Leadfield Correction within a Two-Stage Champagne Framework for Head Motion Artifact Suppression in MEG

Head motion is a major interference source in magnetoencephalography (MEG). In MEG systems employing rigid, fixed sensor arrays, the immobility of the sensors combined with unconstrained head movement leads to mismatches between the leadfield and the actual head position. This spatial discrepancy significantly degrades both waveform fidelity and source localization accuracy. To address this issue, we propose a motion-informed leadfield correction approach: leadfield correction within the Two-Stage Champagne framework (LC-TSC). By dynamically correction the leadfield with the actual head position and mapping the estimated sources to a common reference head position, the method effectively compensates for distortions caused by motion-induced leadfield mismatch, thereby enabling direct suppression of motion artifacts at the source level. Simulation results demonstrate that LC-TSC substantially outperforms benchmark algorithms in both localization precision and waveform quality, offering a robust and feasible solution for motion-resilient analysis in fixed-array MEG systems.