Deterministic Finite Automata in the Detection of Epileptogenesis in a Noisy Domain

In the continuing endeavor to create a robust methodology to mine epileptiform activity from Electroencephalograms (EEG) by combining the methodologies of Deterministic Finite Automata (DFA) and Knowledge Discovery in Data Mining (KDD), this paper presents a more efficient model for identifying epileptiform activity and seizures from Electroencephalograms (EEG) than previous experiments by the authors. We present a a DFA that separates the states between seizure and non-seizure using robust testing and additional algorithms to increase accuracy in the presence of noisy signals. Our results are capable of identifying seizures even when significant artifact and noise is present.