Examination of the prediction of the planar piecewise continuous lumped muscle parameter model for walking gait with ankle-foot orthosis

Goal: This work examines the use of a previously described piecewise continuous lumped muscle parameter (PPCLMP) model for predicting selected gait parameters for walking without and with ankle-foot orthoses (AFOs) of varying stiffnesses.Methods: Two AFOs with low (3.4 Nm/deg) and high (6.9 Nm/deg) stiffnesses were tested on the left leg of six healthy subjects to examine the model prediction on the influence of different AFO stiffnesses on gait.Results: The model prediction errors ranged from 0 % to 70 % for step lengths with root mean square error (RMSE) of 0.15 m and ranged from 0 % to 67 % for swing time with RMSE of 0.07 s. The prediction precision of step length was more consistent among subjects than of swing time.Discussions and conclusions: The model predicts the observed shortened step lengths and swing times, but there were significant differences between predicted and observed swing times and step lengths. The causes of these differences might be differences in the lumped muscle parameters taken from the literature and those of the subjects tested. Also, the model assumption that muscle stiffness is proportional to joint angle may not be corrected.