Joonsung Lee1, Elfar Adalsteinsson1,2
1Electrical Engineering & Computer Science, Massachusetts Institute of Technology, Cambridge, MA, United States; 2Harvard-MIT Division of Health Sciences &, Massachusetts Institute of Technology, Cambridge, MA, United States
Estimates of brain metabolites using MRSI are severely hampered by strong lipid signals even though the interfering signal arises from regions outside of the brain. Variable-density sampling of k-space with accompanying proportional filtering reduces the spatial side-lobes within the brain due to the extra-brain lipid sources. We present an improved algorithm that achieves a direct design of a subject-specific, optimal, spherically symmetric spatial filter in two or three dimensions to minimize lipid contamination estimated by pre-scans. The proposed method is demonstrated with single-slice MRSI to achieve 30~40 dB or lipid suppression in the absence of OVS or inversion recovery pulses.