Wavelet and Region-Specific EEG Signal Analysis for Studying Post-Stroke Rehabilitation

Post-stroke monitoring is a crucial step for properly studying the progress of stroke patients. The rehabilitation process consists of exercise regimes that help i n constantly engaging the affected part of the brain leading to faster recovery. The work here studies the effectiveness of the rehabilitation regime by investigating several parameters that can play important role in observing the immediate effect of the exercises. Various parameters from different wavelet coefficients were extracted for monitoring re habilitation for up to 90 days. Energy and waveform length show maximum variation when monitoring pre and post-exercise changes. The parameters we re correlated with clinical(FMA) score. Centroid Index gave high correlation value for beta ban d (r = -0.559). Alpha band on the other hand showed a good correlation with all the extracted features, maximum being -0.6988 with energy. So for monitoring post-stroke rehabilitation alph a and beta bands should be focused. Region-specific analyses were also done to monitor changes in different parts of the brain.