Validation of a single inertial sensor for measuring running kinematics overground during a prolonged run

Introduction: The purpose of this study was to validate acceleration data from a single inertial sensor containinga tri-axial accelerometer, whilst running overground during a prolonged run against a motion analysis system. Methods: An inertial sensor was placed on the low back of 10 runners who performed an 8 km run on a treadmill.To provide validation of the sensor, data were collected as runners ran along a runway through a motion analysis system at the beginning and throughout the run.Results: High levels of agreement between the two systems were found in the craniocaudal and mediolateral acceleration, with antero posterior having the least agreement with greatest Typical Error of the Estimate (0.66 sample points). Very high to extremely high correlations across all testing times were found in all three directions of accelerations (r=0.75 to 0.95). Heel strike and toe off events were identified in anteroposterior and craniocaudal acceleration, with high levels of agreement and extremely high correlations (r=0.99) between the two systems.Minimal variation and change in agreement and correlation between the data at each testing time were found. Discussion: This study provides evidence that a single inertial sensor placed on the low back is valid for measuring three-dimensional acceleration in overground running during a prolonged run. Further analysis identified specific events of heel strike and toe off and were comparable between the two systems. The minimal variation and change in agreement between the two systems during the run indicates the adherence method of the inertial sensor was suitable. Conclusions: The results of this study indicate that data collected from a single inertial sensor is highly correlated with simultaneous data collected using a motion analysis system, and has the capability to identify heelstrike and toe off events in overground running throughout a prolonged fatiguing run