Computational Intelligence for Movement Sciences. Neural Networks, Support Vector Machines, and Other Emerging Techniques

411 Pages·2016·4.6 MB·English

by  Begg R.

Computational Intelligence
for Movement Sciences:
Neural Networks and Other
Emerging Techniques
Rezaul Begg, Victoria University, Australia
Marimuthu Palaniswami, The University of Melbourne, Australia
IDEA GROUP PUBLISHING
Hershey • London • Melbourne • Singapore
Acquisitions Editor: Michelle Potter
Development Editor: Kristin Roth
Senior Managing Editor: Amanda Appicello
Managing Editor: Jennifer Neidig
Copy Editor: Bernard J. Kieklak, Jr.
Typesetter: Sharon Berger
Cover Design: Lisa Tosheff
Printed at: Integrated Book Technology
Published in the United States of America by
Idea Group Publishing (an imprint of Idea Group Inc.)
701 E. Chocolate Avenue, Suite 200
Hershey PA 17033
Tel: [phone]
Fax: [phone]
E-mail: [email protected]
Web site: http://www.idea-group.com
and in the United Kingdom by
Idea Group Publishing (an imprint of Idea Group Inc.)
3 Henrietta Street
Covent Garden
London WC2E 8LU
Tel: [phone]
Fax: [phone]
Web site: http://www.eurospanonline.com
Copyright © 2006 by Idea Group Inc. All rights reserved. No part of this book may be reproduced,
stored or distributed in any form or by any means, electronic or mechanical, including photocopying,
without written permission from the publisher.
Product or company names used in this book are for identification purposes only. Inclusion of the
names of the products or companies does not indicate a claim of ownership by IGI of the trademark
or registered trademark.
Library of Congress Cataloging-in-Publication Data
Computational intelligence for movement sciences : neural networks, support vector machines, and
other emerging techniques / Rezaul Begg and Marimuthu Palaniswami, editors.
p. ; cm.
Includes bibliographical references and index.
Summary: "This book provides information regarding state-of-the-art research outcomes and cutting-
edge technology on various aspects of the human movement"--Provided by publisher.
ISBN 1-59140-836-9 (hardcover) -- ISBN 1-59140-837-7 (softcover)
1. Musculoskeletal system--Physiology. 2. Human locomotion. 3. Artificial intelligence. 4. Neural
networks (Computer science)
[DNLM: 1. Movement--physiology. 2. Artificial Intelligence. 3. Biomechanics. 4. Models,
Biological. 5. Neural Networks (Computer) WE 103 C7383 2006] I. Begg, Rezaul. II. Palaniswami,
Marimuthu.
QP301.C588 2006
612.7'0285--dc22
2005032065
British Cataloguing in Publication Data
A Cataloguing in Publication record for this book is available from the British Library.
All work contributed to this book is new, previously-unpublished material. The views expressed in this
book are those of the authors, but not necessarily of the publisher.
IGP Forthcoming Titles in the
Computational Intelligence and
Its Applications Series
Advances in Applied Artificial Intelligence
(March 2006 release)
John Fulcher
ISBN: 1-59140-827-X
Paperback ISBN: 1-59140-828-8
eISBN: 1-59140-829-6
Computational Economics: A Perspective from Computational
Intelligence
(November 2005 release)
Shu-Heng Chen, Lakhmi Jain, and Chung-Ching Tai
ISBN: 1-59140-649-8
Paperback ISBN: 1-59140-650-1
eISBN: 1-59140-651-X
Computational Intelligence for Movement Sciences: Neural Networks,
Support Vector Machines and Other Emerging Techniques
(February 2006 release)
Rezaul Begg and Marimuthu Palaniswami
ISBN: 1-59140-836-9
Paperback ISBN: 1-59140-837-7
eISBN: 1-59140-838-5
An Imitation-Based Approach to Modeling Homogenous
Agents Societies
(July 2006 release)
Goran Trajkovski
ISBN: 1-59140-839-3
Paperback ISBN: 1-59140-840-7
eISBN: 1-59140-841-5
It’s Easy to Order! Visit www.idea-group.com!
717/533-8845 x10
Mon-Fri 8:30 am-5:00 pm (est) or fax 24 hours a day 717/533-8661
IDEA GROUP PUBLISHING
Hershey • London • Melbourne • Singapore
Excellent additions to your library!
Computational Intelligence
for Movement Sciences:
Neural Networks and Other
Emerging Techniques
Table of Contents
Preface .........................................................................................................................vii
Section I: Methods and Tools for Movement Analysis
Chapter I
Overview of Movement Analysis and Gait Features.......................................................1
Russell Best, Victoria University, Australia
Rezaul Begg, Victoria University, Australia
Chapter II
Inertial Sensing in Biomechanics: A Survey of Computational Techniques Bridging
Motion Analysis and Personal Navigation.................................................................. 70
Angelo M. Sabatini, ARTS Lab, Scuola Superiore Sant’Anna, Italy
Chapter III
Monitoring Human Movement with Body-Fixed Sensors and its Clinical
Applications...............................................................................................................101
Kamiar Aminian, Ecole Polytechnique Fédérale de Lausanne (EPFL),
Switzerland
Chapter IV
Computational Intelligence Techniques...................................................................139
Bharat Sundaram, The University of Melbourne, Australia
Marimuthu Palaniswami, The University of Melbourne, Australia
Alistair Shilton, The University of Melbourne, Australia
Rezaul Begg, Victoria University, Australia
Section II: Advances in Gait Analysis and Modelling
Chapter V
Modelling of Some Aspects of Skilled Locomotor Behaviour Using Artificial
Neural Networks.......................................................................................................172
Stephen D. Prentice, University of Waterloo, Canada
Aftab E. Patla, University of Waterloo, Canada
Chapter VI
Visualisation of Clinical Gait Data Using a Self-Organising Artificial Neural
Network..................................................................................................................... 197
Gabor J. Barton, Liverpool John Moores University, UK
Chapter VII
Neural Network Models for Estimation of Balance Control, Detection of
Imbalance, and Estimation of Falls Risk...................................................................217
Michael E. Hahn, Montana State University, USA
Arthur M. Farley, University of Oregon, USA
Li-Shan Chou, University of Oregon, USA
Chapter VIII
Recognition of Gait Patterns Using Support Vector Machines ............................... 243
Rezaul Begg, Victoria University, Australia
Marimuthu Palaniswami, The University of Melbourne, Australia
Section III: Applications in Rehabilitation and Sport
Chapter IX
Control of Man-Machine FES Systems..................................................................... 263
Rahman Davoodi, University of Southern California, USA
Gerald E. Loeb, University of Southern California, USA
Chapter X
Evolutionary Methods for Analysis of Human Movement.........................................281
Rahman Davoodi, University of Southern California, USA
Gerald E. Loeb, University of Southern California, USA
Chapter XI
Dynamic Pattern Recognition in Sport by Means of Artificial Neural Networks ... 299
Jürgen Perl, Johannes Gutenberg University, Germany
Peter Dauscher, Johannes Gutenberg University, Germany
Section IV: Computational Modelling for Predicting Movement Forces
Chapter XII
Estimation of Muscle Forces About the Ankle During Gait in Healthy and
Neurologically Impaired Subjects ............................................................................320
Daniel N. Bassett, University of Delaware, USA
Joseph D. Gardinier, University of Delaware, USA
Kurt T. Manal, University of Delaware, USA
Thomas S. Buchanan, University of Delaware, USA
Chapter XIII
Computational Modelling in Shoulder Biomechanics..............................................348
David C. Ackland, The University of Melbourne, Australia
Cheryl J. Goodwin, The University of Texas at Austin, USA
Marcus G. Pandy, The University of Melbourne, Australia
About the Authors.....................................................................................................385
Index ........................................................................................................................391
vii
Preface
Studies into human movement sciences have been usually undertaken from an
interdisciplinary perspective. Individuals and groups who are involved in movement
science research come from a number of diverse backgrounds, including: biomechan-
ics, biomedical engineering, health science, exercise science, sports science, computer
science, clinical science, physiotherapy, prosthetics and orthotics, to name a few. Re-
search and development in movement sciences are progressing quite rapidly. The main
aims of these advances are to gain a better understanding of the normal and abnormal
human movement characteristics, and also to develop new and innovative ways of
combating the rising health care costs around the globe. Analysis of gait and other
human movements has proved very useful in revealing many useful insights into the
recognition and assessment of movement abnormalities. In recent times, gait analysis
is taken almost as a routine procedure in aiding many diagnostic and rehabilitative
procedures. Common application examples include: the design of a rehabilitation pro-
gram to assist the disabled, the planning and assessment of surgical outcomes, the
recognition of gaits due to falls-risk in the elderly and also for the improvement of
sports techniques and performance.
Computational intelligence (CI) encompasses approaches primarily based on arti-
ficial neural networks, fuzzy logic rules, evolutionary algorithms and support vector
machines. These methods have been applied to solve many complex and diverse prob-
lems. Recent years have seen many new developments in CI techniques and conse-
quently this has led to an exponential increase in the number of applications in a variety
of areas including engineering, finance, social and biomedical. In particular, CI tech-
niques are increasingly being used in biomedical and human movement areas because
of the complexity of the biological systems as well as the limitations of the existing
quantitative techniques in modelling. The contents of this book cover a wide range of
relevant applications in human movement sciences written by leading researchers and
academicians in the area. Altogether, the book has 13 chapters organized into the fol-
lowing four sections:
• Section I: Methods and Tools for Movement Analysis
• Section II: Advances in Gait Analysis and Modelling
• Section III: Applications in Rehabilitation and Sport
• Section IV: Computational Modelling for Predicting Movement Forces
viii
Section I has four chapters that are aimed to provide the readers with a compre-
hensive overview of the various approaches and techniques for analysing human move-
ments. The first chapter provides a comprehensive overview of the traditional move-
ment analysis techniques, potential errors and noise contents in the captured data and
some of the major data processing and analysis techniques. Feature extraction is an
important process in movement analysis tasks and forms an integral part of a comput-
erized data analysis procedure. An extensive overview of the techniques that could be
used to derive features from the processed data is presented. In addition to the labora-
tory-based measurement techniques, alternative approaches using body-mounted in-
ertial sensing have received considerable attention in recent years, especially for am-
bulatory monitoring of human motion during various activities. One major advantage of
body-mounted inertial sensing in the biomedical domain is its capability to objectively
determine a person’s level of functional ability in independent living. The next two
chapters focus on such techniques and their applications including: sensor configura-
tions and reviews, computational techniques for automatic recognition of activity, quan-
titative analysis of motor performance, and personal navigation systems. Significant
clinical applications (e.g., in orthopedics, Parkinson disease, aging, etc.) as well as
potential applications in nanotechnology, materials sciences, and advanced mobile and
ubiquitous body movement are discussed. The final chapter in this section presents a
brief description of the major computational intelligence techniques for pattern recog-
nition and modelling tasks that often appear in biomedical, health and human move-
ment research. These include techniques such as artificial neural networks, fuzzy logic
rules, evolutionary approaches, support vector machines, and also approaches that
combine two or more techniques (hybrid).
Section II includes four chapters that focus on applications of neural networks
and other CI techniques for analysing, modelling and visualizing gait data. Neural
networks have been predominantly used in most gait recognition, classification and
modelling tasks. The first chapter in this section examines the use of artificial neural
networks to model and probe the control of walking movements. Chapter VI describes
self-organising artificial neural networks and their use to reduce the complexity of gait
data and to improve visualisation of large amounts of complex data in a two-dimen-
sional map.
The next two chapters focus on recognition of gait changes due to aging and
falling behaviour. With significant increase in the number of aging population around
the world, and falls in older adults being a major public health issue, researchers are
looking for ways to reduce the falls risk incidence in older adults and to improve aging
health care. Chapter VII looks into gait pattern changes in the elderly and demonstrates
the effectiveness of artificial neural network modelling in mapping gait measurements
onto reduced function in the balance control system. Chapter VIII describes an auto-
mated gait pattern recognition system using gait classifier based on support vector
machines.
Section III explores applications of CI techniques in rehabilitation and sport-
related areas. Chapter IX provides a brief description of different methods for the
control of man-machine FES systems and talks about a clinical FES system to demon-
strate the successful application of these strategies. Specially, application involving
FES systems for the restoration of movement to the paralyzed limbs in spinal cord
injury patients is discussed. In Chapter X, evolutionary methods are introduced —
these are intelligent systems that can further our understanding of human movement
ix
and help us devise new treatments. The evolutionary methods resemble the biological
processes that led to the development of natural human movement, and they can solve
optimization and learning problems that cannot be solved with existing methods. Evo-
lutionary computation is well suited to parallel processing that can reduce the compu-
tation time significantly and can be applied extensively to treating human movement
disorders. In this chapter, the authors discuss applications of evolutionary methods to
explore feasible movement patterns that can be used to prescribe movement therapies
to improve the existing functions or design FES control systems to restore the lost
movement.
Chapter XI provides applications of self-organising artificial neural networks for
analysing sport games, motor activities or rehabilitation. Such processes are often
characterized by a complex structure. Measurements considering them may produce a
huge amount of data. This chapter presents neural network approaches and examples
of application in the field of sport.
Section IV focuses on computational modelling approaches to estimate internal
joint forces during movement. Specifically, Chapter XII deals with biomechanical model
of the forces about the ankle joint applicable to both unimpaired and neurologically
impaired subjects. An EMG-driven hybrid forward and inverse dynamics model of the
ankle is employed and optimization procedures are discussed that are used to tune the
physiological parameters for the model for each subject.
The final chapter describes computational modelling of the shoulder complex.
Following a brief background in anatomy and biomechanics of the shoulder complex
the authors present a review of the essential functions of the shoulder and the impor-
tant features of practical shoulder models. Computational modelling techniques, and
also in vivo and in vitro methods for verifying computational models are briefly dis-
cussed and a summary of the emerging trends are presented to indicate the clinical
impact of computational modeling.
The book contains information regarding state-of-the art research outcomes and
cutting-edge technology from the leading scientists and researchers working on vari-
ous aspects of the human movement. It is hoped that the book will be of enormous help
to a broad spectrum of readerships including researchers, professionals, lecturers and
graduate students from a wide range of disciplines. It is our belief that the ideas pre-
sented in this book will trigger further works and research efforts in this rapidly expand-
ing multidisciplinary area.
Rezaul Begg, Victoria University, Australia
Marimuthu Palaniswami, The University of Melbourne, Australia
Editors