A method for automatically detection of absence epilepsy discharges in EEG signals

 

Yu. V. Obukhov ¹, I. A. Kershner ¹, V. V. Gnezditskiy ²

 ¹ Kotel'nikov Institute of Radioengineering and Electronics of RAS

² Scientific Neurology Centre

 

The paper is received on November 18, 2016

 

Abstract. The method for automatically detection of absence epilepsy is described. The method is based on EEG signals wavelet spectra ridges analysis. To remove signal trend, the discrete eight-order Butterworth filter with 2-124 Hz bandwidth was used for the processing of data. The detection method of Epileptic discharges is based on the analysis of Morlet wavelet spectrogram ridges. Actually, the wavelet spectrogram consists of the background as well as the ridges. Epileptic discharges have high PSD in comparison to the background. Ridges with maximal PSD are considered. To filter the background from the spectrogram it was proposed to analyze the ridges PSD histograms. The histogram shows steep decrease in a particular PSD values, and these values are selected as adaptive thresholds for the detection of epilepsy seizures. The algorithm of automatic epileptic seizures detection can be realized as EEG time windows analysis by proposed approach. Designed method is illustrated by absence epilepsy seizures detection.

Key words: EEG, ridges of wavelet spectrogram, absence epilepsy, seizures detection.

References

1.    Word Yealh Organization [electronic resource].  URL: http://www.who.int/topics/epilepsy/en/

2.    Hopfengärtner R., Kasper B. S., Graf W., et al. Automatic seizure detection in long-term scalp EEG using an adaptive thresholding technique: a validation study for clinical routine. Clinical Neurophysiology. 2014;125(7):1346–1352. doi: 10.1016/j.clinph.2013.12.104.

3.    Paul Fergus, et. al. Automatic Epileptic Seizure Detection Using Scalp EEG and Advanced Artificial Intelligence Techniques. //Biomed Res Int. 2015; 2015: 986736. Published online 2015 Jan 29. doi:  10.1155/2015/986736 PMCID: PMC4325968.

4.    S. Divya, S. Suja Priyadharsini. Classification of EEG Signal for Epileptic Seizure Detectionusing EMD and ELM. // INTERNATIONAL JOURNAL FOR TRENDS IN ENGINEERING & TECHNOLOGY VOLUME 3 ISSUE 2 – FEBRUARY 2015 – ISSN: 2349 – 9303 68

5.    Bosnyakova D., Obukhov Yu. Extraction of dominant features in biomedical signals. // Pattern Recogn. Image Anal. 2005. 15. Š. 513-515.

6.    Gabova A.V., Bosnyakova D.Yu., Bosnyakov M.S. et. al. The Time–Frequency Structure of the Spike–Wave Discharges in Genetic Absence Epilepsy. // Doklady Biological Sciences, Vol. 396, 2004, pp. 194–197. doi:10.1023/B:DOBS.0000033274.44589.df

7.    Bosnyakova D., Gabova A., Kuznetsova G.  et al.  Time-frequency analysis of spike-wave discharges using the modified wavelet transform. // J. Neurosci. Methods. 2006. 165. Š. 80-85.

8.    Bosnyakova  D., Gabova A.,  Zharikova A. et al. Some peculiarities of time-frequency dynamics of spike-wave discharges in human and rats. // Clin. Neurophysiol.  2007. 118. Š. 1736-1743.

9.    D. Y Ko. Epileptiform Discharges. Medscape Apr 2. 2014.  URL: http://emedicine.medscape.com/article/1138880-overview#showall .

10.Antciperov V.E., Analysis of brain epileptic discharges hyper synchronization based on special conical representation of EEG signals. // Journal of Radio Electronics, 2014, N. 11.  Available at http://jre.cplire.ru/koi/nov14/18/text.pdf. (In Russian)

11.Kim L. Powell et al. Seizure expression, behavior, and brain morphology differences in colonies of Genetic Absence Epilepsy Rats from Strasbourg // Epilepsia. – 2014. - Vol. 55. – 12. – P. 1959-1968.

12.Mallat Stéphane. A Wavelet Tour of Signal Processing  (Third Edition). The Sparse Way. ISBN: 978-0-12-374370-1, http://www.sciencedirect.com/