Application of Advanced Signal and Information Processing to EEG Data

The Electroencephalogram (EEG) uses a multi-electrode array to measure electrical potentials related to brain activity. It is a low cost non-intrusive technique with diagnostic application to many situations including: battle fatigue, stress, head injury, epilepsy, and sleep disorders. The primary means of EEG analysis today is the manual reading of long paper strip charts of multiple electrode output time series. This is time consuming, costly, and subjective. EEG's are noisy. Signals associated with muscular activity can mask the signals of interest. This SBIR proposes to apply advanced signal and information technology, developed for DoD applications, to problems of EEG analysis. Advanced signal processing and display techniques will be used to enahnce signals and suppress noise to reveal new features. Objective metrics will be developed which can be extracted from the data by machine. This will provide the first step toward automatic screening of large volumes of EEG data. Advanced array processing techniques will be used in conjunction with improved electrical models of the skull to provide localization capability, important in diagnosis of head injury, stroke, epilepsy and brain tumors. The project involves collaboration between ORINCON and Dr. William Peterson, an EEG expert at D.C. General Hospital. ORINCON has developed and successfully applied advanced signal and information processing techniques to a variety of DoD problems under sponsorship of ARPA, the Navy and other agencies. Anticipated Benefits: Application of advanced signal and information processing to EEG data can improve diagnostic accuracy, reduce subjectivity, and reduce the cost of analysis of EEG data. Our goal is to develop an inexpensive PC based system to automatically screen data and assist in diagnosis. There is a large market for this system consisting of the hundreds of military and civilian facilities and hospitals where EEG's are conducted.