Gait stability assessment during perturbed walking

• Author / Creator

  Hosein Bahari

• Falling is a significant source of morbidity and even mortality among the elderly population. 10% to 20% of these falls lead to severe injuries that require medical attention. Understanding the mechanisms behind the loss-of-balance and fall initiation can help prevent the consequences of falls by designing preventive strategies and training to help individuals maintain their balance. Over 50% of falls occur during locomotion. The risk of falling during walking has been assessed using stability measures based on either the dynamic stability of the system or the biomechanical modeling of the human body. The former approach is better suited for balance assessment during unperturbed walking whereas the latter is suitable for both perturbed and unperturbed walking circumstances. The biomechanical mechanisms behind the loss-of-balance and falling during walking have been studied for the cases of slip and non-slip conditions. However, they have not been studied for the complex perturbation scenarios. The objectives of this study were to, first, characterize the biomechanical mechanisms of loss-of-balance during perturbed walking as a function of perturbation conditions, body motion patterns, and gait parameters, and second, propose and experimentally validate stability measures for perturbed walking conditions. We used a seven-segment model of the human body and presented a non-linear, optimization-based model of perturbed walking in the sagittal plane. We analyzed human body motions during swing phase of walking and quantified the effects of various types of external perturbations on the base of support (BOS) on the walking stability. We estimated the boundaries (limits) of the feasible stability region (FSR) as a function of the type, dominant frequency, and dominant amplitude of perturbations. To extend the use of obtained FSRs to continuous perturbed or unperturbed walking, we introduced the concept of “extended feasible stability region” (ExFSR) which included the region between the leading foot’s lower FSR threshold and swinging foot’s upper FSR threshold. Based on this concept, we introduced novel stability measures for a whole step duration in a walking trial. To experimentally validate our obtained feasible stability regions and proposed stability measures, we collected experimental data from two groups of nondisabled individuals and individuals with disability during perturbed walking. To this end, we used a Computer-Assisted Rehabilitation Environment (CAREN) to create various perturbation conditions. We evaluated the validity of the obtained feasible stability regions at the toe-off instant and ExFSR during a whole step duration by investigating their specificity in predicting the loss-of-balance (i.e., false prediction of the loss-of-balance). The specificity of obtained feasible stability regions and ExFSR were comparable to those reported in the literature for other conditions. Also, we evaluated the credibility of our proposed stability measures by showing their correlation with other biomechanical and variability-based stability measures previously introduced in the literature. This study introduced biomechanical measures to characterize the risk of loss-of-balance in the sagittal plane during perturbed walking as a function of perturbation conditions, body motion patterns, and gait parameters. These measures can be used for physiologically and biomechanically meaningful assessment of walking stability for individuals with walking disorders. The outcome of this research can contribute to our understanding of human balance control for biped walking under the effect of external perturbations, and the development of rehabilitative programs in interactive training environments such as the CAREN.

• Subjects / Keywords

  • Dynamic optimization
  • Fall prevention
  • Forward dynamics
  • Musculoskeletal modeling
  • Walking stability
  • Computational biomechanics
  • Stability measure

• Graduation date

  Spring 2019

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-k130-5j16

• License

  Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.