论文部分内容阅读
Since epileptic seizures occur irregularly and unpredictably,automatic seizure detection in EEG recording is significant in the diagnosis of patients with epilepsy.This paper proposed a new epileptic EEG classification approach based on spectral features and extreme learning machine.The spectral power in each 2 Hz width sub-band was measured.Then the principal spectral features were fed to the extreme learning machine(ELM)to discriminate between seizure and non-seizure EEG epochs.The experiments demonstrate that the proposed system can achieve satisfying detection accuracy with low cost of false detection.ELM classifier can reach faster learning speed than conventional learning algorithm.