Quantitative metabolic reference for healthy human cerebrum derived from group averaged 9.4T 1H MRSI data

Drawing inspiration from previous works using 1H FID MRSI, this study quantifies metabolite concentrations at 9.4 T in the human cerebrum of a volunteer cohort and performs a respective group analysis to derive region specific metabolite concentrations. Voxel-specific corrections were performed for both water and individual metabolites, as well as used tissue specific T1-relaxation times. Anatomical and magnetic resonance spectroscopic imaging data were collected using MP2RAGE and FID MRSI sequences, and subsequent data underwent a series of preprocessing techniques. Results showed consistent metabolite maps for key metabolites (NAA, tCr, Glu, tCho and mI), while instability in data quality was noted for lower slices. This study not only showcases the potential of metabolite quantification and mapping at 9.4 T but also underscores the necessity for meticulous data processing to ensure accurate metabolite representations. Comparisons with earlier works and single voxel results validate the methodologies adopted. HighlightsO_LIQuantitative whole cerebrum metabolite maps in the human brain acquired at 9.4 T. C_LIO_LIGroup averaged regional metabolite concentrations. C_LIO_LIGroup summed regional metabolite spectra. C_LI