Medical Image Registration - 1st Edition

Visit the link below to download the full version of this book:

https://medidownload.com/product/medical-image-registration-1st-edition/

Click Download Now

0064_frame_Fm.fm Page 6 Wednesday, May 16, 2001 10:51 AM

Editors

Joseph Hajnal is head of physics and engineering at the Robert Steiner Mag-
netic Resonance Unit, Hammersmith Hospital, Imperial College School of
Medicine and Medical Research Council Clinical Sciences Centre, as well as
leader of a research team funded by Marconi Medical Systems. He trained as a
physicist at Bristol University, England, and earned a Ph.D. in the physics of
electromagnetic waves before working in Australia at Melbourne University
and the Australian National University on interactions between atomic
beams and laser light. In 1990 Dr. Hajnal began research in medical imaging
with a special interest in magnetic resonance imaging. His current research
interests include magnetic resonance data acquisition and processing, image
registration, and data fusion, as well as novel scanner technology. He has
published more than 80 papers in peer-review journals and currently holds
six grants from a variety of funding organizations.

Derek Hill earned a B.Sc. in physics from Imperial College, London in 1987,
a M.Sc. in medical physics in 1989 and a Ph.D. in medical image registration
from the University of London in 1994. He is currently a senior lecturer in the
department of radiological sciences in the medical school of King’s College,
London. Dr. Hill’s research interests include rigid and nonrigid registration,
intraoperative brain deformation, interventional magnetic resonance imag-
ing, MR imaging, and the study of brain shape and connectivity. He has more
than 100 publications in these areas, of which over 30 are peer-review journal
articles. During the last five years, Dr. Hill has held five project grants as prin-
ciple investigator. He has worked extensively with industrial partners on col-
laborative research to develop imaging technology.

David Hawkes has 25 years experience in medical imaging in hospital and

academic environments. He founded the Computational Imaging Science
Group in 1989 at Guy’s Hospital and recently formed the Medical Imaging
Science Interdisciplinary Research Group at the newly merged Guy’s, King’s,
and St. Thomas’ School of Medicine, King’s College, London. Dr. Hawkes
trained as a physicist at Oxford. He earned a master’s degree in radiobiology
at Birmingham and a Ph.D. on parametrization of the x-ray attenuation coef-
ficient for dual energy computed tomography at Surrey. His current research
interests include image matching, data fusion, visualization, shape represen-
tation, surface geometry, and modeling tissue deformation, with applications
in image-guided interventions, augmented reality in surgery, 3D ultrasound,
0064_frame_Fm.fm Page 7 Wednesday, May 16, 2001 10:51 AM

and interventional magnetic resonance imaging. He has been on the scientific

committees of 30 international meetings and is on the editorial boards of five
journals. Dr. Hawkes is currently principal investigator of five U.K. Engineer-
ing and Physical Sciences Research Council (EPSRC) project grants and
manager of three industrially sponsored projects. He has more than 150 pub-
lications in the area of medical imaging.
0064_frame_Fm.fm Page 8 Wednesday, May 16, 2001 10:51 AM

Contributors

Dale L. Bailey Department of Nuclear Medicine, Guy’s, King’s, and

St. Thomas’ School of Medicine, Guy’s Hospital, London, U.K.

Phillipe Batchelor Department of Radiological Sciences, Guy’s, King’s, and

St. Thomas’ School of Medicine, Guy’s Hospital, London, U.K.

David Bell Physics Department, Royal Marsden NHS Trust, Sutton, Surrey,
U.K.

Graeme M. Bydder Robert Steiner MR Unit, MRC Clinical Sciences Centre,

Imperial College School of Medicine, Hammersmith Hospital, London,
U.K.

Matthew J. Clarkson Department of Radiological Sciences, Guy’s, King’s,

and St. Thomas’ School of Medicine, Guy’s Hospital, London, U.K.

D. Louis Collins McConnell Brain Imaging Center, Montreal Neurological

Institute, Montreal, Quebec, Canada

Philip J. Edwards Department of Radiological Sciences, Guy’s, King’s, and

St. Thomas’ School of Medicine, Guy’s Hospital, London, U.K.

Alan C. Evans McConnell Brain Imaging Center, Montreal Neurological

Institute, Montreal, Quebec, Canada

J. Michael Fitzpatrick Department of Electrical Engineering and Computer

Science, Vanderbilt University, Nashville, Tennessee, U.S.A.

Joseph V. Hajnal Robert Steiner MR Unit, MRC Clinical Sciences Centre,

Imperial College School of Medicine, Hammersmith Hospital, London,
U.K.

David J. Hawkes Department of Radiological Sciences, Guy’s, King’s, and

St. Thomas’ School of Medicine, Guy’s Hospital, London, U.K.

Derek L.G. Hill Department of Radiological Sciences, Guy’s, King’s, and

St. Thomas’ School of Medicine, Guy’s Hospital, London, U.K.
0064_frame_Fm.fm Page 9 Wednesday, May 16, 2001 10:51 AM

Jozef Jarosz Neuroimaging, King’s Healthcare Trust, Denmark Hill, London,

U.K.

Mark Jenkinson Oxford Centre for Functional MRI of the Brain, Oxford
University, John Radcliffe Hospital, Oxford, U.K.

Louis Lemieux Research Group, Institute of Neurology, University Col-

lege, London, U.K.

Michael I. Miga Department of Biomedical Engineering, Vanderbilt University,

Nashville, Tennessee, U.S.A.

Angela Oatridge Robert Steiner MR Unit, MRC Clinical Sciences Centre,

Imperial College School of Medicine, Hammersmith Hospital, London,
U.K.

Keith D. Paulsen Thayer School of Engineering, Dartmouth College, Hanover,

New Hampshire, U.S.A.

Tomá s̆ Paus McConnell Brain Imaging Center, Montreal Neurological

Institute, Montreal, Quebec, Canada

Graeme P. Penney Department of Radiological Sciences, Guy’s, King’s, and

St. Thomas’ School of Medicine, Guy’s Hospital, London, U.K.

Uwe Pietrzyk Forschungszentrum Jülich GmbH, Institut für Medizin,

Jülich, Germany

Daniel Rueckert Department of Computing, Imperial College of Science,

Technology, and Medicine, London, U.K.

Stephen M. Smith Oxford Centre for Functional MRI of the Brain, Oxford
University, John Radcliffe Hospital, Oxford, U.K.

Alex P. Zijdenbos McConnell Brain Imaging Center, Montreal Neurologi-

cal Institute, Montreal, Quebec, Canada
0064_frame_Fm.fm Page 10 Wednesday, May 16, 2001 10:51 AM

Contents

1. Introduction ..................................................................................... 1
Joseph V. Hajnal, Derek L.G. Hill, David J. Hawkes

Section I Methodology

2. Registration Methodology: Introduction .................................... 11

David J. Hawkes

3. Registration Methodology: Concepts and Algorithms .............. 39

Derek L.G. Hill and Philipe Batchelor

4. Preparation and Display of Image Data ...................................... 71

Joseph V. Hajnal

5. Correcting for Scanner Errors in CT, MRI, SPECT,

and 3D Ultrasound ........................................................................ 87
Louis Lemieux, Dale L. Bailey, and David Bell

6. Detecting Failure, Assessing Success ........................................ 117

J. Michael Fitzpatrick

Section II Applications of Rigid Body Registration

7. Registration and Subtraction of Serial Magnetic Resonance

Images of the Brain: Image Interpretation and
Clinical Applications .................................................................. 143
Angela Oatridge, Joseph V. Hajnal, and Graeme M. Bydder

8. The Role of Registration in Functional Magnetic

Resonance Imaging ..................................................................... 183
Mark Jenkinson and Stephen Smith

9. Registration of MRI and PET Images

for Clinical Applications ............................................................ 199
Uwe Pietrzyk
0064_frame_Fm.fm Page 11 Wednesday, May 16, 2001 10:51 AM

10. Registration of MR and CT Images

for Clinical Applications ............................................................ 217
Derek L.G. Hill and Josef Jarosz

11. Image Registration in Nuclear Medicine .................................. 233

Dale L. Bailey

12. Guiding Therapeutic Procedures ............................................... 253

Philip J. Edwards, David J. Hawkes, Graeme P. Penney,
and Matthew J. Clarkson

Section III Techniques and Applications

of Nonrigid Registration

13. Nonrigid Registration: Concepts, Algorithms,

and Applications ......................................................................... 281
Daniel Rueckert

14. Use of Registration for Cohort Studies ..................................... 303

D. Louis Collins, Alex P. Zijdenbos, Tomá s̆ Pau s̆ ,
and Alan C. Evans

15. Biomechanical Modeling for Image Registration:

Applications in Image-Guided Neurosurgery .......................... 331
Keith D. Paulsen and Michael I. Miga

16. View of the Future ....................................................................... 363

Joseph V. Hajnal, Derek L.G. Hill, and David J. Hawkes

Index ..................................................................................................... 369

0064_frame_C01.fm Page 1 Wednesday, May 16, 2001 9:54 AM

1
Introduction

Joseph V. Hajnal, Derek L.G. Hill, and David. J. Hawkes

CONTENTS
1.1 Preface..............................................................................................................1
1.2 Historical Background...................................................................................4
1.3 Overview of the Book....................................................................................6
References ................................................................................................. 7

1.1 Preface
Image registration is the process of aligning images so that corresponding fea-
tures can easily be related. The term is also used to mean aligning images with
a computer model or aligning features in an image with locations in physical
space. The images might be acquired with different sensors (e.g., sensitive to
different parts of the electromagnetic spectrum) or the same sensor at different
times. Image registration has applications in many fields; the one that is
addressed in this book is medical imaging. This encompasses a wide range of
image usage, but the main emphasis is on radiological imaging.
The past 25 years have seen remarkable developments in medical imaging
technology. Universities and industry have made huge investments in
inventing and developing the technology needed to acquire images from
multiple imaging modalities. Medical images are increasingly widely used in
healthcare and biomedical research; a very wide range of imaging modalities
is now available. X-ray computed tomography (CT) images are sensitive to
tissue density and atomic composition, and the x-ray attenuation coefficient
and magnetic resonance imaging (MR) images are related to proton density,
relaxation times, flow, and other parameters. The introduction of contrast
agents provides information on the patency and function of tubular struc-
tures such as blood vessels, the bile duct, and the bowel, as well as the state

0-8493-0064-9/01/$0.00+$.50
© 2001 by CRC Press LLC 1
0064_frame_C01.fm Page 2 Tuesday, May 22, 2001 3:12 PM

2 Medical Image Registration

of the blood-brain barrier. In nuclear medicine, radiopharmaceuticals intro-

duced into the body allow delineation of functioning tissue and measurement
of metabolic and pathophysiological processes. Ultrasound detects subtle
changes in acoustic impedance at tissue boundaries and diffraction patterns
in different tissues, providing discrimination of different tissue types. Doppler
ultrasound provides images of flowing blood. Endoscopy and surgical
microscopy provide images of visible surfaces deep within the body. These
and other imaging technologies now provide rich sources of data on the phys-
ical properties and biological function of tissues at spatial resolutions from 5
mm for nuclear medicine down to 1.0 or 0.5 mm for MR and CT, and 20 to 100
 m for optical systems. Each successive generation of image acquisition sys-
tem has acquired images faster, with higher resolution and improved image
quality, and together these have been harnessed for great clinical benefit.
Since the mid 1980s medical image registration has evolved from being
perceived as a rather minor precursor to some medical imaging applications
to a significant subdiscipline in itself. Entire sessions are devoted to the topic
1,2
in major medical imaging conferences, and workshops have been held on
3
the subject. Image registration has also become one of the more successful
areas of image processing, with fully automated algorithms available in a
number of applications.
Why has registration become so important? Medical imaging is about
establishing shape, structure, size, and spatial relationships of anatomical
structures within the patient, together with spatial information about function
and any pathology or other abnormality. Establishing the correspondence of
spatial information in medical images and equivalent structures in the body
is fundamental to medical image interpretation and analysis.
In many clinical scenarios, images from several modalities may be acquired
and the diagnostician’s task is to mentally combine or ‘‘fuse’’ this information to
draw useful clinical conclusions. This generally requires mental compensation
for changes in subject position. Image registration aligns the images and so
establishes correspondence between different features seen on different imag-
ing modalities, allows monitoring of subtle changes in size or intensity over
time or across a population, and establishes correspondence between images
and physical space in image guided interventions. Registration of an atlas or
computer model aids in the delineation of anatomical and pathological struc-
tures in medical images and is an important precursor to detailed analysis.
It is now common for patients to be imaged multiple times, either by
repeated imaging with a single modality, or by imaging with different modal-
ities. It is also common for patients to be imaged dynamically, that is, to have
sequences of images acquired, often at many frames per second. The ever
increasing amount of image data acquired makes it more and more desirable
to relate one image to another to assist in extracting relevant clinical informa-
tion. Image registration can help in this task: intermodality registration
enables the combination of complementary information from different
modalities, and intramodality registration enables accurate comparisons
between images from the same modality.
0064_frame_C01.fm Page 3 Wednesday, May 16, 2001 9:54 AM

Introduction 3

International concern about escalating healthcare costs drives develop-

ment of methods that make the best possible use of medical images and, once
again, image registration can help. However, medical image registration does
not just enable better use of images that would be acquired anyway, it also
opens up new applications for medical images. These include serial imaging
to monitor subtle changes due to disease progression or treatment; perfusion
or other functional studies when the subject cannot be relied upon to remain
in a fixed position during the dynamic acquisition; and image-guided inter-
ventions, in which images acquired prior to the intervention are registered
with the treatment device, enabling the surgeon or interventionalist to use
the preintervention images to guide his or her work. Image registration has
also become a valuable technique for biomedical research, especially in neu-
roscience, where imaging studies are making substantial contributions to our
understanding of the way the brain works. Image registration can be used to
align multiple images from the same individual (intrasubject registration) and
to compare images acquired from different subjects (intersubject registration).
All the images that we wish to register or manipulate in any other way on
a computer must be available in digital form. This means that most medical
images are made up of a rectangular array of small square or rectangular ele-
ments called pixels (an abbreviation of picture elements); each pixel has an
associated image intensity value. This array provides the coordinate system
of the image, and an element in the image can be accessed by its two-dimen-
sional position within this array. A typical CT slice will be formed of 512  512
pixels, and each will correspond to an element of the cut through the patient
2
of about 0.5  0.5 mm . This dimension determines the limiting spatial reso-
lution of the image. 2D images are often stacked together to form a 3D vol-
ume, and many images are now acquired directly as 3D volumes. Each pixel
will now correspond to a small volume element of tissue, or voxel. If the slice
spacing in high resolution CT is, say, 1.5 mm, the voxel size will be 0.5  0.5 
3
1.5 mm . The number stored in each voxel—the voxel image intensity—will
be some average of a physical attribute measured over this volume. In MR,
3
voxels are generally slightly larger, typically 1  1  3–5 mm in size.
Radiologists have traditionally reviewed medical images by viewing them
as film transparencies on a back-illuminated light box. Most imaging modal-
ities involve some digital manipulation and computation, and so these
images are now often stored in digital form and displayed on a workstation.
Digital storage greatly facilitates further digital manipulation, such as regis-
tration of the images and fusion of the information from the different modal-
ities. Subjective judgments of the relative size, shape, and spatial
relationships of visible structures and physiology inferred from intensity dis-
tributions are used for developing a diagnosis, planning therapy, and moni-
toring disease progression or response to therapy. A key process when
interpreting these images together is the explicit or implicit establishment of
correspondence between different points in the images. The spatial integrity
of the images can allow very accurate correspondence to be determined.
Once correspondence has been established in a verifiable way, multiple
0064_frame_C01.fm Page 4 Wednesday, May 16, 2001 9:54 AM

4 Medical Image Registration

images can be interpreted as single unified data sets and conclusions drawn
with increased confidence. Creating this single unified data set is the process
of “fusion.” In many instances, new information becomes available that
could not have been deduced from inspection of individual images in loose
association with one another.

1.2 Historical Background

Although this is the first book dedicated to medical image registration, it is
not a new topic. Image registration has been widely used for many years in
x-ray angiography. It is common to acquire x-rays before and after injection
of intravascular contrast and then subtract these images in order to visualize
the blood vessels in isolation. This technique almost invariably uses digital
systems now, but optical subtraction using photographic methods has been
extremely effective. A negative of the radiograph taken before the arrival of
the contrast material, the “mask,” was positioned on a light box over the
radiograph taken after the arrival of contrast and an additional film was
taken. If the patient moved between the acquisition of the precontrast mask
image and the image with contrast, then the subtracted image would contain
edge artifacts. Translating and rotating the films prior to optical subtraction
greatly reduced these artifacts. Photographic subtraction was also used with
MR to correct for patient motion and generate images showing where gado-
4
linium contrast had been taken up.
Image-guided surgery was the first application of medical image registration.
Indeed, the very first radiograph acquired for this purpose was reported to
have been in Birmingham, U.K., only two weeks after the discovery of x-rays
was published in December 1895. A patient had broken a needle in her hand.
A radiograph was taken and the casualty officer aligned the plate with the
5
hand in order to successfully guide his scalpel to removal of the needle.
Other early examples included battlefield surgery for removal of shrapnel by
registering a calibrated pair of x-ray films to the patient so that the x-ray
could guide the surgeon precisely to the target in 3D. The stereotactic frame
was proposed for image-guided neurosurgery as a means of localizing target
structures with respect to anatomical features identified in the patient’s
6,7
radiographs and or a standard atlas. The frame is rigidly fixed to the skull
and defines a coordinate system for both imaging and treatment. Stereotactic
neurosurgery became more widely used when the technology was comput-
8,9
erized and combined with CT, and then with multiple preoperative imag-
10
ing modalities. Stereotactic neurosurgery can only be used for a small
proportion of neurosurgical procedures, because the frame has to be attached
to the patient prior to imaging and left on until surgery, and the presence of
the frame restricts the types of surgery that can be performed, often just to
biopsy and electrode implantation. These problems were overcome with the
0064_frame_C01.fm Page 5 Wednesday, May 16, 2001 9:54 AM

Introduction 5

development of more sophisticated registration techniques leading to the

11
introduction of frameless stereotaxy in the mid 1980s, though it was another
decade before frameless stereotactic systems obtained the regulatory
approval necessary for widespread use in health care.
In image-guided interventions, correspondence is established between
image and the physical space of the patient during the intervention. Estab-
lishing this correspondence allows the image to be used to guide, direct, and
monitor therapy, akin to providing a 3D map for navigation, with the aim of
making the intervention more accurate, safer, and less invasive for the patient.
In the last few years, image registration techniques have entered routine clinical
use in image-guided neurosurgery systems and computer-assisted orthopedic
surgery. Systems incorporating image registration are sold by a number of
manufacturers.
Stereotactic frames can also be used for intermodality image registration, but
their use is restricted to highly invasive surgical procedures because of the
need for rigid fixation to the skull. Various relocatable frames were proposed
to avoid this invasiveness, but beginning in the mid 1980s, registration algo-
rithms were devised that were “retrospective,” that is, did not require special
measures to be taken during image acquisition in order for registration to be pos-
12–14
sible. Various approaches were introduced in the mid 1980s. These tech-
niques were devised to make it possible to combine images of the same patient
taken with different modalities and they required substantial user interaction.
Another major step forward in image registration came in the first half of
the 1990s with the development of retrospective registration algorithms that
15,16
were fully or virtually fully automated for both intramodality and
17–21
intermodality registration. A significant breakthrough in the mid 1990s
was the development of image alignment measures and registration algo-
rithms based on entropy and, in particular, mutual information—measures
22
first derived from the information theory developed by Shannon in 1948.
Recently the focus of research in medical image registration has returned to
intramodality rather than intermodality registration, and to extending regis-
tration algorithms to handle the more complicated transformations needed
to model soft tissue deformation and intersubject registration.
Detailed atlases or computer models of anatomy are becoming available, in
4
particular from high resolution sources such as the Visible Human datasets.
The Montreal Brain Atlas has been generated by averaging images of the
brain across a population. Establishing spatial correspondence between these
atlases and an individual’s images allows for easier interpretation and, in
particular, enables computer assistance in delineation of anatomical struc-
tures of interest.
Rapid advances in the power of computer technology and in the perfor-
mance of new registration algorithms and displays mean that image manip-
ulation deemed impractical or far too computationally expensive only a few
years ago can now be undertaken on the PCs available on most people’s
desks. In our experience, initial work on voxel similarity measures was prov-
ing successful in the laboratory in 1994 but took an hour or more to complete.
0064_frame_C01.fm Page 6 Wednesday, May 16, 2001 9:54 AM

6 Medical Image Registration

Between then and late 2000, desktop workstations have increased in speed by
nearly two orders of magnitude while their cost has been reduced by an order
of magnitude, so the same calculations can now be achieved in a few minutes
on desktop PCs. The first algorithms for nonrigid registration required large
amounts of interaction or were prohibitively slow, but more recent work has
resulted in highly automated algorithms that can run on standard hardware
in minutes rather than days.

1.3 Overview of the Book

This book is divided into three sections. Section I, Methodology, introduces
the wide variety of techniques used for medical image registration. The con-
cepts behind registration techniques are introduced for a general audience in
Chapter 2, and, for those who wish to understand the underlying algorithms
or implement registration methods themselves. Behind the techniques are
described in more detail in Chapter 3. The necessary additional considerations
behind acquiring and preparing data for image registration are discussed in
Chapter 4, and correcting errors in the scanners is addressed in Chapter 5. In
the final chapter in this section, Chapter 6, the essential problem of detecting
when the algorithms have failed is discussed, and how accurately the algo-
rithms have aligned the images is assessed.
Section II describes the relatively mature applications of image registration
in which the images can be aligned by global translation and rotation
alone—so-called rigid-body registration. The chapters in this section are
written by researchers with many years of experience in the applications of
serial MR registration (Chapter 7), functional MRI (Chapter 8), registration of
PET and MRI (Chapter 9), registration of MRI and CT (Chapter 10), registra-
tion in nuclear medicine (Chapter 11), and the use of registration in guided
therapeutic procedures (Chapter 12).
Section III focuses on the less mature but rapidly developing field of nonrigid
image registration. The topic is introduced in Chapter 13, and alternative
approaches are reviewed and examples given of one approach for a variety
of applications. The problem of combining images from multiple subjects in
cohort studies is examined in detail in Chapter 14, and the rather different
approach of using biomechanical models to achieve registration is considered
in Chapter 15. Section III tells less of a finished story than Sections I and II, due
to the rapid evolution of techniques in this area. The goal of this section is
to give insight into some of the applications that drive nonrigid registration
and the different approaches being devised.
An observation as this book was prepared (late 2000) was that there are
literally hundreds of papers in the literature describing medical image regis-
tration methods and applications, and yet this technology is currently used
very little in clinical practice, with the exception of image-guided surgery.
0064_frame_C01.fm Page 7 Wednesday, May 16, 2001 9:54 AM

Introduction 7

Why is this? Maybe the clinical applications are not relevant for day-to-day
patient management. This does not appear to be a sustainable view either
from the literature or from the view of centers using this technology. It may
be that image registration generally forms only one part of a complete image
analysis application, and other components, notably image segmentation and
labeling, are still not sufficiently robust or automated for routine clinical use. In
many applications, this is undoubtedly the case. Perhaps some of the problems
with segmentation will be solved by nonrigid atlas registration (see Chapter 14).
Another important factor is that to achieve widespread use, the clinical com-
munity and the medical imaging industry that supports it must embrace this
new technology more effectively. This will require investment in order to: 1)
ensure that technical validation and clinical evaluation are effective and
timely; 2) proceed rapidly down the path of standardization and integration
of information sources in healthcare so that innovative products from small
companies can be incorporated earlier and more cheaply into the healthcare
environment; 3) ensure that image registration becomes automatic or virtu-
ally automatic so that it is robust and transparent to the user; and, finally, 4)
ensure that the clinical community, and its scientific and technical support
staffs, are made fully aware of the power of this new technology and that
medical practice evolves to take full advantage of it.
We hope that this book will go some way in encouraging goals 1, 2, and 3
above by contributing to goal 4.
We are very grateful to the contributing authors for sharing our vision that
the time is right for a book on medical image registration, for the effort they
have put into their chapters, and, in particular, for responding to our sugges-
tions for making the book a more coherent whole.

References
1. K.M. Hanson, Medical Imaging 2000: Image Processing. SPIE 3979, 2000.
2. C. Taylor and A. Colchester, Medical Image Computing and Computer Assisted
Intervention—MICCAI’99. Lecture Notes in Computer Science 1679, Heidelberg:
Springer-Verlag, 1999.
3. F. Pernus, S. Kovacic, H.S. Stiehl, and M.A. Viergever, Biomedical Image Registra-
tion Proc WBIR’99. Slovenia Pattern Recognition Society, Ljubljana, 1999.
4. G.A.S. Lloyd, P.G. Barker, and P.D. Phelps, Subtraction gadolinium-enhanced
magnetic resonance for head and neck imaging, Br. J. Radiol., vol. 66, pp. 12–16,
1993.
5. S. Webb, The Physics of Medical Imaging. Institute of Physics Publishing, Bristol,
1988.
6. R.H. Clarke and V. Horsley, On a method of investigating the deep ganglia and
tracts of the central nervous system (cerebellum), Br. Med. J., vol. 2, pp. 1799–
1800, 1906.
7. E.A. Spiegel, H.T. Wycis, M. Marks, and A.J. Lee, Stereotaxic apparatus for
operations on the human brain, Science, vol. 106, pp. 349–350, 1947.
0064_frame_C01.fm Page 8 Wednesday, May 16, 2001 9:54 AM

8 Medical Image Registration

8. R.A. Brown, T.S. Roberts, and A.G. Osborn, Stereotaxic frame and computer
software for CT-directed neurosurgical localization, Invest. Radiol., vol. 15,
pp. 308–312, 1980.
9. L. Leksell and B. Jernberg, Stereotaxis and tomography: a technical note, Acta
Neurochir., vol. 52, pp. 1–7, 1980.
10. T.M. Peters, J.A. Clark, A. Olivier, E.P. Marchand, G. Mawko, M. Dieumegarde,
L. Muresan, and R. Ethier, Integrated stereotaxic imaging with CT, MR imaging,
and digital subtraction angiography, Radiology, vol. 161, pp. 821–826, 1986.
11. D.W. Roberts, J.W. Strohbehn, J.F. Hatch, W. Murray, and H. Kettenberger, A
frameless stereotaxic integration of computerized tomographic imaging and the
operating microscope, J. Neurosurg., vol. 65, pp. 545–549, 1986.
12. G.T.Y. Chen, M. Kessler, and S. Pitluck, Structure transfer between sets of three-
dimensional medical imaging data, in Computer Graphics 1985, pp. 171–175,
Dallas: National Computer Graphics Association, 1985.
13. G.Q. Maguire, Jr., M.E. Noz, E.M. Lee, and J.H. Schimpf, Correlation methods
for tomographic images using two- and three-dimensional techniques, in Infor-
mation Processing in Medical Imaging 1985, S. L. Bacharach, Ed., pp. 266–279,
Dordrecht: Martinus Nijhoff Publishers, 1986.
14. C.-T. Chen, C.A. Pelizzari, G.T.Y. Chen, M.D. Cooper, and D.N. Levin, Image
analysis of PET data with the aid of CT and MR images, in Information Processing
in Medical Imaging 1987, C. N. de Graaf and M. A. Viergever, Eds., pp. 601–611,
New York: Plenum Press, 1988.
15. R.P. Woods, S.R. Cherry, and J.C. Mazziotta, Rapid automated algorithm for align-
ing and reslicing PET images, J. Comput. Assist. Tomogr., vol. 16, pp. 620–633, 1992.
16. J.V. Hajnal, N. Saeed, E.J. Soar, A. Oatridge, I.R. Young, and G.M. Bydder, A
registration and interpolation procedure for subvoxel matching of serially ac-
quired MR images, J. Comput. Assist. Tomogr., vol. 19, pp. 289–296, 1995.
17. R.P. Woods, J.C. Mazziotta, and S.R. Cherry, MRI-PET registration with automat-
ed algorithm, J. Comput. Assist. Tomogr., vol. 17, pp. 536–546, 1993.
18. P. Viola and W. Wells, Alignment by maximization of mutual information, in
Proc. 5th Int. Conf. Computer Vision, pp. 16–23, 1995.
19. A. Collignon, F. Maes, D. Delaere, D. Vandermeulen, P. Suetens, and G. Marchal,
Automated multi-modality image registration based on information theory, in
Information Processing in Medical Imaging 1995, Y. Bizais, C. Barillot, and R. Di
Paola, Eds., pp. 263–274, Dordrecht: Kluwer Academic, 1995.
20. C. Studholme, D.L.G. Hill, and D.J. Hawkes, Automated 3D registration of MR
and CT images of the head, Med. Image Anal., vol. 1, pp. 163–175, 1996.
21. C. Studholme, D.L.G. Hill, and D.J. Hawkes, Automated 3D registration of MR
and PET brain images by multi-resolution optimisation of voxel similarity mea-
sures, Med. Phys., vol. 24, pp. 25–35, 1997.
22. C.E. Shannon, The mathematical theory of communication (parts 1 and 2), Bell
Syst. Tech. J., vol. 27, pp. 379–423 and 623–656, 1948. Reprint available from
http://www.lucent.com.
0064_frame_C02.fm Page 9 Wednesday, May 16, 2001 10:04 AM

Section I

Methodology
0064_frame_C02.fm Page 10 Wednesday, May 16, 2001 10:04 AM