Meeting Banner
Abstract #2036

Removal of Lipid Nuisance Signals in MRSI Using a Spatial-Spectral Lipid Model

Chao Ma1, 2, Fan Lam1, 3, Ryan Larsen3, Bradley P. Sutton, 34, Zhi-Pei Liang, 35

1Department of Electrical and Computer Engineering , University of Illinois at Urbana-Champaign, Urbana, IL, United States; 2Beckman Institute, University of Illinois at Urbana-Champaign, Urbana, IL, United States; 3Beckman Institute, University of Illinois at Urbana-Champaign, Urbana, IL, United States; 4Department of Bioengineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States; 5Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States

Nuisance lipid signals from the subcutaneous lipid layer of the brain often cause significant difficulties for spectral quantification of the brain metabolites. Removal of the lipid signals in brain MRSI is desirable but challenging because they appear as multiple-peak, broad spectra that overlap with the spectra of important brain metabolites (e.g., lactate and NAA). In this work, we introduce a model-based post-processing method for effective removal of the lipid signals.

Abstract PDF

Keywords

acquisition address advance advantages almost anatomical another appear assigned assuming basis bioengineering brain broad cause challenging champaign close coefficients completely computer constraints coverage dataset denotes designed desirable difficulties dong effective effectively electrical engineering evaluated example excitation experiment extended field finally fourth frequency function glover harmonics important in vivo incorporated incorporates inhomogeneity institute introduce inversion known lactate layer least limited linkage lipid lipids metabolite metabolites model models norm novel nuisance often optimization original outer overlap peak pixel post priori problem procedure processing propose proposed protect prove quality quantification reconstruct reconstruction recovery removal remove removed represent representative residual resolution respectively retrieval selective setting significantly simulation simultaneously solving space spatial spectra spectral spectrum squares subcutaneous subtracted support suppress suppressed suppression taking temporal third tissue transform types useful variety volume