Quantitative EEG Based on Direct Recording and Source Localization Estimation in Repetitive Hand Motor Activity

Abstract

Electroencephalography (EEG) has been widely used in clinical practice as a tool to guide an efficient post-stroke functional recovery program and adjust the rehabilitation training program. EEG has some advantages in temporal resolution, portability, and relatively low cost. Motor-related brain activities can be quantified by measuring the mu rhythm’s event-related desynchronization (ERD), acquired from a direct surface EEG recording. The common problem of direct surface EEG is the crosstalk between brain regions that are unrelated to the sensorimotor cortex. This problem can be solved through the EEG source localization technique. Essentially, the idea of source reconstruction is to estimate the primary cortical current densities during the given EEG recording across the head surface. The recent work aims to compare EEG power spectral density (PSD) analysis and source localization. Modification of exact-Low Resolution Brain Electromagnetic Tomography (eLORETA) was used as an EEG source localization method. EEG data recording was performed on a healthy, able-bodied male subject using 21 electrodes based on the case study’s 10-20 electrode placement system. The subject was asked to perform basic hand movements, i.e., grasping with and without hand load. Mu power was calculated from the electrodes that are located around the motor cortex (C3 and C4), while the average current density was calculated from the Brodmann Area (BA) 4 and 6, both during grasping and relaxation (baseline). The comparison showed that during grasping, there was a decrease in aver-age current density around BA-4 and BA-6 and a decrease in mu power at C3 and C4 when compared to baseline. These results indicate: (1) the activity of mu rhythm’s ERD around the sensorimotor cortex during hand movement; and (2) the possibility of using the EEG source localization method as an alternative biofeedback modality.