The identification of electrographic seizures during long-term EEG monitoring in the neonate is currently based upon visual interpretation of the graphic record, a process that is very time-consuming. While significant progress has been made in the automated detection of seizures in the adult population, relatively little work has been done in the neonatal area. Therefore, the major objective of this project is the development of techniques for the reliable automated detection of electrographic seizures in the neonatal EEG. We propose a multi-stage, hybrid approach to detection that will employ a combination of signal processing, pattern recognition, neural networks, and expert rules. Through the successive stages of the detection process, multichannel neonatal EEG data containing all types of background activity and artifacts will be analyzed to detect and classify electrographic seizures. We postulate that the varied types and morphologies of seizures in the neonatal EEG, as compared to seizures in the adult EEG, can best be detected and classified using this hybrid approach. We will test the methods we develop on data recorded from infants in the Clinical Research Center for Neonatal Seizures, The Methodist Hospital, in Houston, Texas. We expect that the information we gain from the research will lead to the development of a practical seizure detection system and further our long-term goals of reduced expense in the reading and interpretation of neonatal EEGs, and also facilitate the efficient collection of seizure parameters for possible use in future research studies.
| Mitra, Joyeeta; Glover, John R; Ktonas, Periklis Y et al. (2009) A multistage system for the automated detection of epileptic seizures in neonatal electroencephalography. J Clin Neurophysiol 26:218-26 |
| Karayiannis, Nicolaos B; Mukherjee, Amit; Glover, John R et al. (2006) Detection of pseudosinusoidal epileptic seizure segments in the neonatal EEG by cascading a rule-based algorithm with a neural network. IEEE Trans Biomed Eng 53:633-41 |
