A Collection of Contemporary Research Articles in Electronics, Communication & Computation- Book ISBN- 978-81-948050-7-6

Authors: N Shweta, Nagendra H

Abstract: Seizure is a period of symptoms due to abnormal neuronal activity in the brain. An epileptic seizure is one of the most serious neurological disorders. Approximately 50 million people have suffered from a seizure. Electroencephalogram (EEG) is the measurement and recording of electrical activity in the brain for diagnosis purpose. Detection of epileptic seizures could be very useful for patient safety. The various method has been proposed to detect the seizure. The main objective is to analyze and detection of the seizure using a single-channel, i.e. F4-C4 (Frontal-central). In the proposed algorithm analysis of EEG signals using single-channel frontal- central (F4-C4) are formed by extracting the alpha-beta gamma delta and theta bands from the EEG signals using a bandpass filter, then the classifiers using neural network and support vector machine is used separately to detect the seizure. The experimental result shows that the proposed method effectively detects the seizure in the EEG signal using the F4-C4 Channel and also showed a reasonable accuracy in detection.

Keywords: - Electroencephalography (EEG), Seizure, Discrete Wavelet transform (DWT) Artificial neural network (ANN) Support vector machine (SVM).