Detecting Epileptic Seizures Using Abe Entropy, Line Length and SVM Classifier

Epilepsy is a 4th prevalent neurological disorder which affects the individuals in all ages around the world. Epilepsy disorder is characterized by the abnormal movements of human muscles, called seizure, as a result of the abnormality in the brain electrical activity. The electroencephalogram (EEG) can serve as a powerful tool for detecting Epilepsy. In this paper, the most commonly used Andrzejak database is utilized for building an automated system for epilepsy detection. Digital Wavelet Transform (DWT) is applied on the segmented EEG signals to extract the five EEG sub-bands (delta, theta, alpha, beta, and gamma). Approximation and Abe entropies along with line length are calculated for the extracted sub-bands. Support Vector Machine (SVM) classifier with Radial Basis Function (RBF) kernel function is used to distinguish between three classes: (1) normal, (2) interictal (seizure free interval), and (3) ictal (during seizure). The best accuracies achieved are 93.75%, 98.75% and 98.125% for normal, interictal and ictal classes respectively. These accuracies are achieved using the combination of both Abe entropy and line length features together.

[1]  Daniel Graupe,et al.  A neural-network-based detection of epilepsy , 2004, Neurological research.

[2]  Abdulhamit Subasi.  Automatic detection of epileptic seizure using dynamic fuzzy neural networks , 2006, Expert Syst. Appl..

[3]  U. Rajendra Acharya,et al.  Erratum to "Entropies for detection of epilepsy in EEG" [Comput. Methods Prog. Biomed. 80 (2005) 187-194] , 2006, Comput. Methods Programs Biomed..

[4]  K Lehnertz,et al.  Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity: dependence on recording region and brain state. , 2001, Physical review. E, Statistical, nonlinear, and soft matter physics.

[5]  Christian Beck,et al.  Generalised information and entropy measures in physics , 2009, 0902.1235.

[6]  Duoqian Miao,et al.  Best basis-based wavelet packet entropy feature extraction and hierarchical EEG classification for epileptic detection , 2011, Expert Syst. Appl..

[7]  Sabine Van Huffel,et al.  Line length as a robust method to detect high-activity events: Automated burst detection in premature EEG recordings , 2014, Clinical Neurophysiology.

[8]  K. Srinivasa,et al.  Epileptic Seizures Classification from EEG Signals using Neural Networks , 2012 .

[9]  V. Srinivasan,et al.  Approximate Entropy-Based Epileptic EEG Detection Using Artificial Neural Networks , 2007, IEEE Transactions on Information Technology in Biomedicine.

[10]  Vicente Alarcon-Aquino,et al.  Epilepsy Seizure Detection in EEG Signals Using Wavelet Transforms and Neural Networks , 2015 .

[11]  Abdulhamit Subasi,et al.  Neural Networks with Periodogram and Autoregressive Spectral Analysis Methods in Detection of Epileptic Seizure , 2004, Journal of Medical Systems.

[12]  Sohaib Majzoub,et al.  E EG - Based Epileptic Seizures Detection with Adaptive Learning Capability , 2017 .

[13]  Daniel Rivero,et al.  Automatic epileptic seizure detection in EEGs based on line length feature and artificial neural networks , 2010, Journal of Neuroscience Methods.

[14]  Chia-Ping Shen,et al.  Epileptic Seizure Detection for Multichannel EEG Signals with Support Vector Machines , 2011, 2011 IEEE 11th International Conference on Bioinformatics and Bioengineering.

[15]  Daniel Rivero,et al.  Epileptic seizure detection using multiwavelet transform based approximate entropy and artificial neural networks , 2010, Journal of Neuroscience Methods.

[16]  Stefano Di Gennaro,et al.  CLASSIFICATION OF EEG SIGNALS FOR DETECTION OF EPILEPTIC SEIZURES BASED ON WAVELETS AND STATISTICAL PATTERN RECOGNITION , 2014 .

[17]  Mohamed F. Tolba,et al.  EEG based epilepsy detection using approximation entropy and different classification strategies , 2017, 2017 Eighth International Conference on Intelligent Computing and Information Systems (ICICIS).