Validity of subthalamic-cortical coherency observed in patients with Parkinson's disease

Simultaneous magnetoencephalography (MEG) and local field potential (LFP) recordings in patients with Parkinson's disease (PD) undergoing deep brain stimulation (DBS) treatment is a promising tool for both clinical application and basic research. Recordings can be accomplished during the time interval between electrode insertion and its connection to the pulse generator while electrodes are being externalized. In nine PD patients, coherence (COH) between LFP and MEG signals was calculated from the data of a 5-min simultaneous MEG-LFP rest recording. For the observed COH patterns, a validation procedure is introduced based on time-shift principal component analysis (TSPCA), which was originally developed to suppress background signals from MEG. Here TSPCA is used as a regression of the MEG signal with filtered versions of the LFP signal to intentionally remove COH. The channel mean of the original COH is compared with the residual channel mean COH after TSPCA application. COH peaks are suppressed in the 15- to 30-Hz range; at lower frequencies, the results are less obvious due to the presence of an artifact caused by a weak remanent magnetization of the externalization wires. The COH suppression is statistically significant for four out of nine subjects, and validation has been achieved, as the COH suppression yields the hypothesized outcome.