Dynamic 2D self-phase-map Nyquist ghost correction for simultaneous multi-slice echo planar imaging.

Area of Science:

• Magnetic Resonance Imaging
• Neuroimaging

Background:

• Simultaneous multislice echo planar imaging (SMS-EPI) accelerates data acquisition in functional MRI (fMRI).
• Nyquist ghosting is a common artifact in EPI data that can obscure true signals.
• Accurate reconstruction is crucial for reliable fMRI results.

Purpose of the Study:

• To develop a novel reconstruction pipeline for SMS-EPI data.
• To intrinsically correct for dynamic, slice-specific Nyquist ghosting.
• To improve the quality of time-series fMRI data.

Main Methods:

• A model-based framework was employed for joint reconstruction of SMS-EPI data.
• The method estimates a 2D self-phase map to correct dynamic slice-specific phase errors.
• Data from all coils and echoes were combined after unaliasing using slice GeneRalized Autocalibrating Partial Parallel Acquisition (sGRAPPA).

Main Results:

• The proposed dynamic 2D phase correction reduced ghost-to-signal ratios (GSRs) to 1.37 ± 0.11%, compared to 2.66 ± 0.16% for standard 1D correction.
• Temporal variations in GSR were also reduced.
• The reconstruction introduced minimal smoothing in fMRI activation maps, comparable to a 1x1x1 mm³ Gaussian kernel.

Conclusions:

• The developed reconstruction pipeline effectively reduces slice-specific phase errors and Nyquist ghosting in SMS-EPI.
• This method is suitable for fMRI studies, as regularization effects on activation maps are minimal.
• Improved image quality facilitates more accurate analysis of brain activity.