A Muscle Model Incorporating Fiber Architecture Features for the Estimation of Joint Stiffness During Dynamic Movement

Abstract

Quantifying human joint stiffness in vivo during movement remains challenging. Well established stiffness estimation methods include system identification and the notion of quasi-stiffness, with experimental and conceptual limitations, respectively. Joint stiffness computation via biomechanical models is an emerging solution to overcome such limitations. However, these models make assumptions that hamper their generalization across muscle architectures. Here we present a stiffness formulation that considers the muscle's pennation angle, and its comparison to a simpler formulation that does not. Model-based stiffness estimates are evaluated against joint-perturbation-based system identification. Results on muscles with different pennation angle show that our formulation seamlessly adjusts the muscle-tendon units' stiffness depending on their architecture. At the joint level, our new model improved the stiffness estimations. Our study's relevance is the creation and validation of a modeling formulation that does not require joint perturbation. This will enable better estimations and understanding of stiffness properties and human movement.

Unifying system identification and biomechanical formulations for the estimation of muscle, tendon and joint stiffness during human movement

“…We propose high-density electromyography (HD-EMG)-driven musculoskeletal modeling techniques [48] for the estimation of muscle, tendon, and joint forces as a function of measured HD-EMG signals and joint angles [49][50][51]. This methodology was recently extended to estimate muscle, tendon, MTU-equivalent, as well as joint stiffness as proposed by Sartori et al [45] and generalized to account for pennation angles in [52]. In this section we provide a further refined formulation.…”

Estimating Biological Stiffness Without Relying on External Joint Perturbations: A Musculoskeletal Modeling Framework

“…In this way, we have a dynamically consistent framework that properly projects the stiffness from the smaller anatomical levels, i.e., muscles and tendons, to the joint level. This muscle stiffness formulation, generalizable to any pennation angle, can lead to significant differences in MTU stiffness estimations in pennated muscles, further improving joint stiffness estimation, as we showed in [85].…”

“…We propose high-density electromyography (HD-EMG)-driven musculoskeletal modeling techniques [82] for the estimation of muscle, tendon, and joint forces as a function of measured HD-EMG signals and joint angles [28,83,84]. This methodology was recently extended to estimate muscle, tendon, MTU-equivalent, as well as joint stiffness as proposed by Sartori et al [42] and generalized to account for pennation angles in [85]. In this section we provide a further refined formulation.…”