Cross-sectional validation of inertial measurement units for estimating trunk flexion kinematics during treadmill disturbances

Abstract

Postural perturbation training has been shown to reduce falls. The key outcome measures are trunk flexion kinematics, at recovery step. Motion capture is typically used, but requires space and trained staff. Small, inexpensive, portable inertial measurement units (IMU) are preferred for routine clinical care. IMUs have been validated for trunk motion during walking and running on a treadmill, however treadmill fall prevention training generates higher accelerations. The purpose of this study was to validate the IMU estimate of trunk kinematics against motion capture during treadmill disturbances. Ten healthy young adults had an IMU with a retro-reflective marker triad placed on their sternum to estimate trunk kinematics. Disturbances, increasing in magnitude, were delivered until the harness supported at least 50% of the subject's weight. Equivalence testing (α = 0.05) demonstrated the trunk angle (TA) and angular velocity (TAV) measured by the IMU and motion capture were equivalent. The 95% Confidence Intervals (TA: [-1, 1], TAV: [0, 17]) were within the equivalence interval (TA: [−2, 2], TAV: [−20, 20]) and the p-Values (TA: 0.005, TAV: 0.011) were less than alpha. This data confirms that IMUs provide a valid method for measuring trunk kinematics during treadmill perturbation training.