Yu, Tung Fai (2014) Musculo-skeletal modelling and parameterisation in vivo. PhD thesis, University of Warwick.

Preview

PDF WRAP_THESIS_Yu_2014.pdf - Submitted Version - Requires a PDF viewer. Download (8Mb) | Preview

Request Changes to record.

Abstract

This thesis describes the development of an anatomically meaningful
musculo-skeletal model of the human arm, incorporating two modified Hill
muscle models representing the elbow flexor and extensor muscles. In vivo
experimental methods to determine parameter values are presented. The stimulus
for this work was to enable the prediction of movement, to support development
of prostheses and orthoses such as Functional Electrical Stimulation (FES).
A key problem in model based movement studies is that the passive
parameter values in the Hill muscle models and the joint had not been
experimentally determined in vivo. The result has been an inability for predictive
models to generate realistic predictions of human movement dynamics.
In the model, movement dynamics of the forearm was described using the
Newton-Euler method, which was validated from analysis of physical pendulum.
Structural identifiability analyses of the muscle models ensured that values for the
model parameters could be uniquely determined from perfect noise free data.
A novel experimental procedure termed the passive movement method is
described, which exclusively parameterised the model’s passive components.
Simulated model dynamics were fitted to measured movements of the freely
swinging forearm under gravity. Model values were obtained on an individual
subject basis. The average muscle model spring and damping constants for four
healthy subjects were 143N/m and 1.73Ns/m respectively.
Separately, the force/length characteristics of the muscles’ active
component, the contractile element (CE), were obtained from measurements of
isometric maximum voluntary contraction (MVC) at different elbow angles. The
results for the five healthy subjects showed good agreement with results reported
in the literature.
A preliminary experiment was performed to predict elbow flexion
movement under FES. An electrical stimulus that generated a specified isometric
elbow flexion moment (10% of MVC) was applied to generate elbow flexion
movement. Simulated FES arm movement was compared with the measured
results. The simulated change in elbow angle did not agree with the measured
data. A major cause for this was believed to be skin movement causing a change
in the current path across the muscle fibres, thus affecting the force generated.
The passive movement method described in this thesis filled an important
chapter to fully parameterise musculo-skeletal models in vivo. Although in the
FES movement experiment, simulated change in elbow angle generated by FES
did not agree with measured data, the shape of the dynamic response in the fitted
simulated movement showed good agreement with the measured FES movement.

Item Type:	Thesis (PhD)
Subjects:	Q Science > QH Natural history > QH301 Biology Q Science > QP Physiology T Technology > TA Engineering (General). Civil engineering (General)
Library of Congress Subject Headings (LCSH):	Musculoskeletal system -- Mechanical properties, Biomechanics, Musculoskeletal system -- Mathematical models, Electric stimulation
Official Date:	March 2014
Dates:	Date Event March 2014 Submitted
Institution:	University of Warwick
Theses Department:	Department of Physics
Thesis Type:	PhD
Publication Status:	Unpublished
Supervisor(s)/Advisor:	Wilson, Adrian
Sponsors:	Engineering and Physical Sciences Research Council (EPSRC)
Extent:	xvii, 223 leaves : illustrations.
Language:	eng

Request changes or add full text files to a record

Repository staff actions (login required)

View Item

Downloads