Comparison of fractal dimension estimation algorithms for epileptic seizure onset detection

Polychronaki, GE; Ktonas, P; Gatzonis, S; Asvestas, PA; Spanou, E; Siatouni, A; Tsekou, H; Sakas, D; Nikita, KS

dc.contributor.author	Polychronaki, GE	en
dc.contributor.author	Ktonas, P	en
dc.contributor.author	Gatzonis, S	en
dc.contributor.author	Asvestas, PA	en
dc.contributor.author	Spanou, E	en
dc.contributor.author	Siatouni, A	en
dc.contributor.author	Tsekou, H	en
dc.contributor.author	Sakas, D	en
dc.contributor.author	Nikita, KS	en
dc.date.accessioned	2014-03-01T02:45:11Z
dc.date.available	2014-03-01T02:45:11Z
dc.date.issued	2008	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/32192
dc.subject	Epileptic Seizure	en
dc.subject	Estimation Algorithm	en
dc.subject	Fractal Dimension	en
dc.subject	K Nearest Neighbor	en
dc.subject	Onset Detection	en
dc.subject	Time Series	en
dc.subject	False Positive Rate	en
dc.subject.other	EEG analysis	en
dc.subject.other	EEG recording	en
dc.subject.other	Epileptic seizures	en
dc.subject.other	False positive rates	en
dc.subject.other	Fractal dimension estimation	en
dc.subject.other	K-nearest neighbors	en
dc.subject.other	k-NN algorithm	en
dc.subject.other	Natural measure	en
dc.subject.other	Bioinformatics	en
dc.subject.other	Electroencephalography	en
dc.subject.other	Finite difference method	en
dc.subject.other	Fractal dimension	en
dc.subject.other	Online searching	en
dc.subject.other	Partial discharges	en
dc.subject.other	Time series analysis	en
dc.subject.other	Algorithms	en
dc.title	Comparison of fractal dimension estimation algorithms for epileptic seizure onset detection	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/BIBE.2008.4696822	en
heal.identifier.secondary	http://dx.doi.org/10.1109/BIBE.2008.4696822	en
heal.identifier.secondary	4696822	en
heal.publicationDate	2008	en
heal.abstract	The fractal dimension (FD) is a natural measure of the irregularity of a curve. In this study the performances of two FD-based methodologies are compared in terms of their ability to detect the onset of epileptic seizures in scalp EEC The FD algorithms used is Katz's, which has been broadly utilized in the EEG analysis literature, and the k-nearest neighbor (k-NN), which is applied in this study in a time series sense for the first time. 244.9 hours of EEG recordings, including 16 seizures in 3 patients, were analyzed. Both approaches achieved 100% sensitivity with a false positive rate of 0.85 FP/h for the k- NN algorithm and 1 FP/h for Katz's algorithm. The corresponding detection delays were 6.5 s and 10.5 s on the average, respectively. The k-NN algorithm seems to outperform Katz's algorithm. Results are satisfactory in comparison to other methodologies applied on scalp EEG and proposed in the literature.	en
heal.journalName	8th IEEE International Conference on BioInformatics and BioEngineering, BIBE 2008	en
dc.identifier.doi	10.1109/BIBE.2008.4696822	en