Fractal theory based Non-linear analysis of sEMG

This research examines the use of fractal theory to study the properties of surface electromyogram (sEMG). The paper reports identifying a new fractal feature, maximum fractal length (MFL) that, along with fractal dimensions, has been found to be useful in modelling the muscle activity. Experimental results demonstrate that the combination of fractal dimension and maximum fractal length of sEMG recordings is suitable for characterising the activation of subtle gestures. The log-log plots demonstrate the presence of patterns of these features for each of the hand gesture that can be related to the muscle anatomy. The results indicate that there is small inter-experimental variation but there are large inter-subject variations. This inter-subject variation may be attributable to anatomical differences for the different subjects. The possible applications of this research include use of classifying sEMG for controlling machines and computers.