Quantification of Cycling Smoothness in Children with Cerebral Palsy

Smoothness is a hallmark of skilled, coordinated movement, however, mathematically quantifying movement smoothness nuanced. Several metrics exist, each having its own limitations and may be specific to particular motion such as upper limb reaching. To date, there no consensus on which metric the most appropriate for assessing cycling in children with cerebral palsy (CP). We evaluated ability four preexisting metrics, dimensionless jerk, spectral arc length measure, roughness index, cross-correlation; two new root mean square error, quantify dataset from CP (mean age 13.7 ± 2.6 years). First, measure repeatability distinguishing between different levels un-smoothness, we applied set simulated crank signals known number aberrant revolutions using subjects’ actual angle data. Second, used discriminant function analysis statistically compare strength six relative other, discriminate smooth obtained typically developed (TD), control group 14.9 1.4 years), less smooth, halted CP. Our results show that (1) ArcL showed highest accurately an unsmooth when same were presented order, (2) DJ had discriminatory differentiate motion. Combining analysis, was repeatable sensitive identifying unsmooth, can hence future studies changes pre- vs. post-interventions. Further, this serve tool track motor recovery not just individuals but other patient populations similar neurological deficits present halted,