Comparison of 4 different smoothness metrics for the quantitative assessment of movement's quality in the upper limb of subjects with cerebral palsy

Abstract

This article discusses the results obtained by computing 4 different quantitative metrics for the assessment of the upper limb, using the elbow angle as the datum for the experiment. This angle was obtained from a commercial Motion Tracking system with 19 subjects: 10 with cerebral palsy (CP) and 9 with no neurological impairment. Subjects performed one task, consisting in reaching a small object (elbow extension) and bringing it to their mouths (elbow flexion). A MATLAB toolbox was implemented in order to compute the logarithmic dimensionless jerk metric, the mean arrest period ratio, the spectral arc-length metric and the peaks metric, as well as the maximum peak angle, minimum peak angle, maximum peak velocity, maximum peak acceleration and minimum angular acceleration. Results show the consistency of the spectral arc-length metric to display the differences between CP and healthy subjects, while the peaks metric and the logarithmic dimensionless jerk metric tend to be useful under some restrictions. With this analysis, a preliminary motor impairment assessment can be deducted, by comparing the metrics data from the CP subjects with the control group.