Correcting frequency and phase offsets in MRS data using robust spectral registration.

Area of Science:

• Neuroimaging
• Spectroscopy
• Biomedical Engineering

Background:

• Magnetic Resonance Spectroscopy (MRS) is crucial for in vivo metabolite quantification.
• Frequency and phase offsets are common challenges in MRS data acquisition.
• Existing methods for MRS data correction can be insufficient, especially with complex editing techniques.

Purpose of the Study:

• To introduce and validate a novel algorithm, robust spectral registration (rSR), for retrospective correction of frequency and phase offsets in MRS data.
• To demonstrate the effectiveness of rSR in aligning individual transients and improving spectral quality.
• To compare the performance of rSR against existing correction methods.

Main Methods:

• rSR employs subroutines for robust alignment of transients, handling frequency/phase offsets and signal contamination.
• Automated removal of lipid and water signals is performed prior to alignment.
• Time-domain correction uses nonlinear least-squares optimization to align transients to a weighted average reference.
• Frequency-domain alignment targets subtraction artifacts in edited datasets.
• Weighted averaging down-weights lower-quality transients during signal averaging.

Main Results:

• rSR successfully aligned complex multiplexed edited MRS data.
• The algorithm demonstrated improved spectral quality, particularly in datasets with significant distortion.
• rSR reduced subtraction artifacts more effectively than a previous multistep method in pediatric GABA-/GSH-edited HERMES datasets (64% for GABA, 75% for GSH).

Conclusions:

• rSR provides a robust and effective solution for frequency and phase correction in MRS.
• The algorithm overcomes major challenges in MRS data processing, enhancing spectral quality and reliability.
• rSR shows significant promise for improving the analysis of MRS data, especially in pediatric studies.