UIC-2013-129 – Method for Reducing Chemical Shift Displacement Errors in 2D MR Spectroscopy (sLASER-first-COSY)

Magnetic Resonance (MR) spectroscopy is a non-invasive diagnostic test that measures biochemical changes in the brain to differentiate between different neurological conditions and tumor types. Two-dimensional (2D) localized chemical shift correlated spectroscopy (L-COSY) is one of the simplest and the most useful methods applied for 2D MR spectroscopy. Conventional 2D L-COSY consists of a slice-selective 90° radiofrequency (RF) excitation pulses followed by a slice-selective 180° refocusing pulse and a terminal slice selective 90° mixing pulse. Due to limited bandwidths of the routine 90° and 180° pulses, L-COSY has relatively large chemical shift displacement errors. This issue is more severe at higher magnetic field strengths such as 3 Tesla and 7 Tesla. To mitigate the impacts of this issue, an adiabatic localized correlation spectroscopy (AL-COSY) using semi-localization by adiabatic selective refocusing (semi-LASER) was proposed by one group to suppress the chemical shift displacement errors. However, because the second 90° pulse (the mixing pulse for correlation spectroscopy) is slice selective in both L-COSY and AL-COSY, the limited bandwidth (BW) of this second 90° pulse will induce spatially dependent magnetization transfer that results in attenuated cross-peaks in 2D L-COSY and AL-COSY spectra. In view of the aforementioned limitations inherent in conventional 2D L-COSY and AL-COSY, UIC inventors have created an improved 2D adiabatic localized correlation spectroscopy using semi-LASER (sLASER) for voxel localization but with the slice-selective first 90° pulse and the non-slice selective second 90° mixing pulse to solve the problem of spatially dependent magnetization transfer. This novel technique is called “sLASER-first-COSY spectroscopy.” Phantom, ex vivo, and in vivo human brain experiment have successfully demonstrated that sLASER-first-COSY yields stronger cross peaks and higher ratios of cross peak volumes to diagonal peak volumes, compared with conventional L-COSY and AL-COSY methods. Jonathan Gortat jgortat@otm.uic.edu (312) 413-1643