New Automatic EEG Epileptic Seizure Detection Approach Using Sliding Discrete Fourier Transform and Machine Learning Techniques

Automatic seizure detection is important for fast detection because the expert denoted and searching for seizures in the long signal takes time. The most common way to detect seizures automatically is to use an electroencephalogram (EEG). In this study, sliding discrete Fourier transform (SDFT) is applied for conversion to a frequency domain by using a simple IIR structure with other fourteen features extracted from the EEG database of Bonn University. These fifteen features used as input to classifier for seizure detection, a two-classifier feedforward neural network (FFNN) and an adaptive network-based fuzzy inference system (ANFIS) used. The results appear that the highest accuracy is 99.74% with FFNN and 99.67 ANFIS. Also, when measuring the feature importance in classification for each feature extraction method and compare the results, the SDFT has more importance than other features used in this study for the classification.