LAUSR

Epilepsy is a common neurological disorder that is difficult to treat. Monitoring brain activity using electroencephalography (EEG) has become an important tool for the diagnosis of epilepsy. In this paper, we propose a method for EEG seizure detection by decomposing EEG signals for up to six wavelet scales without downsampling. Then, we perform the fast Fourier transform with wavelet scales 3, 4, 5, and 6, and take the magnitude of the Fourier coefficients as features for seizure detection. We use a nearest neighbour classifier to classify the input EEG signal into the seizure or no seizure class. Experiments demonstrate that the proposed method is comparable and sometimes better than our previous dual-tree complex wavelet–Fourier method for the University of Bonn EEG database. Furthermore, the proposed method is very competitive with a number of existing EEG seizure detection methods. We found that various types of wavelet filter all achieved perfect classification rates (100%) for EEG seizure detection.