Experimental study of prosthesis modifications based on passive dynamic walking model: A limit cycle stability analysis

The aim of the current study was to examine the effect of asymmetric prosthesis derived from the passive dynamic walking (PDW) concept on dynamic stability and symmetry in unilateral transfemoral amputees. Seventeen transfemoral amputees were asked to walk on 77.8 m in preferred speed under four conditions: 1) no added mass, 2) the knee joint relocated downwards by 18% of the total shank length, shank mass decreased by 68%, thigh mass increased by 7%, 3) the knee joint relocated downwards by 37% of the total shank length, shank mass decreased by 68%, thigh mass increased by 7% and 4) thigh mass increased 17%, shank mass decreased by 38%. Trunk accelerations were recorded by a triaxle accelerometer, attached at the L3 level of spine. For each condition, stability (orbital and local), intra-limb step length and step time variability, step length and step time symmetry, were estimated from the vertical acceleration. Our findings showed no significant change in the orbital stability (P = 0.627) and the local stability (P = 0.748). In addition, no significant difference was found in the step length symmetry (P = 0.891), intralimb step length variability (P > 0.234), the step time symmetry (P = 0.960) and intra-limb step time variability (P > 0.847) with the new prosthetic configurations. Our empirical findings indicate that contrary to the modeling predictions, manipulating physical parameters does not improve gait pattern in terms of stability, variability and symmetry in transfemoral amputees. It suggests that proposed modifications based on PDW cannot be directly applied to real human conditions without further elaborations. (C) 2020 Elsevier Ltd. All rights reserved.