Abstract
Abstract 1 S. Faul, 2 G. Boylan, 1 L. Marnane, 1 G. Lightbody, and 3 S. Connolly ( 1 Department of Electrical and Electronic Engineering, University College Cork, Cork, Ireland , 2 Department of Paediatrics and Child Health, University College Cork, Ireland , 3 Department of Clinical Neurophysiology, St. Vincent's Hospital, Dublin, Ireland ) Purpose: Seizures are a common neurological emergency in the neonatal intensive care unit. Clinical signs of neonatal seizures can be very subtle or entirely subclinical and hence the electroencephalogram (EEG) is the only reliable tool for their detection. However, constant supervision by trained specialists is still needed for adequate seizure detection. An automated seizure detection system would remove the need for constant EEG supervision. The aim of this study was to use a digital signal processing theory to develop a multistage seizure detection system. Method: The proposed system breaks the seizure detection problem up into three sections: preprocessing, feature extraction and classification. The preprocessing element, based on independent component analysis, incorporates multichannel analysis and reduces the effect of artifacts. Defining features are then extracted from the resulting signals using techniques from a wide range of signal processing techniques, from frequency analysis to modelling. Finally, a classification network makes the final decision based on changes in the characteristic feature set. Results: Preliminary tests carried out on neonatal EEG data, extracting 10 features, detected 45 out of 46 seizures (97.82%) with a mean delay of only 11.4 seconds between commencement of the seizure and its detection. Conclusion: The neonatal seizure detection system proposed in this study produces rates and speed of detection surpassing previously documented methods. An implementation of the system will undergo clinical tests after further simulation and development. This system will greatly aid clinical neurophysiologists in detecting seizures rapidly, thus minimising the impact on the infant.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
