Measuring upper limb function in children with hemiparesis with 3D inertial sensors

Christopher J. Newman, Roselyn Bruchez, Sylvie Roches, Marine Jequier Gygax, Cyntia Duc, Farzin Dadashi, Fabien Massé, Kamiar Aminian

Upper limb assessments in children with hemiparesis rely on clinical measurements, which despite standardization are prone to error. Recently, 3D movement analysis using optoelectronic setups has been used to measure upper limb movement, but generalization is hindered by time and cost. Body worn inertial sensors may provide a simple, cost-effective alternative. We instrumented a subset of 30 participants in a mirror therapy clinical trial at baseline, post-treatment, and follow-up clinical assessments, with wireless inertial sensors positioned on the arms and trunk to monitor motion during reaching tasks. Inertial sensor measurements distinguished paretic and non-paretic limbs with significant differences (P < 0.01) in movement duration, power, range of angular velocity, elevation, and smoothness (normalized jerk index and spectral arc length). Inertial sensor measurements correlated with functional clinical tests (Melbourne Assessment 2); movement duration and complexity (Higuchi fractal dimension) showed moderate to strong negative correlations with clinical measures of amplitude, accuracy, and fluency. Inertial sensor measurements reliably identify paresis and correlate with clinical measurements; they can therefore provide a complementary dimension of assessment in clinical practice and during clinical trials aimed at improving upper limb function.