Seizure detection in EEG signals using support vector machines
A linear Support Vector Machine (SVM) classifier is designed to detect and classify seizures in EEG signals based on a few simple features such as mean, variance, dominant frequency, and the mean power spectrum. The SVM classifier is tested on a benchmark EEG database. Using a combination of these features, classification rates up to 98% were achieved. The proposed classifier that utilizes a few simple features is computationally efficient to be deployed in a real-time seizure monitoring system.