Brainwave Classification for Acute Ischemic Stroke Group Level Using k-NN Technique

In this study, Relative Power Ratio (RPR) technique is used to analyze the Power Spectral Density (PSD) of the EEG signal. RPR technique is used to observe the difference value of power spectral between sub bands for difference level of group for stroke. In this research, more than hundred stroke patients brainwave activity with open eyes (OE) session are used. Then, they are group into Early Group (EG), Intermediate Group (IG) and Advance Group (AG). The different groups were classified by using k-nearest neighbour (k-NN) method. There are significant different of the EEG signals due to the stroke level. Beta, Alpha, Theta and Delta bands were used as input signals for the classification model. In this study, results from k-NN classification are 100% and 85 % accuracy for training and testing data set respectively. These results proved that k-NN can be used in order to predict the stroke group levels.