Brain network biomarker characterization by localized EEG-based neuroimaging signal data analysis

Fow-Sen Choa, Department of Computer Science and Electrical Engineering
Chintan Patel, Department of Computer Science and Electrical Engineering
Anupam Joshi, Department of Computer Science and Electrical Engineering
Deepa Gupta, Department of Computer Science and Electrical Engineering

In this work, we plan to develop methods that can extract functional brain network biomarkers for identifying human subjects that suffer from neurological disorders. Conventional EEG based brain connectivity studies are mostly describing connections among scalp electrode locations, which challenge the functional meaning interpretation of brain activities. In this work, we use real time sLORETA source localization to link measurement with brain functional networks. We then use energy landscape-based analysis to obtain brain subnetwork energy states and their statistical distributions. Through t-test of each brain state we can obtain connectivity biomarkers and can separate control and patient groups. The EEG data was recorded at UMB from 22 schizophrenia patients and 27 healthy controls in response to transcranial magnetic stimulation administered on the left motor cortex, vertex, vermis, and prefrontal cortex.