A New Wavelet-Based Neural Network for Classification of Epileptic-Related States using EEG

A New Wavelet-Based Neural Network for Classification of Epileptic-Related States using EEG. Juárez-Guerra, E., Alarcon-Aquino, V., Gómez-Gil, P., Ramírez-Cortés, J., M., & García-Treviño, E., S. Journal of Signal Processing Systems, 92(2):187-211, 2, 2020.

Website doi abstract bibtex 3 downloads

In this paper, we present a novel neural network able to classify epileptic seizures using electroencephalogram (EEG) signals, called “Multidimensional Radial Wavelons Feed-Forward Wavelet Neural Network” (MRW-FFWNN). The network is part of a classification system, which distinguishes among three brain states related to epilepsy namely ictal, interictal and healthy. Efficient methods for pre-processing EEG’s, extracting features and getting the final class decisions were selected using a statistical three-fold cross-validation method, which assures the robustness of the system and its generalization ability. The following methods were systematically analyzed to find the most appropriate for this problem: 1) Infinite Impulse Response (IIR) and Finite Impulse Response (FIR) filters for noise reduction; 2) discrete Wavelet Transform (DWT) and Maximal Overlap Discrete Wavelet Transform (MODWT) for frequency decomposition of the EEG signals; 3) average correlation and maximum voting correlation for selecting a suitable mother wavelet for frequency decomposition; 4) Binary-tree and one-vs-one (OVO) decomposition strategies for primary binary classification; 5) voting and weighted-voting strategy aggregation strategies for the final classification. The integrated system was assessed using a three-fold cross validation, applied to a benchmark provided by the University of Bonn, getting an average accuracy of 93.33% when tested using sets Z, S and F and 95.0% when sets Z, S, F and O were used. The proposed network got competitive accuracy, compared with other state-of-the art classifiers, training in almost a half of the time than the ones with similar accuracy.

@article{
 title = {A New Wavelet-Based Neural Network for Classification of Epileptic-Related States using EEG},
 type = {article},
 year = {2020},
 keywords = {EEG analysis,Epileptic seizure detection,Machine learning classification,Wavelet selection,Wavelet-based neural networks},
 pages = {187-211},
 volume = {92},
 websites = {http://link.springer.com/10.1007/s11265-019-01456-7},
 month = {2},
 day = {17},
 id = {1fbc70b1-99e2-337b-afcd-9db2a3e7c397},
 created = {2022-08-29T17:42:21.104Z},
 file_attached = {false},
 profile_id = {940dd160-7d67-3a5f-b9f8-935da0571367},
 group_id = {92fccab2-8d44-33bc-b301-7b94bb18523c},
 last_modified = {2022-08-29T17:42:21.104Z},
 read = {false},
 starred = {false},
 authored = {false},
 confirmed = {true},
 hidden = {false},
 private_publication = {false},
 abstract = {In this paper, we present a novel neural network able to classify epileptic seizures using electroencephalogram (EEG) signals, called “Multidimensional Radial Wavelons Feed-Forward Wavelet Neural Network” (MRW-FFWNN). The network is part of a classification system, which distinguishes among three brain states related to epilepsy namely ictal, interictal and healthy. Efficient methods for pre-processing EEG’s, extracting features and getting the final class decisions were selected using a statistical three-fold cross-validation method, which assures the robustness of the system and its generalization ability. The following methods were systematically analyzed to find the most appropriate for this problem: 1) Infinite Impulse Response (IIR) and Finite Impulse Response (FIR) filters for noise reduction; 2) discrete Wavelet Transform (DWT) and Maximal Overlap Discrete Wavelet Transform (MODWT) for frequency decomposition of the EEG signals; 3) average correlation and maximum voting correlation for selecting a suitable mother wavelet for frequency decomposition; 4) Binary-tree and one-vs-one (OVO) decomposition strategies for primary binary classification; 5) voting and weighted-voting strategy aggregation strategies for the final classification. The integrated system was assessed using a three-fold cross validation, applied to a benchmark provided by the University of Bonn, getting an average accuracy of 93.33% when tested using sets Z, S and F and 95.0% when sets Z, S, F and O were used. The proposed network got competitive accuracy, compared with other state-of-the art classifiers, training in almost a half of the time than the ones with similar accuracy.},
 bibtype = {article},
 author = {Juárez-Guerra, E. and Alarcon-Aquino, V. and Gómez-Gil, P. and Ramírez-Cortés, J. M. and García-Treviño, E. S.},
 doi = {10.1007/s11265-019-01456-7},
 journal = {Journal of Signal Processing Systems},
 number = {2}
}

Downloads: 3