An Approach for Experimental Investigation of Muscle Activities of the Upper Limbs (Right versus Left Arm) of Healthy Subjects and Post-stroke Patients - A Preliminary Study

The experimental estimation of muscle activities in post-stroke patients with limb motor deficit is often based on surface electromyography signals (EMGs). Besides EMGs processing and interpretation difficulties, the differences in motor control of dominant and non-dominant arm have to be taken into account since the stroke can injure either left or right limb. Different motor tasks (under static conditions - different postures, and dynamic ones, i.e. elbow flexions/extensions in the sagittal and in the horizontal planes) were performed consecutively with the dominant and the non-dominant arm of two healthy volunteers and with the non-affected and the affected arm of two stroke patients. Surface EMGs of seven muscles were recorded using a Noraxon telemetric system. Two custom-made programs written in MATLAB were developed, separately for processing the data from static and dynamic motor tasks. At first the EMG signals were filtered, rectified and smoothed. For static tasks power-frequency analysis was performed, calculating different parameters (mean and median frequency, area under power function, mean values of the smoothed signals, which would be used as coefficients for normalization, etc.). For dynamic tasks one trial of flexion/extension motions was chosen for visual expectation through the saved video records. An analysis of synergistic and antagonistic muscle activities was performed.