Item

Abstract

Purpose/Introduction: Despite of the known difficulties (longer T1 and shorter T2 relaxation times, higher SAR and stronger B0 and B1 inhomogeneities), the feasibility of in-vivo 1H MRS at a field strength of 9.4 T has been recently demonstrated in [1]. Since the main advantage of CSI over SVS is providing the additional information about the spatial distribution of metabolites, the aim of this study was to examine the feasibility of CSI in the human brain at a field strength of 9.4 T. Subjects and Methods: In-vivo measurements (3 healthy volunteers) were performed on a whole body 9.4 T MR scanner (Siemens, Erlangen, Germany). CSI data were collected with a custom built 8 channel transmit coil combined with a 24 channel receiving helmet. Spectra were acquired with a modified STEAM sequence (TR=2000 ms, TE=20 ms, TM=11 ms, spectral bandwidth: 4000 Hz, voxel size: 10 mm3 isotropic). The main modification of the pulse sequence consisted in replacing the 90° h-sinc RF pulses with 90° hermite pulses, which allowed keeping the reference voltage below the hardware limitations and minimized the RF power deposition. Raw spectra were quantified with LCModel, version 6.2-2B [2] using a basis set (17 metabolites) simulated in VeSPA software (http://scion.duhs.duke.edu/vespa/). Results: It can be seen that all the spectra are decent in terms of SNR and spectral resolution (Fig. 1). Especially in two regions; between ~2.4 and ~2.6 and ~1.9 and ~2.3 ppm: is the overlap between the resonances of Gln, Glu, NAA and NAAG reduced. Quantification with LCModel software (Fig. 2) confirmed those observations. Peaks of NAA, NAAG and Glu were assigned with Cramer-Rao lower bounds (CRLB) below 13. Moreover, 13 metabolites of the 17 included in the basis set were quantified with CLRB below 20. Comparable results have been obtained for the other volunteers. Discussion/Conclusion: It has been demonstrated that 1H MRSI at a field strength of 9.4 T is feasible. All acquired spectra demonstrated good agreement in terms of spectral quality with the results published previously [1, 3]. In conclusion, the enhanced sensitivity in the detection of metabolite resonances with CSI at ultra-high fields combined with the additional information about spatial distribution of the metabolites can provide new insights to the assessment of various pathologies.