Epilepsy is a pathological condition characterized by the spontaneous and unforeseeable occurrence of seizures, during which the perception or behaviour of patients is disturbed. An automatic early detection of the seizure onsets would help the patients and observers to take appropriate precautions. Various methods have been proposed to predict the onset of seizures based on electroencephalography (EEG) recordings. The use of non-linear features motivated by the higher-order spectra (HOS) has been reported to be a promising approach to differentiate between normal, background (pre-ictal), and epileptic EEG signals. In this work, these features are used to train both a Gaussian mixture model classifier and a support vector machine classifier. Results show that the classifiers were able to achieve 93.11 per cent and 92.67 per cent classification accuracy respectively, with selected HOS-based features. About 2 h of EEG recordings from ten patients were used in this study.
CockerellO. C.JohnsonA. L.SanderJ. W.HartY. M.GoodridgeD. M.ShorvonS. D.‘Mortality from epilepsy: Results from a prospective population-based study.’Lancet344 (1994): 918–921
3.
CallawayE.HarrisP. R.‘Coupling between cortical potentials from different areas.’Science183 (1974): 873–875
4.
BabloyantzA.NicolisC.SalazarJ. M.‘Evidence of chaotic dynamics of brain activity during the sleep cycle.’Phys. Lett. A111 (1985): 152–157
5.
MormannF.KreuzT.AndrzejakR. G.PeterD.LehnertzK.ElgerC. E.‘Epileptic seizures are preceded by a decrease in synchronization.’Epilepsy Res.53 (2003): 173–185
6.
MormannF.ThomasK.ChristophR.AndrzejakR.KraskovA.DavidP.ElgerC. E.LehnertzK.‘On the predictability of epileptic seizures.’Clini. Neurophysiology116 (2005): 569–587
7.
OsorioI.FreiM. G.WilkinsonS. B.‘Real-time automated detection and quantitative analysis of seizures and short-term prediction of clinical onset.’Epilepsia39 (6) (1998): 615–627
8.
OsorioI.FreiM. G.ManlyB. F.SunderamS.BhavarajuN. C.WilkinsonS. B.‘An introduction to contingent (closed-loop) brain electrical stimulation for seizure blockage, to ultra-short-term clinical trials, and to multidimensional statistical analysis of therapeutic efficacy.’J. Clin. Neurophysiology18 (2001): 533–544
9.
OsorioI.FreiM. G.GiftakisJ.PetersT.IngramJ.TurnbullM.HerzogM.RiseS.SchaffnerM. T.WennbergR. A.WalczakT. S.RisingerM. W.Ajmone-MarsanC.‘Performance reassessment of a real-time seizure-detection algorithm on long ECoG series.’Epilepsia43 (12) (2002): 1522–1535
10.
NiederhoeferC.GollasF.ChernihovskyiA.LehnertzK.TetzlaffR.‘Detection of seizure precursors in the EEG with cellular neural networks.’Epilepsia45 (Suppl. 7) (2004): 245
11.
TetzlaffR.NiederhöferC.FischerP.‘Feature extraction in epilepsy using a cellular neural network based device — first results.’ In Proceedings of the 2003 International Symposium on Circuits and systems (ISCAS ′03)
12.
KaplanA. Y.‘Segmental structure of EEG more likely reveals the dynamic multistability of the brain tissue than the continual plasticity one.’ In Proceedings of the Sixth International Conference on Neural information processing (ICONIP ′99), Perth, Australia1999, vol. 2, pp. 633–638 (IEEE, New York).
13.
KemalM. K.GulerI.AlperD.MehmetA.‘Comparison of STFT and wavelet transform methods in determining epileptic seizure activity in EEG signals for real-time application.’Computers Biology Medicine35 (2005): 603–616
14.
StamC. J.PijnJ. P.SuffczynskiP.Lopez da SilvaF. H.‘Dynamics of the human alpha rhythm: Evidence for nonline.’Clin. Neurophysiology110 (10) (1999): 1801–1813
15.
ZhuoS. M.GanJ. Q.SepulvedaF.‘Classifying mental tasks based on features of higher-order statistics from EEG signals in brain—computer interface.’Inf. Sci.178 (6) (2008): 1629–1640
16.
ShenM.ChanF. H. Y.SunL.BeadleB. J.‘Parametric bispectral estimation of EEG signals in different functional states of the brain.’IEE Proc. — Sci., Measmt Technol.147 (6) (2000): 374–377
17.
ChuaK. C.ChandranV.AcharyaU. R.LimC. M.‘Analysis of epileptic EEG signals using higher order spectra.’J. Med. Engng Technol.33 (1) (2009): 42–50
AndrzejakR. G.LehnertzK.RiekeC.ElgerC. E.‘Indications of nonlinear deterministic and finite dimensional structures in time series of brain electrical activity: Dependence on recording region and brain state.’Phys. Rev. E64 (2001): 061 907.
20.
ChandranV.ElgarS.‘Pattern recognition using invariants defined from higher order spectra — one-dimensional inputs.’IEEE Trans. Signal Processing41 (1) (1993): 205–212
21.
NgT. T.ChangS. F.SunQ.‘Blind detection of photomontage using higher order statistics.’ In Proceedings of the 2004 International Symposium on Circuits and systems (ISCAS ′04), Vancouver, Canada, 23–26 May 2004 vol. 5, pp. V-688–V-691 (IEEENew York).
22.
InouyeT.ShinosakiK.SakamotoH.ToiS.UkaiS.IyamaA.KatsudaY.HiranoM.‘Quantification of EEG irregularity by use of the entropy of the power spectrum.’Electroencephalography Clin. Neurophysiology79 (3) (1991): 204–210
23.
SwamiA.MendelC. M.NikiasC. L.Higher-order spectral analysis (HOSA) toolbox, version 2.0.3 (2000): (United Signals & Systems, Inc., Culver City, California, USA)
24.
VapnikV.Statistical learning theory (John Wiley1998)
25.
MullerK. R.MikaS.RatschG.TsudaK.ScholkopfB.‘An introduction to kernel based learning algorithms.’IEEE Trans. Neural Networks12 (2) (2001): 181–201
26.
27.
DeLeoJ. M.‘Receiver operating characteristic laboratory (ROCLAB): Software for developing decision strategies that account for uncertainty.’ In Proceedings of the Second International Symposium on Uncertainty modeling and analysis
28.
DowneyT. J.MeyerD. J.PriceR. K.SpitznagelE. L.‘Using the receiver operating characteristic to assess the performance of neural classifiers.’ In Proceedings of the International Joint Conference on Neural networks (IJCNN ′99)
29.
RavelliF.AntoliniR.‘Complex dynamics underlying the human electroencephalogram.’Biol. Cybernetics67 (1992): 57–65
30.
LittB.EstellerR.EchauzJ.D'AlessandroM.ShorR.HenryT.PennellP.EpsteinC.BakayR.DichterM.VachtsevanosG.‘Epileptic seizures may begin hours in advance of clinical onset: A report of five patients.’Neuron30 (2001): 51–64
31.
NicolelisM. A. L.‘Actions from thoughts.’Nature409 (2001): 403–407
32.
ElgerC. E.‘Future trends in epileptology.’Curr. Opin. Neurology14 (2001): 185–186
33.
UthmanB. M.ReichlA. M.DeanJ. C.EisenschenkS.GilmoreR.ReidS.RoperS. N.WilderB. J.‘Effectiveness of vagus nerve stimulation in epilepsy patients: A 12-year observation.’Neurology63 (6) (2004): 1124–1126
IasemidisL. D.ShiauD.-S.SackellaresJ. C.PardalosP. M.PrasadA.‘Dynamical resetting of the human brain at epileptic seizures: Application of nonlinear dynamics and global optimization techniques.’IEEE Trans. Biomed. Engng51 (3) (2004): 493–506
36.
IasemidisL. D.ShiauD. S.SackellaresJ. C.PardalosP.‘Transition to epileptic seizures: Optimization.’ In Discrete mathematical problems with medical applications, DIMACS Series in Discrete Mathematics and Theoretical Computer Science, vol. 55 (Eds DuD. Z.PardalosP. M.WangJ.), 1999, pp. 55–74 (American Mathematical SocietyProvidence, Rhode Island).
37.
IasemidisL. D.PardalosP.SackellaresJ. C.ShiauD.‘Quadratic binary programming and dynamical system approach to determine the predictability of epileptic seizures.’J. Combinatorial Optimization5 (2001): 9–26
38.
YaariY.BeckH.‘‘Epileptic neurons’ in temporal lobe epilepsy.’Brain Pathology12 (2002): 234–239
39.
PetitmenginC.BaulacM.NavarroV.‘Seizure anticipation: Are neurophenomenological approaches able to detect pre-ictal symptoms?’Epilepsy Behavior9 (2) (2006): 298–306
40.
SleighJ. W.OlofsenE.DahanA.de GoedeJ.Steyn-RossA.‘Entropies of the EEG: The effects of general anaesthesia.’ In Proceedings of the Fifth International Conference on Memory, Awareness and Consciousness, New York, USA, 2001 available from http://waikato.researchgateway.ac.nz/handle/10289/770 (accessed 23 March 2009).
41.
AcharyaU. R.FaustO.KannathalN.ChuaT.-L.LaxminarayanS.‘Nonlinear analysis of EEG signals at different sleep stages.’Computer Meth. Programs Biomedicine, Sweden80 (1) (2005): 37–45
