A novel selection of optimal statistical features in the DWPT domain for discrimination of ictal and seizure-free electroencephalography signals
Properly determining the discriminative features which characterize the inherent behaviors of electroencephalography (EEG) signals remains a great challenge for epileptic seizure detection. In this present study, a novel feature selection scheme based on the discrete wavelet packet decomposition and...
| Main Authors: | , , |
|---|---|
| Formato: | Artigo |
| Idioma: | English |
| Publicado em: |
Springer Nature
2018
|
| Assuntos: | |
| Acesso em linha: | http://eprints.uthm.edu.my/5453/1/AJ%202018%20%28190%29.pdf |
| _version_ | 1825710025211904000 |
|---|---|
| author | Ong, Pauline Zainuddin, Zarita Kee, Huong Lai |
| author_facet | Ong, Pauline Zainuddin, Zarita Kee, Huong Lai |
| author_sort | Ong, Pauline |
| collection | UTHM |
| description | Properly determining the discriminative features which characterize the inherent behaviors of electroencephalography (EEG) signals remains a great challenge for epileptic seizure detection. In this present study, a novel feature selection scheme based on the discrete wavelet packet decomposition and cuckoo search algorithm (CSA) was proposed. The normal as well as epileptic EEG recordings were frst decomposed into various frequency bands by means of wavelet packet decomposition, and subsequently, statistical features at all developed nodes in the wavelet packet decomposition tree were derived. Instead of using the complete set of the extracted features to construct a wavelet neural networks-based classifer, an optimal feature subset that maximizes the predictive competence of the classifer was selected by using the CSA. Experimental results on the publicly available benchmarks demonstrated that the proposed feature subset selection scheme achieved promising recognition accuracies of 98.43–100%, and the results were statistically signifcant using z-test with p value <0.0001. |
| first_indexed | 2024-03-05T21:51:20Z |
| format | Article |
| id | uthm.eprints-5453 |
| institution | Universiti Tun Hussein Onn Malaysia |
| language | English |
| last_indexed | 2024-03-05T21:51:20Z |
| publishDate | 2018 |
| publisher | Springer Nature |
| record_format | dspace |
| spelling | uthm.eprints-54532022-01-09T07:37:18Z http://eprints.uthm.edu.my/5453/ A novel selection of optimal statistical features in the DWPT domain for discrimination of ictal and seizure-free electroencephalography signals Ong, Pauline Zainuddin, Zarita Kee, Huong Lai RC Internal medicine Properly determining the discriminative features which characterize the inherent behaviors of electroencephalography (EEG) signals remains a great challenge for epileptic seizure detection. In this present study, a novel feature selection scheme based on the discrete wavelet packet decomposition and cuckoo search algorithm (CSA) was proposed. The normal as well as epileptic EEG recordings were frst decomposed into various frequency bands by means of wavelet packet decomposition, and subsequently, statistical features at all developed nodes in the wavelet packet decomposition tree were derived. Instead of using the complete set of the extracted features to construct a wavelet neural networks-based classifer, an optimal feature subset that maximizes the predictive competence of the classifer was selected by using the CSA. Experimental results on the publicly available benchmarks demonstrated that the proposed feature subset selection scheme achieved promising recognition accuracies of 98.43–100%, and the results were statistically signifcant using z-test with p value <0.0001. Springer Nature 2018 Article PeerReviewed text en http://eprints.uthm.edu.my/5453/1/AJ%202018%20%28190%29.pdf Ong, Pauline and Zainuddin, Zarita and Kee, Huong Lai (2018) A novel selection of optimal statistical features in the DWPT domain for discrimination of ictal and seizure-free electroencephalography signals. Pattern Analysis and Applications, 21. pp. 515-527. ISSN 1433-7541 |
| spellingShingle | RC Internal medicine Ong, Pauline Zainuddin, Zarita Kee, Huong Lai A novel selection of optimal statistical features in the DWPT domain for discrimination of ictal and seizure-free electroencephalography signals |
| title | A novel selection of optimal statistical features in the DWPT domain for discrimination of ictal and seizure-free electroencephalography signals |
| title_full | A novel selection of optimal statistical features in the DWPT domain for discrimination of ictal and seizure-free electroencephalography signals |
| title_fullStr | A novel selection of optimal statistical features in the DWPT domain for discrimination of ictal and seizure-free electroencephalography signals |
| title_full_unstemmed | A novel selection of optimal statistical features in the DWPT domain for discrimination of ictal and seizure-free electroencephalography signals |
| title_short | A novel selection of optimal statistical features in the DWPT domain for discrimination of ictal and seizure-free electroencephalography signals |
| title_sort | novel selection of optimal statistical features in the dwpt domain for discrimination of ictal and seizure free electroencephalography signals |
| topic | RC Internal medicine |
| url | http://eprints.uthm.edu.my/5453/1/AJ%202018%20%28190%29.pdf |
| work_keys_str_mv | AT ongpauline anovelselectionofoptimalstatisticalfeaturesinthedwptdomainfordiscriminationofictalandseizurefreeelectroencephalographysignals AT zainuddinzarita anovelselectionofoptimalstatisticalfeaturesinthedwptdomainfordiscriminationofictalandseizurefreeelectroencephalographysignals AT keehuonglai anovelselectionofoptimalstatisticalfeaturesinthedwptdomainfordiscriminationofictalandseizurefreeelectroencephalographysignals AT ongpauline novelselectionofoptimalstatisticalfeaturesinthedwptdomainfordiscriminationofictalandseizurefreeelectroencephalographysignals AT zainuddinzarita novelselectionofoptimalstatisticalfeaturesinthedwptdomainfordiscriminationofictalandseizurefreeelectroencephalographysignals AT keehuonglai novelselectionofoptimalstatisticalfeaturesinthedwptdomainfordiscriminationofictalandseizurefreeelectroencephalographysignals |