Electroencephalogram Signals Processing for the Diagnosis of Petit mal and Grand mal Epilepsies Using an Artificial Neural Network

 

M. R. Arab^*1, A. A. Suratgar², V. M. Martínez–Hernández², A. Rezaei Ashtiani³

 

¹ Department of Biomedical Engineering, Arak Medical University, Arak, Iran. *a–suratgar@aut.ac.ir

² Department of Electrical Engineering, Arak University, Arak, Iran.

² Department of Electrical Engineering, Amirkabir University of Technology (Tehran Polytechnic), Tehran, Iran.

³ Department of Neurology, Arak Medical University, Arak, Iran.

 

ABSTRACT

In this study, a novel wavelet transform–neural network method is presented. The presented method is used for the classification of grand mal (clonic stage) and petit mal (absence) epilepsies into healthy, ictal and interictal (EEGs). Preprocessing is included to remove an artifact occurred by blinking and a wandering baseline (electrodes movement) as well as an eyeball movement artifact using the Discrete Wavelet Transformation (DWT). Denoising EEG signals from the AC power supply frequency with a suitable notch filter is another job of preprocessing. The preprocessing enhanced speed and accuracy of the processing stage (wavelet transform and neural network). The EEGs signals are categorized into normal and petit mal and clonic epilepsy by an expert neurologist. The categorization is confirmed by the Fast Fourier Transform (FFT) analysis. The dataset includes waves such as sharp, spike and spike–slow wave. Through the Countinous Wavelet Transform (CWT) of EEG records, transient features are accurately captured and separated and used as classifier input. We introduce a two–stage classifier based on the Learning Vector Quantization (LVQ) neural network localized in both time and frequency contexts. The particular coefficients of the Continuous Wavelet Transform (CWT) are networks. The simulation results are very promising and the accuracy of the proposed method obtained is of about 80%.

Keywords: Tonic–clonic epilepsy, petit mal epilepsy, Continuous Wavelet Transform (CWT), absence epilepsy.

 

RESUMEN

En este estudio, se presenta un nuevo método basado en redes neuronales y la transformada ondicular. El método presentado se usa para la clasificación de la epilepsia gran mal (clónica) y pequeño mal (de ausencia) en saludable, ictal e interictal (EEG). Se incluye el pre procesamiento para eliminar un artefacto causado por el parpadeo y una línea de base errante (movimiento de electrodos) así como un artefacto producido por el movimiento ocular usando la Transformada Ondicular Discreta (DWT). Otra función del pre procesamiento es la eliminación de ruido de las señales de EEG de la frecuencia de la fuente de alimentación AC con un filtro de eliminación adecuado. El pre procesamiento aumentó la velocidad y precisión de la etapa de procesamiento (transformada ondicular y red neuronal). Un neurólogo experto clasifica las señales de EEG en epilepsia normal, pequeño mal y clónica. La clasificación se corrobora por medio del análisis con Transformada Rápida de Fourier (FFT). El conjunto de datos incluye ondas tales como agudas, puntas y punta–onda lenta. Mediante la Transformada Ondicular Continua (CWT) de los registros del EEG, se capturan y separan correctamente características transitorias y se usan como entrada del clasificador. Introducimos un clasificador de dos etapas basado en redes de cuantización vectorial (LVQ) localizado en los contextos tiempo y frecuencia. Los coeficientes particulares de la Transformada Ondicular Continua (CWT) son redes. Los resultados de la simulación son muy prometedores y la exactitud del método propuesto es de alrededor del 80%.

 

DESCARGAR ARTÍCULO EN FORMATO PDF

 

References

[1] W. G. Bradley, W. H. Trescher and R. P. Lesser, Neurology in Clinical Practice. 3rd Ed., 2004, pp: 710–762.         [ Links ]

[2] F. H. Duffy, V. G. Iyer and W. W. Surwill, Clinical Electroencephalography and Topographic Brain Mapping. Springer Verlag, New York, 1989, pp: 210–270.         [ Links ]

[3] H. Adeli, N. Dadmeher and S. Ghosh–Dastidar, 2007. Mixed–band wavelet–chaos–neural network methodology for epilepsy and epileptic seizure detection. IEEE Trans. Biomed. Eng., Vol:54, 2007, pp: 545–1551.         [ Links ]

[4] A. Subasi, E. Ercelebi, Classification of EEG signals using neural network and logistic regression. Comput. Methods Program. Biomed., Vol:78, 2005, pp: 87–99.         [ Links ]

[5] T. Zikov, S. Bibian, G. A. Dumont and M. Huzmezan, A Wavelet based de–noising, technique for occular artifact correction of electroencephalogram, 2007, pp:101–106.         [ Links ]

[6] M. Latka, Z. Was, A. Kozik and B. J. West, Wavelet analysis of epileptic spikes. Phys. Rev. E Stat. Nonlin Soft Matter Phys., Vol: 67, 2003, pp: 495–500.         [ Links ]

[7] A. Petrosian, D. Prokhorov, R. Homan, R. Dascheiff and D. Wunsch, Recurrent neural network based prediction of epileptic seizures in intercranial and extracranial EEG. Neurocomputing, Vol:30, 2000, pp:201–218.         [ Links ]

[8] W. Weng, K. Khorasani, An adaptive structure neural network with application to EEG automatic seizure detection. Neural Network, Vol:9, 1996, pp:1223–1240.         [ Links ]

[9] H. R. Mohseni, A. Maghsoudi and M. Shamsollahi, Seizure detection EEG signals: A comparison of different approaches. Proceeding of the IEEE Annual International Conference on Engineering in Medicine and Biology Society, Aug. 30–Sep 3, 2006, pp: 6724–6727.         [ Links ]

[10] R. G. Andrzejak G. Widman, K. Lehnertz, C. Rieke, P. David and C. E. Elger, Epileptic process as nonlinear determinstics dynamics in astochastic environment: An evaluation on mesial temporal lobe epilepsy. Epilepsy Res., Vol: 44, 2001, pp:129–140.         [ Links ]

[11] E. Niedermeyer, Silva FLDa. Electroencephalography. 5th Ed., Lippincott Williams and Wilkings, 2005, pp: 790–800.         [ Links ]

[12] D. Novak, L. Lhotska, EEG and VEP signal processing. Cybernetics, Faculty of Electrical Eng, 2004, pp: 50–53.         [ Links ]

[13] Y. U. Khan, J. Gotman, 2003. Wavelet based automatic seizure detection in intercerebral electroence phalogram . Clin. Neurophsiol., Vol:114, 2003, pp: 898–908.         [ Links ]

[14] S. Mallat, A Wavelet Tour of Signal Processing. 2nd Edn., Academic Press, San Diego, 1999, pp: 637–653.         [ Links ]

[15] M. B. Menhaj, Fundamentals of neural network. Amir Kabir press, 2005, pp:700–730.         [ Links ]

[16] N. Pradhan, P. K. Sadasivan and G. R. Arunodaya, Detection of seizure activity in EEG by an artificial neural network. A preliminary study. Comput. Biomed., Res. Vol:29, 1996, pp:303–313.         [ Links ]

[17] P. Addison, Illustrated Wavelet Transform Handbook. 1st Edn., Taylor and Francis, USA., ISBN: 10: 0750306920, 2002, pp: 7–17.         [ Links ]

[18] M. Aminoff, Electroencephalography: General Principles and Clinical Application Applications. In: Electrodiagnosis in Clinica Neurology. Library of Congress Cataloging in Publication Data, 2005, pp: 37–80.         [ Links ]

[19] M. E. Ohadi, Electroencephalography. Etelat, 1998, pp: 700–790.         [ Links ]

[20] D. Willam, The Nervous System. In: Physiology, Berne, R.M., M.N. Levy, B.M. Koeppen and B.A. Stanton (Eds.). New York, 2004, pp: 62–206.         [ Links ]

[21] I Rampil , A primer for EEG signal processing in anesthesia. Anesthesiology, Vol: 89, 1998, pp: 980–1002.         [ Links ]

[23] A. Mertins, Signal Analysis Wavelet Filter Banks. Time–Frequency Transform and Application., Mc Graw Hill, 2002.         [ Links ]

[24] The Mathwork Inc. MA Matlab User's Guide. Wavelet and Nntool Toolbox. 2007.         [ Links ]

[25] L. D. Iasemidis, D. S. Shiar, C. Wongse, J. C. Sackelles, P. M. Pradalos and J. C. Principe, Adaptive epileptic seizure predication system. IEEE Trans. Biomed. Eng., Vol: 50, 2003, pp: 616–627.         [ Links ]