Dynamical Characterization of Mild Traumatic Brain Injury in Default Mode Network Using Magnetoencephalogram

Dynamical Characterization of Mild Traumatic Brain Injury in Default Mode Network using Magnetoencephalogram

Sreenivasan R. Nadar, Dominic Nathan, John Ollinger, John D. Hughes, Grant H. Bonavia, and Gerard Reidy

The synchronization of oscillatory neuronal activity allows for the formation of diverse spatiotemporal characteristics including metastability patterns, which are suggested to form the underlying substrate of information processing within the central nervous system (CNS). We sought to characterize the distributed oscillatory brain networks of the task-free Default Mode Network (DMN) from MEG data, specifically to compare controls and subjects with mild Traumatic Brain Injury (mTBI). A total of 15 controls (Age 37.07±2.09, F=1) and 15 mTBI subjects (Age 41.8±1.53, F=2) were recruited for this study. Magnetoencephalogram (MEG) data was acquired using an Elekta Neuromag 306 system at 1kHz and the raw data were co-registered with structural T1 images, and visually inspected for bad channels before running Independent Component Analysis based artifact correction. The sensor signals were source localized using the Linearly Constrained Minimum Variance beamformer. The structural and MEG data were transformed to MNI space for group analysis.  Source signals were extracted for the Medial Prefrontal Cortex (MPFC), Anterior Cingulate Cortex (ACC), Posterior Cingulate Cortex (PCC), Angular Gyrus (AG) and Cerebellar Tonsils, which are main areas of the DMN. We computed Granger Causality, synchronization likelihood, and multiscale entropy to compare the dynamics of oscillatory synchronized network within the DMN of controls and mTBI subjects for different frequency bands: alpha (8:14Hz), beta (15-30Hz), gamma (31-50Hz), delta (1-4Hz) and theta (4-8Hz). The preliminary results indicate that control subjects had higher synchronization of MPFC with respect to the left AG in the alpha band and the PCC in the theta band; whereas for the mTBI patients the Left AG had higher causal influence to MPFC in the theta band, ACC in the alpha and Right AG in beta band.  The results suggest that the MPFC and the Left AG have distributed modulatory mechanisms in different frequency bands and the nature of these connection may form the basis for characterizing mTBI as this specific connection is critically involved in language, number processing, spatial cognition, memory retrieval and attention. Of future interest is to determine if functional connectivity and metastability measures can be used at an individual scale, as opposed to a group level, to distinguish between subject populations.