Abstracts

Rationale: Motor mapping is a commonly ordered magnetoencephalography (MEG) procedure during the presurgical evaluation of epilepsy patients. It is optimal to localize the neuromagnetic motor field (MF) prior to movement onset, however, to obtain this transient response typically requires long inter-movement times resulting in a demanding task.  For young children and neurologically-compromised populations, compliance can be a significant problem that increases the likelihood of an incomplete test or poor data quality which can then compromise the validity of the source localization estimate and thus the utility of the procedure. The purpose of this study is to enhance current clinical MEG motor mapping protocols by validating the localizing ability of potentially easier rhythmic tasks for motor mapping. Methods: 12 healthy adults aged 19-30 performed a standard self-paced button-press (SPM) paradigm with an inter-movement interval of 3-5 seconds. Additionally, two rhythmic pinching tasks were performed, one self-paced (RPinch) at approximately 1-2Hz and one auditorily paced (APinch) with at a pseudo-rhythmic pace of 1.75 Hz. Neuromagnetic activity was recorded simultaneously using a 306 channel Triux MEG machine (MEGIN Oy, Ltd). A custom-built pressure sensor was used to record the rhythmic pinch movement of the index and thumb as a separate ADC channel. After routine data preprocessing to remove eye-blink artifacts, the data were epoched and averaged with respect to button press onset or maximal contraction for button-press and rhythmic tasks, respectively. A single Equivalent Current Dipole (ECD) source model was employed using a spherical head-model and the best ECD fit surrounding the peak of the highest amplitude response based on a combination of goodness of fit and volumetric confidence. Source locations were recorded in subject head coordinate space prior to overlay onto volumetric MRI imaging. Euclidean distances were calculated for the location of the rhythmic tasks with respect to that of the standard button-press, and for all tasks with respect to the location within the MRI representing the site of transcranial magnetic stimulation (TMS) motor threshold mapping. Results: The z-score significant localization separation for the APinch task to SPM was 7 mm (p< 0.01) compared to that of 11 mm for RPinch (p<0.02).  The average Euclidean distance to the TMS cortical site was 9.5mm ±3.6 , 9.0mm ±4.1, and 7.9mm±4.5 for APinch, RPinch, and SPM, respectively. Conclusions: The results from this study show that rhythmic tasks generally localize within approximately 10 mm of a TMS verified cortical site and have lower across-task localization variability. The benefit of considering rhythmic tasks within clinical MEG motor mapping battery will be discussed in the context of the localization goals of presurgical mapping and also with respect to TMS as a localization reference. Funding: No funding