Behavioral Health, Mental Health, Substance Abuse
Feasibility of Using an Accelerometer to Assess Gait Asymmetry
(School of Public Health (UMD) Kinesiology Doctoral Student)
Karimpour, Rana (UMD SPH Kinesiology), Krupenevich, Rebecca L. (UMD SPH Kinesiology), Miller, Ross H. (UMD SPH Kinesiology), Shim, Jae Kun (UMD SPH Kinesiology)
Evaluation of gait asymmetry is of interest in both human movement research and clinical settings. Specifically, force asymmetry is frequently attributed to the risk or presence of pathologies and injuries. Although biomechanical research settings allow for assessment of gait asymmetry through the use of motion capture systems and force plates, these methods may not always be feasible in clinical settings. Accelerometers are an easily accessible, inexpensive means of conducting biomechanical analyses. Furthermore, evaluation of gait asymmetry with a wearable device, will provide a tool for the objective assessment of rehabilitation progress, and diagnosis of pathologies associated with force asymmetries during locomotion under free-living conditions. The purpose of this study was to investigate the relationship between asymmetries measured from ground reaction forces (GRF) measured by force plates and data from an accelerometer placed on the lower back. We hypothesized a linear correlation between the asymmetries calculated from each signal’s corresponding peak values. Twenty-five participants (5 male and 20 female, 20.8±0.76yrs) completed six different conditions of walking: one normal walking at their preferred speed, and 5 with simulated asymmetry while being kept at their preferred walking speed. Indices of left/right asymmetry were calculated from the peak three-dimensional magnitude of GRF and the data from the accelerometer. Pearson correlation coefficient was calculated to assess the relationship between the accelerometer and GRF AIs for each participant. Average correlation coefficient was calculated using the Fisher z-transformed scores. There was a significant positive correlation between accelerometer AI and GRF AI (r = 0.69, p < 0.0001). The 99% confidence interval was r = 0.61-0.75. This preliminary finding suggests that an accelerometer may be a feasible tool for measuring force asymmetry in clinical settings. Some factors that should be controlled for in future statistical analyses include inertial parameters, anthropometric measures, and gender of the participants.