Fourier-based decomposition for simultaneous 2-voxel MRS acquisition with 2SPECIAL

Area of Science:

• Magnetic Resonance Imaging
• Spectroscopy
• Neuroimaging

Background:

• Simultaneous multi-voxel spectroscopy (sMVS) is crucial for efficient brain imaging.
• Existing methods can be time-consuming, limiting clinical applications.
• Advanced pulse sequences and decomposition algorithms are needed to improve efficiency.

Purpose of the Study:

• To develop and validate a simultaneous multi-voxel spectroscopy (sMVS) technique at 7 Tesla using a novel multi-banded sequence.
• To introduce and assess a voxel-GRAPPA (vGRAPPA) decomposition method for accurate signal separation.
• To evaluate the performance and limitations of the proposed 2SPECIAL sequence and vGRAPPA algorithm.

Main Methods:

• Implementation of a 2 spin-echo, full-intensity acquired localized (2SPECIAL) sequence with a wideband, uniform rate, smooth truncation (WURST) multi-banded pulse.
• Development of a voxel-GRAPPA (vGRAPPA) algorithm for signal decomposition and comparison with SENSE-based methods.
• Evaluation of two-voxel excitation, adiabatic inversion pulses, and B0/B1 adjustments for in vivo brain applications.

Main Results:

• The 2SPECIAL sequence enabled sMVS with a 21.6% reduction in scan time compared to consecutive acquisitions, maintaining signal-to-noise ratio (SNR).
• The vGRAPPA algorithm accurately reassigned spectral signals to their origin voxels, showing comparable performance to SENSE regarding leakage (<10%) and Cramér-Rao lower bounds (CRLB).
• vGRAPPA reduced motion sensitivity by eliminating the need for additional sensitivity map acquisitions.

Conclusions:

• The 2SPECIAL sequence combined with vGRAPPA offers a significant reduction in measurement time for multi-voxel spectroscopy.
• This approach maintains spectral quality and metabolite quantification accuracy.
• It presents a valuable new option for advanced multi-voxel applications in neuroimaging.