Robotic assay of arm reaching movements in diverse neurologic populations: Can movement features be reliable, disease-specific diagnostic biomarkers?

In this paper, we compared robotic measures of point-to-point arm reaching movements in those without impairment to those with impairment due to different neurological diseases. The goal was to determine if robot-derived performance metrics (robotic assay) are capable of discerning movement features unique to underlying pathology. Subjects without neurological impairment and those with clinically defined neurological movement disorders, including multiple sclerosis, Parkinson's disease, Huntington's disease and cerebellar ataxia performed unassisted, visually evoked center-out movements (14 cm out and back) using a 2 degree-of-freedom planar robot. Kinematic robot-derived measures of movement quality (speed, smoothness) and amount (path length) were computed. Analysis showed that metrics related to reaching speed-specifically, mean speed emerged as the strongest movement biomarker distinguishing between those with and without neurological impairment. Relative time-to-first speed peak, also a metric for reaching speed, emerged as the strongest differentiator between the disease groups. Overall, our results provide a starting point for analysis of functional arm reaching movements in diverse neurologic diseases to establish validity of these trends in larger disease-specific cohorts and afford comparison of robotic assay to clinical scales.