireza Sadeghi-Tarakameh
Center for Magnetic Resonance Research, Department of Radiology, University of Minnesota, Minneapolis, USA
Gilbert Hangel
High-field MR Center HFMR, Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria; Department of Neurosurgery, Medical University of Vienna, Vienna, Austria; Christian Doppler Laboratory for MR Imaging Biomarkers, Vienna, Austria
Fabian Niess
High-field MR Center HFMR, Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria
Yigitcan Eryaman
Center for Magnetic Resonance Research, Department of Radiology, University of Minnesota, Minneapolis, USA
Gregor Adriany
Center for Magnetic Resonance Research, Department of Radiology, University of Minnesota, Minneapolis, USA
Gregory Metzger
Center for Magnetic Resonance Research, Department of Radiology, University of Minnesota, Minneapolis, USA
Wolfgang Bogner
High-field MR Center HFMR, Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria; Christian Doppler Laboratory for MR Imaging Biomarkers, Vienna, Austria; Corresponding authors.
Małgorzata Marjańska
Center for Magnetic Resonance Research, Department of Radiology, University of Minnesota, Minneapolis, USA; Corresponding authors.
Introduction: Ultra-high-field magnetic resonance (MR) systems (7 T and 9.4 T) offer the ability to probe human brain metabolism with enhanced precision. Here, we present the preliminary findings from 3D MR spectroscopic imaging (MRSI) of the human brain conducted with the world's first 10.5 T whole-body MR system. Methods: Employing a custom-built 16-channel transmit and 80-channel receive MR coil at 10.5 T, we conducted MRSI acquisitions in six healthy volunteers to map metabolic compounds in the human cerebrum in vivo. Three MRSI protocols with different matrix sizes and scan times (4.4 × 4.4 × 4.4 mm³: 10 min, 3.4 × 3.4 × 3.4 mm³: 15 min, and 2.75×2.75×2.75 mm³: 25 min) were tested. Concentric ring trajectories were utilized for time-efficient encoding of a spherical 3D k-space with ∼4 kHz spectral bandwidth. B0/B1 shimming was performed based on respective field mapping sequences and anatomical T1-weighted MRI were obtained. Results: By combining the benefits of an ultra-high-field system with the advantages of free-induction-decay (FID-)MRSI, we present the first metabolic maps acquired at 10.5 T in the healthy human brain at both high (voxel size of 4.4³ mm³) and ultra-high (voxel size of 2.75³ mm³) isotropic spatial resolutions. Maps of 13 metabolic compounds (aspartate, choline compounds and creatine + phosphocreatine, γ-aminobutyric acid (GABA), glucose, glutamine, glutamate, glutathione, myo-inositol, scyllo-inositol, N-acetylaspartate (NAA), N-acetylaspartylglutamate (NAAG), taurine) and macromolecules were obtained individually. The spectral quality was outstanding in the parietal and occipital lobes, but lower in other brain regions such as the temporal and frontal lobes. The average total NAA (tNAA = NAA + NAAG) signal-to-noise ratio over the whole volume of interest was 12.1± 8.9 and the full width at half maximum of tNAA was 24.7± 9.6 Hz for the 2.75 × 2.75 × 2.75 mm³ resolution. The need for an increased spectral bandwidth in combination with spatio-spectral encoding imposed significant challenges on the gradient system, but the FID approach proved very robust to field inhomogeneities of ∆B0 = 45 ± 38 Hz (frequency offset ± spatial STD) and B1+ = 65 ± 11° within the MRSI volume of interest. Discussion: These preliminary findings highlight the potential of 10.5 T MRSI as a powerful imaging tool for probing cerebral metabolism. By providing unprecedented spatial and spectral resolution, this technology could offer a unique view into the metabolic intricacies of the human brain, but further technical developments will be necessary to optimize data quality and fully leverage the capabilities of 10.5 T MRSI.
