Estimation of the body center of mass velocity during gait of people with transfemoral amputation from force plate data integration

Abstract

Background: Body Center Of Mass velocity assessment is a prerequisite for several applications in prosthetic control and rehabilitation monitoring. Force plate data integration is a promising alternative to full-body quantitative analysis of segmental kinematics to estimate the velocity. Still, it remains to be implemented and validated for people with transfemoral amputation. Methods: Two methods were used (force plate based and pelvic markers based) for Body Center Of Mass velocity estimation in a clinical context. The two methods were comparatively assessed on overground walking data of eight people with transfemoral amputation in a laboratory equipped with a motion capture system and force plates compared to reference estimation derived from a full body segmental gait analysis. The ‘Methods’ agreement with the reference was quantified from the Bland and Altman procedure. Findings: The estimation of Body Center Of Mass velocity from force plate data integration was considered acceptable in terms of limits of agreement. In addition, the hypotheses used to determine integration constants were evaluated and shown to be reasonable as far as the walking direction is well controlled. Interpretation: Results demonstrate the possibility to use the force plate method to assess the Body Center Of Mass velocity of people with transfemoral amputation for straight walking on level ground. An estimation from the velocity of pelvic markers can also be a relevant alternative as soon as the walking velocity remains low. Further investigation will deal with the impact of the errors on the computation of derived parameters such as individual limb power.