Frame-wise multi-echo distortion correction for superior functional MRI

Area of Science:

• Neuroimaging
• Magnetic Resonance Imaging (MRI)
• Data Processing

Background:

• Functional MRI (fMRI) data suffer from magnetic field (B0) inhomogeneities causing geometric distortions.
• Existing correction methods require separate field map acquisitions, which are sensitive to participant head motion.
• Head motion during field mapping corrupts data, hindering accurate distortion correction.

Purpose of the Study:

• To develop and validate a dynamic, frame-by-frame distortion correction method for fMRI data.
• To leverage multi-echo (ME) fMRI data to correct for magnetic field distortions caused by head motion.
• To assess the impact of motion-corrected fMRI data on resting-state functional connectivity (RSFC) and anatomical alignment.

Main Methods:

• Developed the Multi-Echo DIstortion Correction (MEDIC) pipeline using ME fMRI data.
• MEDIC utilizes differences between echoes to estimate and correct magnetic field inhomogeneities on a frame-by-frame basis.
• Compared MEDIC's performance against the established FSL TOPUP method for distortion correction.

Main Results:

• MEDIC effectively corrects geometric distortions dynamically, frame-by-frame, mitigating head motion artifacts.
• Frame-wise correction with MEDIC significantly reduces the impact of head motion on resting-state functional connectivity (RSFC) maps.
• MEDIC demonstrates improved alignment to anatomical references compared to the standard FSL TOPUP approach.
• The method eliminates the need for separate field map acquisition, simplifying the fMRI workflow.

Conclusions:

• MEDIC provides a robust, motion-robust method for correcting fMRI geometric distortions using ME data.
• This approach enhances the reliability of RSFC analyses and improves anatomical registration in fMRI studies.
• MEDIC offers a significant advancement over traditional single-echo fMRI correction techniques, particularly in the presence of head motion.