Effects of kinematic complexity and number of muscles on musculoskeletal model robustness to muscle dysfunction

• Type of Material: Article
• Authors: M. Hongchul Sohn, Georgia Institute of Technology; Daniel M. Smith, Georgia Institute of Technology; Lena Ting, Emory University
• Language: English
• Date: 2019-07-24
• Publisher: Public Library of Science
• Publication Version: Final Published Version
• Copyright Statement: © 2019 Sohn et al.
• License: Creative Commons Attribution 4.0 International
• Final Published Version (URL): http://dx.doi.org/10.1371/journal.pone.0219779
• Title of Journal or Parent Work: PLoS ONE
• ISSN: 1932-6203
• Volume: 14
• Issue: 7
• Start Page: e0219779
• End Page: e0219779
• Grant/Funding Information: This work was supported by the National Institute for Child Health and Human Development (https://www.nichd.nih.gov/Pages/index.aspx) grant R01HD46922 to LHT.
• Keywords: Science & Technology; FORCE; TENDON; Science & Technology - Other Topics; DYNAMIC SIMULATIONS; ACCELERATION; PATTERNS; GAIT; Multidisciplinary Sciences; REDUNDANCY; PREDICTIONS; ACTIVATION; COMPENSATORY STRATEGIES
• Research Categories: Health Sciences, Rehabilitation and Therapy; Engineering, Biomedical

Abstract

The robustness of motor outputs to muscle dysfunction has been investigated using musculoskeletal modeling, but with conflicting results owing to differences in model complexity and motor tasks. Our objective was to systematically study how the number of kinematic degrees of freedom, and the number of independent muscle actuators alter the robustness of motor output to muscle dysfunction. We took a detailed musculoskeletal model of the human leg and systematically varied the model complexity to create six models with either 3 or 7 kinematic degrees of freedom and either 14, 26, or 43 muscle actuators. We tested the redundancy of each model by quantifying the reduction in sagittal plane feasible force set area when a single muscle was removed. The robustness of feasible force set area to the loss of any single muscle, i.e. general single muscle loss increased with the number of independent muscles and decreased with the number of kinematic degrees of freedom, with the robust area varying from 1% and 52% of the intact feasible force set area. The maximum sensitivity of the feasible force set to the loss of any single muscle varied from 75% to 26% of the intact feasible force set area as the number of muscles increased. Additionally, the ranges of feasible muscle activation for maximum force production were largely unconstrained in many cases, indicating ample musculoskeletal redundancy even for maximal forces. We propose that ratio of muscles to kinematic degrees of freedom can be used as a rule of thumb for estimating musculoskeletal redundancy in both simulated and real biomechanical systems.