Optimization and Data Mining Approaches to Seizure Prediction in Epilepsy Research

W. Art Chaovalitwongse

Department of Industrial and Systems Engineering, Rutgers University

Uncontrolled epilepsy poses a significant burden to society due to associated healthcare cost to treat and control the unpredictable and spontaneous occurrence of seizures. The main objective of this talk is to introduce novel optimization-based data mining approaches to the study of brain physiology, which might be able to revolutionize current diagnosis and treatment of epilepsy. Through quantitative analyses of electroencephalogram

(EEG) recordings, a new data mining paradigm for feature selection and clustering is developed based on the chaos theory and optimization techniques. The experimental results to be presented in this talk will demonstrate that the proposed techniques can be used as a feature

(electrode) selection technique to capture seizure pre-cursors and predict seizures. In addition, not only do the proposed techniques excavate hidden patterns/relationships in EEGs, but also give a greater understanding of brain functions (as well as other complex systems) from a system perspective.