Abstract
Introduction For partial volume correction in 1H MMR spectroscopy the voxel fraction of brain matter (BM) and cerebrospinal fluid (CSF) can be calculated via biexponential fitting of T2 relaxation of the unsuppressed water signal or via segmentation of a high-resolution structural image. To determine their equivalence, we compared CSF percentages obtained using these two methods in 3 brain regions. Material and methods 113 children were scanned on a 3T Siemens Allegra. The protocol included a T1-weighted 3D EPI-navigated multiecho magnetisation prepared rapid gradient echo (MEMPRAGE) and single voxel spectroscopy (SVS) in the basal ganglia (BG), midfrontal gray matter (MFGM) and peritrigonal white matter (PWM) using an EPI-navigated point-resolved spectroscopy (PRESS) sequence. Water unsuppressed SVS measurements were acquired at TE’s 30 ms, 50 ms, 75 ms, 100 ms, 144 ms, 500 ms and 1000 ms. The voxel water signal S(TE) was quantified using LCModel and modeled as a biexponential function of TE. Percentage CSF ( PV CSF ) in the voxel was estimated according to (1), where S CSF and S BM are the signals from CSF and BM at TE = 0 (1) PV CSF = S CSF S CSF + 0.75 × S BM × 100 The T1-weighted image was used to segment the SVS voxel into GM, WM and CSF in SPM12. Results BG: PV CSF from segmentation was PV CSF was zero in more than half the subjects. A paired t-test showed a significant difference between methods (t = −2.9, p = 0.004). PWM: PV CSF from the segmentation method was PV CSF using the relaxation method was larger (t = −7.4, p = 2e−11). MFGM: in the MFGM many points fell around the y = x line. A paired t-test showed no significant difference between methods (t = 0.9, p = 0.4). Conclusion There is most consistency between methods where the PV CSF is small, and less consistency in voxels with larger amounts of CSF.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call