Heuristic Algorithm based Approach to Classify EEG Signals into Normal and Focal

Condition of brain can be examined using the brain-signals and brain-images. Signal based evaluation is simple and offers essential information compared with the image based methods. This paper proposes an approach to evaluate the benchmark EEG signals. The implemented approach initially implements an amplitude based assessment to compute the peak-to-peak voltage value of the EEG signal. Later, it implements time-frequency conversation procedure to transfer the signal into image based on the wavelet transform. Further, the S-transform approach is considered to extract the essential signal features for the classifier system. Firefly-Algorithm (FA) based approach is also considered to choose leading signal features considered to train and test the classifier unit. In this work, classifiers, such as Support-Vector-Machine (SVM), Random-Forest (RF) and K-Nearest Neighbor (KNN) are implemented and the result of this work offered an average accuracy of 80.39%. The works confirms that, proposed procedure offers better result on the chosen EEG signals.