Robust retrospective frequency and phase correction for single-voxel MR spectroscopy

Area of Science:

• Magnetic Resonance Imaging and Spectroscopy
• Biomedical Signal Processing

Background:

• Subject motion and static magnetic field (B0) drift degrade single-voxel MR spectroscopy (SVS) data quality.
• Retrospective correction methods adjust phase and frequency but struggle with large shifts and baseline instability.

Purpose of the Study:

• Develop a novel method (RATS) for retrospective correction of SVS MRS data.
• Enhance tolerance to large frequency shifts (>7 Hz) and baseline instability.

Main Methods:

• RATS incorporates variable-projection and baseline fitting for robustness.
• Compared RATS with time-domain spectral registration (TDSR) using simulated and in-vivo data (glutathione-edited MRS).

Main Results:

• RATS demonstrates improved accuracy and stability for large frequency shifts and unstable baselines.
• Reduced subtraction artifacts observed in RATS-corrected glutathione-edited MRS compared to uncorrected or TDSR-corrected spectra.

Conclusions:

• RATS provides accurate retrospective correction for SVS MRS data with significant frequency shifts and baseline instability.
• The RATS algorithm is rapid, generic, and easily integrated into MRS processing pipelines to improve lineshape, SNR, and quality assessment.