Acute Ischemic Stroke Brainwave Classification Using Relative Power Ratio Cluster Analysis

Abstract

This study proposed the application of cluster analysis to classify the brainwaves of stroke level based on the Relative Power Ratio (RPR) techniques. RPR was performed to determine the brainwave characteristics due to group of stroke level. In this research, we determined seventy four stroke patients brainwave activity with open eyes (OE) session. Then group them into Advance Group (AG), Intermediate Group (IG) and Early Group (EG). Simultaneously, their Electroencephalogram (EEG) was recorded from which the EEG dataset will be calculated using the RPR formula. Beta, Alpha, Theta and Delta Power Spectrum Density (PSD) are used as input for RPR. The results show that by implementing RPR technique, the pattern of group stroke level, especially in the cognitive or thinking abilities can be clearly observed. Then cluster analysis using a factorial ANOVA (analysis of variance) design with factors Group will be deployed to cluster RPR towards the corresponding group of stroke level. It can be concluded that the AG are higher in the cluster 1 for RPR Beta (RPRB) while IG and EG were grouped together because they do not differ from each other in cluster 2. For RPR Alpha (RPRA), EG are higher in the cluster 2 while AG and IG were grouped together in cluster 1. AG was found in both clusters for RPR Theta (RPRT) and AG are lower in cluster 1 for RPR Delta (RPRD). This indicates that the group stroke level can discriminate due to the brainwaves characteristics.