Simple Gait Symmetry Measures Based on Foot Angular Velocity: Analysis in Post Stroke Patients

In this paper, we propose symmetry measures for post stroke assessment based on gait signal profiles from inertial sensors. Ten healthy controls and eight post stroke patients performed 6-Minute Walk Tests while wearing an inertial sensor on top of each shoe. Symmetry measures based on the linear correlation and the normalized sample distance between left and right foot pitch angular velocity showed high discriminating power to differentiate post stroke gait from healthy controls (Cliff's D = 0.95, Wilcoxon test p<0.001). The proposed symmetry measures are simple to estimate and do not require spatiotemporal gait parameters while they provide comparable discriminating power than symmetry measures based on spatiotemporal gait characteristics such as maximum angular velocity and stance ratio of each cycle. The proposed symmetry measures have the potential for generalization in wearable sensor based gait symmetry assessment.