Reliability of Human Running Analysis With Low-Cost Inertial and Magnetic Sensor Arrays

• Published in: IEEE sensors journal Vol. 21; no. 13; pp. 15299 - 15307
• Main Authors: Struber, Lucas, Ledouit, Simon, Daniel, Olivier, Barraud, Pierre-Alain, Nougier, Vincent
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.07.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accuracy; Algorithms; Calibration; Cost analysis; Estimation; Fitness equipment; Gait; gait and running analysis; Inertial sensing devices; Inertial sensors; Kinematics; Low cost; Magnetic sensors; Magnetometers; Misalignment; Optoelectronic devices; Parameters; Quaternions; Rehabilitation; Reliability analysis; Reproducibility; Reproducibility of results; Running; Sensor arrays; Sensors; system validation; Treadmills
• ISSN: 1530-437X; 1558-1748
• DOI: 10.1109/JSEN.2021.3071854

In the last years, the use of inertial sensors has increased due to their time efficiency and portability. This paper introduces a low-cost inertial system for gait and running analysis embedded on running tights, S-Move, designed with an open-source orientation estimation algorithm. Calibration and misalignment correction procedure are presented, as well as methods and algorithms that automatically assess mean stride kinematics and spatiotemporal parameters. The main objective of this study was to validate S-Move reliability through comparison of its repeatability with a commonly used optoelectronic device for angular kinematic stride and an instrumented treadmill for spatiotemporal parameters. Thirty subjects performed three sessions of walking and running being alternatively equipped by two operators. To analyze repeatability of devices, multivariate ANOVAs were performed on the measured differences between sessions. Direct comparisons between devices were also conducted. Devices comparison showed that both devices presented a moderate absolute reproducibility - between 2 and 10° - and a very good relative (after mean suppression) reproducibility - max. 4.5°. A non-negligible measure difference between systems was observed but it could not be attributed to the lack of accuracy of one device. Regarding spatiotemporal parameters, treadmill measurement was very reproducible - about 2%. S-Move reproducibility and accuracy was similar on temporal parameters but lower on spatial parameters - about 5%. However, accuracy was concordant with other studies using inertial devices. Finally, this study showed that accessible, embeddable and automatic gait and running analysis is suitable for sports performance analysis and rehabilitation is possible, at very low costs.