Robust source analysis of oscillatory motor cortex activity with inherently variable phase delay

Abstract

This study evaluated quantitatively the synchronization between the magnetoencephalography (MEG) and electromyography (EMG) signals and developed a novel method for the determination of the synchronization in order to increase the reliability of the source analysis of the oscillatory motor cortex activity. The new method is based on our observation that there are large variances in the time lag due to relatively low muscle-cortex synchronization which reduces the signal-to-noise ratio of the MEG signal when averaged in direct synchrony with the rectified EMG peaks. To improve the localization of the motor cortex activity, time-frequency analysis was performed for each epoch coinciding with an EMG peak to reject the weak oscillatory activity and artifacts. In addition, the MEG signals were shifted to maximize the coherence between MEG and rectified EMG by determining for each accepted epoch the time lag resulting in a maximum cross-correlation. An experiment was carried out using 30 subjects in order to determine the applicability of this method to a real situation. The synchronization and the results of the corresponding source analysis based on the novel method were compared with the data obtained using the non-phase-shift method and Hilbert approach detecting EMG phase. The results showed that the synchronization was significantly enhanced and the signal-to-noise ratio of the MEG signals improved, and that the localized dipoles of all subjects were well clustered at the motor cortex. This method, based on shifting the MEG epochs according to the simultaneously measured time lag, considerably improves the performance of the averaging and localization of the rhythmic activity of the motor cortex.