TY - GEN
T1 - An Intelligent System to Classify Epileptic and Non-Epileptic EEG Signals
AU - Qazi, Emad Ul Haq
AU - Hussain, Muhammad
AU - Aboalsamh, Hatim
AU - Abdul, Wadood
AU - Bamatraf, Saeed
AU - Ullah, Ihsan
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2017/4/21
Y1 - 2017/4/21
N2 - Epilepsy is a neurological disorder disease that affects more than 55 million people in the world. In this paper, we have proposed an efficient intelligent pattern recognition system for the classification of epileptic and non-epileptic electroencephalogram (EEG) signals. For this purpose, we used state-of-the-art machine learning technique, i.e., SVM (support vector machines) to classify epileptic and non-epileptic signals. Two (02) different classes of signals are used in this study, i.e., non-epileptic with open eyes and epileptic in seizure condition. One hundred (100) subjects from each class were employed for extraction of discriminatory features and classification purpose. After pre-processing of EEG signals, we use discrete wavelet transform (DWT) to decompose signals upto level 5. Then various features, i.e., energy, entropy and standard deviation are extracted from wavelet bands. Next, we use these features in the classification of signals. We achieved the classification accuracy of 100 % at delta band (0 to 3 Hz) and theta band (3 to 6 Hz). The comparisons with the previous studies show the significance of this system, which can be utilized in real-time as well as in offline clinical applications.
AB - Epilepsy is a neurological disorder disease that affects more than 55 million people in the world. In this paper, we have proposed an efficient intelligent pattern recognition system for the classification of epileptic and non-epileptic electroencephalogram (EEG) signals. For this purpose, we used state-of-the-art machine learning technique, i.e., SVM (support vector machines) to classify epileptic and non-epileptic signals. Two (02) different classes of signals are used in this study, i.e., non-epileptic with open eyes and epileptic in seizure condition. One hundred (100) subjects from each class were employed for extraction of discriminatory features and classification purpose. After pre-processing of EEG signals, we use discrete wavelet transform (DWT) to decompose signals upto level 5. Then various features, i.e., energy, entropy and standard deviation are extracted from wavelet bands. Next, we use these features in the classification of signals. We achieved the classification accuracy of 100 % at delta band (0 to 3 Hz) and theta band (3 to 6 Hz). The comparisons with the previous studies show the significance of this system, which can be utilized in real-time as well as in offline clinical applications.
KW - Discrete wavelet transform (DWT)
KW - Electroencephalogram (EEG)
KW - Epileptic
KW - Non-Epileptic
KW - Support Vector Machine (SVM)
UR - https://www.scopus.com/pages/publications/85019196770
U2 - 10.1109/SITIS.2016.44
DO - 10.1109/SITIS.2016.44
M3 - Conference Publication
AN - SCOPUS:85019196770
T3 - Proceedings - 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016
SP - 230
EP - 235
BT - Proceedings - 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016
A2 - De Pietro, Giuseppe
A2 - Dipanda, Albert
A2 - Chbeir, Richard
A2 - Gallo, Luigi
A2 - Yetongnon, Kokou
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016
Y2 - 28 November 2016 through 1 December 2016
ER -