Fractal features based technique to identify subtle forearm movements and to measure alertness using physiological signals (sEMG, EEG)

This research paper reports the use of fractal features based technique in physiological signals like surface electromyogram (sEMG), electroencephalogram (EEG) which has gained increasing attention in biosignal processing for medical and healthcare applications. This research reports the use of fractal dimension, a fractal complexity measure in physiological signals and also reports identification of a new feature of sEMG, maximum fractal length (MFL), as a better measure of small or low level changes in the human activity. The authors propose that FD is a useful indicator of the complexity in signals and MFL is a useful indicator of the level of activity, and the combination of these is suitable for identifying actions and gestures corresponding to low-level muscle contraction using surface EMG signal and using EEG to estimate operatorpsilas global level of alertness. The results indicate that MFL is correlated with the fluctuations of the userpsilas task performance and putative level.