Advanced Biomedical Image Analysis - M. Haidekker (Wiley, 2011) WW

541 Pages·2011·5.55 MB·English

P1: OTA/XYZ
P2: ABC

fm
JWBS035-Haidekker
August 31, 2010
19:57
Printer Name: Yet to Come

ADVANCED BIOMEDICAL
IMAGE ANALYSIS

P1: OTA/XYZ
P2: ABC

fm
JWBS035-Haidekker
August 31, 2010
19:57
Printer Name: Yet to Come

ADVANCED BIOMEDICAL
IMAGE ANALYSIS

MARK A. HAIDEKKER

A JOHN WILEY & SONS, INC., PUBLICATION

P1: OTA/XYZ
P2: ABC

fm
JWBS035-Haidekker
August 31, 2010
19:57
Printer Name: Yet to Come

Copyright C⃝ 2011 by John Wiley & Sons, Inc. All rights reserved.

Published by John Wiley & Sons, Inc., Hoboken, New Jersey.
Published simultaneously in Canada.

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or
by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as
permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior
written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to
the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, [phone],
fax [phone], or on the web at www.copyright.com. Requests to the Publisher for permission
should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken,
NJ 07030, [phone], fax [phone], or online at http://www.wiley.com/go/permission.

Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in
preparing this book, they make no representations or warranties with respect to the accuracy or
completeness of the contents of this book and specifically disclaim any implied warranties of
merchantability or fitness for a particular purpose. No warranty may be created or extended by sales
representatives or written sales materials. The advice and strategies contained herein may not be suitable
for your situation. You should consult with a professional where appropriate. Neither the publisher nor
author shall be liable for any loss of profit or any other commercial damages, including but not limited to
special, incidental, consequential, or other damages.

For general information on our other products and services or for technical support, please contact our
Customer Care Department within the United States at [phone], outside the United States at
[phone] or fax [phone].

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may
not be available in electronic formats. For more information about Wiley products, visit our web site at
www.wiley.com.

Library of Congress Cataloging-in-Publication Data:

Haidekker, Mark A.

Advanced biomedical image analysis / Mark A. Haidekker

Includes bibliographical references and index.

ISBN 978-0-470-62458-6

Printed in Singapore

10
9
8
7
6
5
4
3
2
1

P1: OTA/XYZ
P2: ABC

fm
JWBS035-Haidekker
August 31, 2010
19:57
Printer Name: Yet to Come

CONTENTS

Preface
ix

1
Image Analysis: A Perspective
1

1.1
Main Biomedical Imaging Modalities, 3

1.2
Biomedical Image Analysis, 7

1.3
Current Trends in Biomedical Imaging, 12

1.4
About This Book, 15

References, 17

2
Survey of Fundamental Image Processing Operators
23

2.1
Statistical Image Description, 24

2.2
Brightness and Contrast Manipulation, 28

2.3
Image Enhancement and Restoration, 29

2.4
Intensity-Based Segmentation (Thresholding), 42

2.5
Multidimensional Thresholding, 50

2.6
Image Calculations, 54

2.7
Binary Image Processing, 58

2.8
Biomedical Examples, 63

References, 68

3
Image Processing in the Frequency Domain
70

3.1
The Fourier Transform, 71

3.2
Fourier-Based Filtering, 82

v

P1: OTA/XYZ
P2: ABC

fm
JWBS035-Haidekker
August 31, 2010
19:57
Printer Name: Yet to Come

vi
CONTENTS

3.3
Other Integral Transforms: The Discrete Cosine Transform
and the Hartley Transform, 91

3.4
Biomedical Examples, 94

References, 100

4
The Wavelet Transform and Wavelet-Based Filtering
103

4.1
One-Dimensional Discrete Wavelet Transform, 106

4.2
Two-Dimensional Discrete Wavelet Transform, 112

4.3
Wavelet-Based Filtering, 116

4.4
Comparison of Frequency-Domain Analysis to Wavelet
Analysis, 128

4.5
Biomedical Examples, 130

References, 135

5
Adaptive Filtering
138

5.1
Adaptive Noise Reduction, 139

5.2
Adaptive Filters in the Frequency Domain: Adaptive Wiener
Filters, 155

5.3
Segmentation with Local Adaptive Thresholds and Related
Methods, 157

5.4
Biomedical Examples, 164

References, 170

6
Deformable Models and Active Contours
173

6.1
Two-Dimensional Active Contours (Snakes), 180

6.2
Three-Dimensional Active Contours, 193

6.3
Live-Wire Techniques, 197

6.4
Biomedical Examples, 205

References, 209

7
The Hough Transform
211

7.1
Detecting Lines and Edges with the Hough Transform, 213

7.2
Detection of Circles and Ellipses with the Hough Transform, 219

7.3
Generalized Hough Transform, 223

7.4
Randomized Hough Transform, 226

7.5
Biomedical Examples, 231

References, 234

8
Texture Analysis
236

8.1
Statistical Texture Classification, 238

8.2
Texture Classification with Local Neighborhood Methods, 242

8.3
Frequency-Domain Methods for Texture Classification, 254

P1: OTA/XYZ
P2: ABC

fm
JWBS035-Haidekker
August 31, 2010
19:57
Printer Name: Yet to Come

CONTENTS
vii

8.4
Run Lengths, 257

8.5
Other Classification Methods, 263

8.6
Biomedical Examples, 265

References, 273

9
Shape Analysis
276

9.1
Cluster Labeling, 278

9.2
Spatial-Domain Shape Metrics, 279

9.3
Statistical Moment Invariants, 285

9.4
Chain Codes, 287

9.5
Fourier Descriptors, 291

9.6
Topological Analysis, 295

9.7
Biomedical Examples, 301

References, 307

10
Fractal Approaches to Image Analysis
310

10.1
Self-Similarity and the Fractal Dimension, 311

10.2
Estimation Techniques for the Fractal Dimension in
Binary Images, 319

10.3
Estimation Techniques for the Fractal Dimension in
Gray-Scale Images, 327

10.4
Fractal Dimension in the Frequency Domain, 331

10.5
Local H¨older Exponent, 337

10.6
Biomedical Examples, 340

References, 345

11
Image Registration
350

11.1
Linear Spatial Transformations, 352

11.2
Nonlinear Transformations, 355

11.3
Registration Quality Metrics, 360

11.4
Interpolation Methods for Image Registration, 371

11.5
Biomedical Examples, 379

References, 382

12
Image Storage, Transport, and Compression
386

12.1
Image Archiving, DICOM, and PACS, 389

12.2
Lossless Image Compression, 392

12.3
Lossy Image Compression, 400

12.4
Biomedical Examples, 408

References, 411

13
Image Visualization
413

13.1
Gray-Scale Image Visualization, 413

P1: OTA/XYZ
P2: ABC

fm
JWBS035-Haidekker
August 31, 2010
19:57
Printer Name: Yet to Come

viii
CONTENTS

13.2
Color Representation of Gray-Scale Images, 416

13.3
Contour Lines, 422

13.4
Surface Rendering, 422

13.5
Volume Visualization, 427

13.6
Interactive Three-Dimensional Rendering and Animation, 433

13.7
Biomedical Examples, 434

References, 438

14
Image Analysis and Visualization Software
441

14.1
Image Processing Software: An Overview, 443

14.2
ImageJ, 447

14.3
Examples of Image Processing Programs, 452

14.4
Crystal Image, 456

14.5
OpenDX, 461

14.6
Wavelet-Related Software, 466

14.7
Algorithm Implementation, 466

References, 473

Appendix A: Image Analysis with Crystal Image
475

Appendix B: Software on DVD
497

Index
499

P1: OTA/XYZ
P2: ABC

fm
JWBS035-Haidekker
August 31, 2010
19:57
Printer Name: Yet to Come

PREFACE

Medical imaging is one of the great revolutions in medicine. Traditionally, explo-
rative surgery had to be performed to look inside a patient’s body, even to perform a
diagnosis. Slightly more than a century ago, x-rays were discovered. With it came the
ability to look inside the body without surgery. X-ray imaging was rapidly adopted
in medical centers worldwide. A new medical subject area was created—radiology.
For decades, the radiologist had a basic set of tools, the x-ray tube, a fluoroscope, a
film cassette with an image intensifier screen, and a light box. For decades, progress
was incremental. X-ray tubes were improved, film and intensifier were made more
sensitive, radiation exposure was reduced, and contrast agents were introduced and
improved. It took a second, independent revolution to propel biomedical imaging to
today’s level: the invention of the programmable computer and its subsequent minia-
turization. The availability of powerful digital data processing hardware and newly
developed image processing methods paved the way for new imaging modalities:
computed tomography, magnetic resonance imaging, ultrasound imaging, and func-
tional imaging. These new imaging modalities had in common that computer-based
data processing was required for image formation. Medical imaging experienced a
second wave of rapid progress near the end of the twentieth century when tomography
methods were developed and improved. Tomography means imaging by sections, and
the origin of the word lies in the Greek � ´o�o� for “to cut” and ��´��� for “to write.”
With the availability of imaging modalities that produced three-dimensional recon-
structions of a patient’s body came the need for computerized image processing and
computerized image visualization. The expertise of the radiologist in interpreting an
image played–and still plays—a major role, but more and more tasks could be given
to the computer, and the interpretation of computer-processed images became eas-
ier, more objective, and more accurate. Concurrently, scientists became interested in

ix