Computational simulations for the study of torque variability during a plantar flexion.

Author(s)

Renato Naville Watanabe

• Document type: Master's Dissertation
• Press: São Paulo.
• Institution: Universidade de São Paulo (USP). Escola Politécnica (EP/BC)
• Defense date: 2012-02-01

Examining board members

André Fábio Kohn; Arturo Forner Cordero; Marcos Duarte

• Advisor: André Fábio Kohn

Abstract

The muscle force variability, usually in a constant and isometric force task, has been studied both experimentally and using computational tools. However, most studies using computer simulations have been made on tasks that use only one muscle, usually in the hand. As far as is known, no study has analyzed, either theoretically or experimentally, the overall behavior of the torque variability during plantar flexion. Therefore, this work aims to study the plantar flexion torque variability by means of mathematical models and simulations, comparing the results with those obtained in human experiments carried out locally. A first attempt was made using a Hill-type muscle contraction model activated by the electromyogram obtained from each of the three triceps surae muscles. This approach was not successful in terms of reproducing the torque variability results obtained from humans, although it estimated well the average value of plantar flexion torque. This inability to reproduce the torque variability found in experimental data was probably due to the information loss in the electromyogram of the spike times of motoneurons. In a second approach, the firing of individual motoneuron were obtained from a neuromuscular simulator developed locally, called ReMoto, capable of providing the spike times of all motoneuron models that activate each muscle and the respective muscle force. The latter is generated in the simulator from the forces generated by each motor unit that composes the muscle. However, the ReMoto original version was almost completely parameterized using data from cats and, hence, it was necessary to modify various parameter values (such as motor unit twitchs and action potential amplitudes) and models (such as the recruitment threshold and force saturation) before using the simulator to study torque variability in humans. Besides the second order twitch model already implemented in the original version of the simulator, two other models were implemented in this work. One was a more refined twitch model and the second was a Hill-type model modified to be activated by the ReMoto simulator motoneuron pool. New simulations were run with the new version of the simulator (adapted to human data) and the fittings to the experimental data (torque and electomyogram envelope variability) were good, suggesting that the models in the simulator are a reasonable representation of what occurs in the living human being. (AU)

• FAPESP's process: 10/12934-0 - Computational simulations for the study of torque variability during a plantar flexion.
• Grantee: Renato Naville Watanabe
• Support Opportunities: Scholarships in Brazil - Master