Automatic EEG seizure detection using dual-tree complex wavelet-Fourier features |
| |
| Affiliation: | 1. Institute for Infocomm Research, Singapore;2. University of Hong Kong, Hong Kong;3. Centre for Computer and Information Security Research, School of Computer Science and Software Engineering, University of Wollongong, Australia;1. Research Institute of Computer Science, Technical University of Loja, San Cayetano alto, Loja, Ecuador;2. Department of Computing, Polytechnic University of Madrid, Boadilla del Monte, Madrid, Spain;1. Department of Information Technology, College of Computers and Information Technology, Taif University, Taif, Saudi Arabia;2. Department of Electrical and Computer Engineering, Concordia University, Montreal H3G 1T7, QC, Canada;3. The Concordia Institute for Information Systems Engineering (CIISE), Concordia University, Montreal H3G 1T7, QC, Canada;1. CMR Institute of Technology, AECS Layout, Bangalore, Karnataka 560037, India;2. Christ University, Hosur Road, Bangalore, Karnataka 560029, India;1. Universidade Federal de Ouro Preto, Computing Department, Ouro Preto, MG, Brazil;2. Universidade Federal de Minas Gerais, Computer Science Department, 31.270-010 Belo Horizonte, MG, Brazil;1. School of Mathematical Science, Anhui University, Hefei, Anhui 230601, China;2. School of Business, Anhui University, Hefei, Anhui 230601, China |
| |
| Abstract: | Epilepsy is one of the most common neurological disorders with 0.8% of the world population. The epilepsy is unpredictable and recurrent, so it is very difficult to treat. In this paper, we propose a new Electroencephalography (EEG) seizure detection method by using the dual-tree complex wavelet (DTCWT) – Fourier features. These features achieve perfect classification rates (100%) for the EEG database from the University of Bonn. These classification rates outperform a number of existing EEG seizure detection methods published in the literature. However, it should be mentioned that several recent works also achieved this perfect classification rate (100%). Our proposed method should be as good as these works since our method only performs the DTCWT transform for up to 5 scales and our method only conducts the FFT to the 4th and 5th scales of the DTCWT decomposition. In addition, we could replace the conventional FFT in our method by sparse FFT so that our method could be even faster. |
| |
| Keywords: | Electroencephalography (EEG) Seizure detection Dual-tree complex wavelet transform (DTCWT) Fourier transform |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
