Joonsung Lee1, Borjan Gagoski1, Michale Hamm2, Elfar Adalsteinsson1,3
1Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, USA; 2Siemens Medical Solutions USA, Inc, Cambridge, MA, USA; 3Harvard-MIT Division of Health Sciences of Technology, Massachusetts Institute of Technology, Cambridge, MA, USA
Estimates of metabolites in the brain using large-FOV CSI are severely hampered by strong interfering lipid signals arising from subcutaneous tissue. A novel approach using variable-density sampling in k-space with accompanying proportional filtering is presented. Healthy volunteers were scanned on a 3T Siemens Trio scanner with a 32-channel head coil. The variable spiral trajectory was implemented with a conventional PRESS excitation of 3-cm thick axial section with a voxel size of 0.47 cc, and FOVxy = 24 cm, FOVz= 6.2cm, in a scan time of 16 min. Even in peripheral regions of the cortex, NAA is distinguishable from the lipids.