Citation

Bahari H, Vette AH, Hebert JS, Rouhani H. J. Biomech. 2019; ePub(ePub): ePub.

Affiliation

Department of Mechanical Engineering, University of Alberta, Donadeo Innovation Centre for Engineering, Edmonton, Alberta T6G 1H9, Canada; Glenrose Rehabilitation Hospital, Alberta Health Services, 10230 - 111 Avenue NW, Edmonton, Alberta T5G 0B7, Canada. Electronic address: hrouhani@ualberta.ca.

Copyright

(Copyright © 2019, Elsevier Publishing)

DOI

10.1016/j.jbiomech.2019.109315

PMID

31455499

Abstract

The biomechanical mechanisms of loss of balance have been studied before for slip condition but have not been investigated for arbitrary perturbation profiles under non-slip conditions in sagittal plane. This study aimed to determine the thresholds of center of mass (COM) velocity and position relative to the base of support (BOS) that predict forward and backward loss of balance during walking with a range of BOS perturbations. Perturbations were modeled as sinusoidal BOS motions in the vertical or anterior-posterior direction or as sagittal rotation. The human body was modeled using a seven-link model. Forward dynamics alongside with dynamic optimization were used to find the thresholds of initial COM velocity for each initial COM position that would predict forward or backward loss of balance. The effects of perturbation frequency and amplitude on these thresholds were modeled based on the simulation data. Experimental data were collected from 15 able-bodied individuals and three individuals with disability during perturbed walking. The simulation results showed similarity with the stability region reported for slip and non-slip conditions. The feasible stability region shrank when the perturbation frequency and amplitude increased, especially for larger initial COM velocities. 89.5% (70.9%) and 82.4% (68.2%) of the measured COM position and velocity combinations during low (high) perturbations were located inside the simulated limits of the stability region, for able-bodied and disabled individuals, respectively. The simulation results demonstrated the effects of different perturbation levels on the stability region. The obtained stability region can be used for developing rehabilitative programs in interactive environments.

Copyright © 2019 Elsevier Ltd. All rights reserved.

Language: en

Keywords

Dynamic optimization; Fall prevention; Forward dynamics; Musculoskeletal modeling; Walking stability