Validation of a single-sensor sternal accelerometry system for quantifying spatiotemporal gait stability in healthy adults

Clinical balance assessment during gait often relies on observational scales prone to ceiling effects, which may overlook subtle motor deficits. We introduce and validate MotionTracking, a single-sensor inertial measurement system designed to quantify spatiotemporal indicators of dynamic gait stability from trunk accelerations. Twenty healthy adults (mean age 46 ± 16.8 years) performed 10-meter walking task while wearing the MotionTracking system on the upper sternum. Simultaneous 2D video recording served as an independent temporal reference for gait-event timing. Triaxial accelerations were processed using a zero-phase, second-order 9 Hz low-pass filter to prevent phase distortion. Heel-strike (HS) and toe-off (TO) events were identified as local minima in the vertical and anteroposterior acceleration signals, respectively. From these HS → TO → HS sequences, we derived standard spatiotemporal parameters alongside symmetry ratio, cadence and propulsion efficiency. Single-support time estimated by MotionTracking showed excellent correlation with 2D video (Pearson’s r = 0.67, p < 0.001) and negligible bias on Bland–Altman analysis (bias = − 0.0007; lower limits −0.029 and upper limit 0.020). Linear regression models revealed significant age-related declines in symmetry, propulsion efficiency, Stance and SS duration, alongside an increase in DS duration. The MotionTracking system’s single-sensor configuration and zero-phase event detection provide a portable, reproducible, and quantitative assessment of dynamic gait stability. These results support MotionTracking as a potential practical tool for longitudinal monitoring and data-informed decision-making in gait and balance assessment.