All studies were performed with a 4T magnet (Oxford Magnet Technology, Oxford, UK) with an INOVA console (Varian, Palo Alto, CA) as described before (3 (link),12 (link)). The scanner was equipped with a gradient system (40 mT/m, 400 μs rise-time), which included second-order shim coils with maximum shim strengths of XZ = YZ = 1.1 mT/m, Z2 = 2.8 mT/m, and XY = X2Y2 = 0.5 mT/m. Spectra from the occipital lobe were acquired with a quadrature 14 cm half volume coil (19 (link)) and spectra from the cerebellum and brainstem with a TEM volume coil (20 (link)). Sagittal, coronal and transverse multi-slice images were obtained with a fast spin echo sequence (repetition time TR = 4.5 s, echo train length = 8, echo spacing = 15 ms, 9 slices, 2 averages) for volume selection. All first- and second-order shims were adjusted using FASTMAP with EPI readout (21 (link)). Proton spectra from the occipital cortex (2.5 × 2.5 × 2.5 cm3), vermis (1.0 × 2.5 × 2.5 cm3), cerebellar hemisphere (1.7 × 1.7 × 1.7 cm3) and pons (1.6 × 1.6 × 1.6 cm3) were acquired both with the redesigned semi-LASER sequence (TR/TE = 4500/24 ms, spectral width = 6000 Hz, number of points = 4096, Fig. 1 ) and a STEAM sequence (TR/TM/TE = 4500/42/5 ms, spectral width = 6000 Hz, number of points = 4096) (11 (link)). Unsuppressed water spectra acquired from the same VOI were used to correct for eddy current effects and as an internal reference for metabolite quantification. The internal reference water spectrum was acquired without the OVS pulses to avoid magnetization transfer (MT) effects. Acquiring the water spectrum without the OVS resulted in 2–4% increase of signal in phantom measurements (without MT effects), while a ~37% higher water signal intensity was obtained without OVS relative to with OVS in vivo, indicating that ~34% of this increase was due to the removal of the MT affect on water.
Data were acquired in single scan mode, i.e. each single FID was saved separately. When the SNR was sufficient, these individual FIDs were frequency and phase corrected. When the SNR of single scan data was not sufficient (cerebellar and brainstem STEAM data), data were averaged over 4 – 8 scans and then corrected for frequency variations. Single scan or averaged data blocks that showed evidence for substantial motion (out-of-phase and/or lower intensity signal as determined by the operator) were excluded from final summation. Motion artifacts were rarely observed in this cohort of healthy volunteers.
Data were acquired in single scan mode, i.e. each single FID was saved separately. When the SNR was sufficient, these individual FIDs were frequency and phase corrected. When the SNR of single scan data was not sufficient (cerebellar and brainstem STEAM data), data were averaged over 4 – 8 scans and then corrected for frequency variations. Single scan or averaged data blocks that showed evidence for substantial motion (out-of-phase and/or lower intensity signal as determined by the operator) were excluded from final summation. Motion artifacts were rarely observed in this cohort of healthy volunteers.
