51955

Download the full version of the ebook at
https://ebookultra.com
Fundamentals of Medical Imaging 2nd Edition

Paul Suetens
https://ebookultra.com/download/fundamentals-of-
medical-imaging-2nd-edition-paul-suetens/
Explore and download more ebook at https://ebookultra.com

Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.
Fundamentals of Digital Imaging H. J. Trussell
https://ebookultra.com/download/fundamentals-of-digital-imaging-h-j-
trussell/
ebookultra.com
Imaging Dopamine 1st Edition Paul Cumming
https://ebookultra.com/download/imaging-dopamine-1st-edition-paul-
cumming/
ebookultra.com
Image Radiation Biology of Medical Imaging 1st Edition

Edition David Greenwood
https://ebookultra.com/download/image-radiation-biology-of-medical-
imaging-1st-edition-edition-david-greenwood/
ebookultra.com
Fundamentals of Musculoskeletal Imaging 3rd Edition

Contemporary Perspectives in Rehabilitation Lynn Mckinnis
https://ebookultra.com/download/fundamentals-of-musculoskeletal-
imaging-3rd-edition-contemporary-perspectives-in-rehabilitation-lynn-
mckinnis/
ebookultra.com
Medical Infrared Imaging Principles and Practices 1st
Edition Diakides
https://ebookultra.com/download/medical-infrared-imaging-principles-
and-practices-1st-edition-diakides/
ebookultra.com
Imaging systems for medical diagnostics fundamentals

technical solutions and applications for systems applying
ionization radiation nuclear magnetic resonance and
ultrasound Oppelt
https://ebookultra.com/download/imaging-systems-for-medical-
diagnostics-fundamentals-technical-solutions-and-applications-for-
systems-applying-ionization-radiation-nuclear-magnetic-resonance-and-
ultrasound-oppelt/
ebookultra.com
Ultrasound Imaging and Therapy Imaging in Medical

Diagnosis and Therapy 1st Edition Aaron Fenster (Editor)
https://ebookultra.com/download/ultrasound-imaging-and-therapy-
imaging-in-medical-diagnosis-and-therapy-1st-edition-aaron-fenster-
editor/
ebookultra.com
Fundamentals of Mathematical Analysis 1st Edition Paul J.

Sally Jr
https://ebookultra.com/download/fundamentals-of-mathematical-
analysis-1st-edition-paul-j-sally-jr/
ebookultra.com
Medical Imaging Based on Magnetic Fields and Ultrasounds

1st Edition Fanet
https://ebookultra.com/download/medical-imaging-based-on-magnetic-
fields-and-ultrasounds-1st-edition-fanet/
ebookultra.com
Fundamentals of Medical Imaging 2nd Edition Paul
Suetens Digital Instant Download
Author(s): Paul Suetens
ISBN(s): 9780511596407, 0511596405
Edition: 2
File Details: PDF, 18.68 MB
Year: 2009
Language: english
Fundamentals of
Medical Imaging
Second Edition
Paul Suetens
Katholieke Universiteit Leuven
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore,
São Paulo, Delhi, Dubai, Tokyo
Cambridge University Press

The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521519151
© First edition © Cambridge University Press 2002
Second edition © P. Suetens 2009
This publication is in copyright. Subject to statutory exception and to the
provision of relevant collective licensing agreements, no reproduction of any part
may take place without the written permission of Cambridge University Press.
First published in print format 2009
ISBN-13 978-0-511-59640-7 eBook (NetLibrary)
ISBN-13 978-0-521-51915-1 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy

of urls for external or third-party internet websites referred to in this publication,
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
Every effort has been made in preparing this publication to provide accurate and
up-to-date information which is in accord with accepted standards and practice at
the time of publication. Although case histories are drawn from actual cases, every
effort has been made to disguise the identities of the individuals involved.
Nevertheless, the authors, editors and publishers can make no
warranties that the information contained herein is totally free from error, not least
because clinical standards are constantly changing through research and
regulation. The authors, editors and publishers therefore disclaim all liability for
direct or consequential damages resulting from the use of material contained in
this publication. Readers are strongly advised to pay careful attention to
information provided by the manufacturer of any drugs or equipment that they
plan to use.
Contents
Preface page vii Image quality 90

Acknowledgments ix Equipment 95
Clinical use 98
Biologic effects and safety 102
1 Introduction to digital image Future expectations 104
processing 1
Digital images 1 5 Nuclear medicine imaging 105
Image quality 2 Introduction 105
Basic image operations 4 Radionuclides 105
Interaction of γ-photons and particles
2 Radiography 14 with matter 108
Introduction 14
Data acquisition 108
X-rays 14
Imaging 111
Interaction with matter 16
Image quality 116
X-ray detectors 17
Equipment 117
Dual-energy imaging 21
Clinical use 122
Image quality 23
Equipment 24
Future expectations 125
Clinical use 25
Future expectations 32 6 Ultrasound imaging 128
Introduction 128
3 X-ray computed tomography 33 Physics of acoustic waves 128
Introduction 33 Generation and detection of
X-ray detectors in CT 34 ultrasound 137
Imaging 35 Gray scale imaging 138
Cardiac CT 46 Doppler imaging 141
Dual-energy CT 48 Image quality 145
Image quality 49 Equipment 149
Equipment 53 Clinical use 152
Clinical use 58 Biologic effects and safety 155
Biologic effects and safety 59 Future expectations 156
Future expectations 63
7 Medical image analysis 159
4 Magnetic resonance imaging 64 Introduction 159
Introduction 64 Manual analysis 160
Physics of the transmitted signal 64 Automated analysis 160
Interaction with tissue 68 Computational strategies for automated
Signal detection and detector 71 medical image analysis 163
Imaging 72 Pixel classification 166
Contents
Geometric model matching using a Virtual reality 205

transformation matrix 170 User interaction 207
Flexible geometric model matching 175 Intraoperative navigation 208
Validation 186 Augmented reality 214
Future expectations 189 Future expectations 218
8 Visualization for diagnosis

and therapy 190 Appendix A: Linear system theory 219
Introduction 190 Appendix B: Exercises 232
2D visualization 192 Bibliography 246
3D rendering 192 Index 248
vi
Preface
This book explains the applied mathematical and the stronger student to read around the subject and
physical principles of medical imaging and image pro- also makes the book a useful purchase for those going
cessing. It gives a complete survey, accompanied by on to do research.
more than 300 illustrations in color, of how medical In Chapter 1, an introduction to digital image pro-
images are obtained and how they can be used for cessing is given. It summarizes the jargon used by
diagnosis, therapy, and surgery. the digital image community, the components defin-
It has been written principally as a course text on ing image quality, and basic image operations used
medical imaging intended for graduate and final-year to process digital images. The theory of linear sys-
undergraduate students with a background in physics, tems, described in Chapter 2 of the first edition, has
mathematics, or engineering. However, I have made been moved to an appendix. It is too high-level for
an effort to make the textbook readable for biomedical the medical reader and a significant part of the engi-
scientists and medical practitioners as well by delet- neering readers of the previous edition considered it
ing unnecessary mathematical details, without giving as redundant. However, many students in physics or
up the depth needed for physicists and engineers. engineering are not familiar with linear system theory
Mathematical proofs are highlighted in separate para- and will welcome this appendix.
graphs and can be skipped without hampering a fluent Chapters 2–6 explain how medical images are
reading of the text. obtained. The most important imaging modalities
Although a large proportion of the book covers today are discussed: radiography, computed tomogra-
the physical principles of imaging modalities, the phy, magnetic resonance imaging, nuclear medicine
emphasis is always on how the image is computed. imaging, and ultrasonic imaging. Each chapter
Equipment design, clinical considerations, and diag- includes (1) a short history of the imaging modality,
nosis are treated in less detail. Premature techniques (2) the theory of the physics of the signal and its inter-
or topics under investigation have been omitted. action with tissue, (3) the image formation or recon-
Presently, books on medical imaging fall into two struction process, (4) a discussion of the image quality,
groups, neither of which is suitable for this read- (5) the different types of equipment in use today, (6)
ership. The first group is the larger and comprises examples of the clinical use of the modality, (7) a brief
books directed primarily at the less numerate pro- description of the biologic effects and safety issues, and
fessions such as physicians, surgeons, and radiologic (8) some future expectations. The imaging modalities
technicians. These books cover the physics and mathe- have made an impressive evolution in a short time with
matics of all the major medical imaging modalities, but respect to quality, size and applicability. This part of
mostly in a superficial way. They do not allow any real the book provides up-to-date information about these
understanding of these imaging modalities. The sec- systems.
ond group comprises books suitable for professional Chapters 7 and 8 deal with image analysis and
medical physicists or researchers with expertise in the visualization for diagnosis, therapy and surgery once
field. Although these books have a numerate approach, images are available. Medical images can, for example,
they tend to cover the topics too deeply for the be analyzed to obtain quantitative data, or they can
beginner and to have a narrower scope than this book. be displayed in three dimensions and actively used
The text reflects what I teach in class, but there is to guide a surgical intervention. Most courses sepa-
somewhat more material than I can cover in a module rate the imaging theory from the postprocessing, but
of 30 contact hours. This means that there is scope for I strongly believe that they should be taken together
Preface
because the topics are integrated. The interest in clin- patient record, and PACS (picture archiving and
ical practice today goes beyond the production and communication systems). However, this focus would
diagnosis of two-dimensional images, and the objec- put the emphasis on informatics, such as databases,
tive then is to calculate quantitative information or networking, internet technology and information
to use the images during patient treatment. The field security, which is not the purpose of this book.
of medical image analysis is in full progress and has New also in this second edition is an appendix with
become more mature during the last decade. This exercises. By solving these exercises the student can
evolution has been taken into account in this second test his or her insight into the matter of this book.
edition. The chapter on image-guided interventions of Furthermore an ancillary website (www.cambridge.
the first edition has been rewritten with a new focus. org/suetens) with three-dimensional animations has
The emphasis now is on three-dimensional image been produced which contains answers to the
visualization, not only to guide interventions, but also exercises.
for diagnostic purposes. In the bibliography, references to untreated top-
Medical imaging and image processing can also ics can be found as well as more specialized works on
be approached from the perspective of information a particular subdomain and some other generic text-
and communication and the supporting technology, books related to the field of medical imaging and image
such as hospital information systems, the electronic processing.
viii
Acknowledgments
My colleagues of the Medical Imaging Research Cen- and Dirk Vandermeulen (image analysis), Dirk
ter have directly and indirectly contributed to the Loeckx (exercises), Christophe Deroose, Steven
production of this book. This facility is quite a Dymarkowski, Guy Marchal and Luc Mortelmans
unique place where engineers, physicists, computer (clinical use). They provided me with pieces of text,
scientists, and medical doctors collaborate in an relevant clinical images and important literature; and
interdisciplinary team. It has a central location in I had indispensable discussions with them concerning
the University Hospital Leuven and is surrounded content and structure.
by the clinical departments of radiology, nuclear A final reading was done by Kristof Baete,
medicine, cardiology, and radiotherapy. Research Bart De Dobbelaer, An Elen, Johannes Keuster-
is focused on clinically relevant questions. This mans, Florence Kremer, Catherine Lemmens, Ronald
then explains the emphasis in this book, which Peeters, Janaki Rangarajan, Annemie Ribbens, Lies-
is on recent imaging technology used in clinical bet Roose, Kristien Smans, Dirk Smeets and Kevin
practice. Suetens.
The following colleagues and former colleagues I would like to express my gratitude to Walter
contributed to the first edition of the book: Bruno Coudyzer for his assistance in collecting radiological
De Man, Jan D’hooge, Frederik Maes, Johan Michiels, data. Special thanks are due to Dominique Delaere,
Johan Nuyts, Johan Van Cleynenbreugel and Koen the information manager of the Medical Imaging
Vande Velde. Research Center, who assisted me for both this and
This second edition came about with substan- the previous edition with the figures, illustrations
tial input from Hilde Bosmans (radiography), Bruno and animations, consistency checking, and the web-
De Man (computed tomography), Stefan Sunaert pages associated with this textbook. Thanks to his
(magnetic resonance imaging), Johan Nuyts (nuclear degree in biomedical engineering, he also made several
medicine), Jan D’hooge (ultrasound), Frederik Maes improvements to the content.
Chapter
1 Introduction to digital image processing
Digital images cyan, magenta, and yellow. Cyan is the color of a mate-
Visible light is essentially electromagnetic radiation rial, seen in white light, that absorbs red but reflects
with wavelengths between 400 and 700 nm. Each green and blue, and can thus be obtained by additive
wavelength corresponds to a different color. On the mixing of equal amounts of green and blue light. Sim-
other hand, a particular color does not necessarily ilarly, magenta is the result of the absorption of green
correspond to a single wavelength. Purple light, for light and consists of equal amounts of red and blue
example, is a combination of red and blue light. In light, and yellow is the result of the absorption of blue
general, a color is characterized by a spectrum of and consists of equal amounts of red and green light.
different wavelengths. Therefore, subtractive mixing of cyan and magenta
The human retina contains three types of photore- gives blue, subtractive mixing of cyan and yellow
ceptor cone cells that transform the incident light with gives green, and subtractive mixing of yellow and
different color filters. Because there are three types of magenta gives red. Subtractive mixing of yellow, cyan,
cone receptors, three numbers are necessary and suf- and magenta produces black (only absorption and no
ficient to describe any perceptible color. Hence, it is reflection) (see Figure 1.1(b)).
possible to produce an arbitrary color by superimpos- Note that equal distances in physical intensity are
ing appropriate amounts of three primary colors, each not perceived as equal distances in brightness. Inten-
with its specific spectral curve. In an additive color sity levels must be spaced logarithmically, rather than
reproduction system, such as a color monitor, these linearly, to achieve equal steps in perceived bright-
three primaries are red, green, and blue light. The ness. Hue refers to the dominant wavelength in the
color is then specified by the amounts of red, green, spectrum, and represents the different colors. Satu-
and blue. Equal amounts of red, green, and blue give ration describes the amount of white light present in
white (see Figure 1.1(a)). Ideal white light has a flat the spectrum. If no white light is present, the satura-
spectrum in which all wavelengths are present. In prac- tion is 100%. Saturation distinguishes colorful tones
tice, white light sources approximate this property. from pastel tones at the same hue. In the color cone
In a subtractive color reproduction system, such as of Figure 1.2, equal distances between colors by no
printing or painting, these three primaries typically are
Hue
Saturation
Brightness
(a) (b)
Figure 1.1 Color mixing: (a) additive color mixing, (b) subtractive
color mixing. Figure 1.2 Hue, brightness, and saturation.
Chapter 1: Introduction to digital image processing
means correspond to equal perceptual differences. The gray values (12 bpp). The problem with too many gray
Commission Internationale de l’Eclairage (CIE) has values, however, is that small differences in brightness
defined perceptually more uniform color spaces like cannot be perceived on the display. This problem can
L ∗ u ∗ v ∗ and L ∗ a ∗ b ∗ . A discussion of pros and cons be overcome for example by expanding a small gray
of different color spaces is beyond the scope of this value interval into a larger one by using a suitable gray
textbook. value transformation, as discussed on p. 4 below.
While chromatic light needs three descriptors or In the process of digital imaging, the continuous
numbers to characterize it, achromatic light, as pro- looking world has to be captured onto the finite num-
duced by a black-and-white monitor, has only one ber of pixels of the image grid. The conversion from a
descriptor, its brightness or gray value. Achromatic continuous function to a discrete function, retaining
light is light with a saturation of 0%. It contains only only the values at the grid points, is called sampling
white light. and is discussed in detail in Appendix A, p. 228.
Given a set of possible gray levels or colors and Much information about an image is contained in
a (rectangular) grid, a digital image attributes a gray its histogram. The histogram h of an image is a prob-
value (i.e., brightness) or a color (i.e., hue, saturation ability distribution on the set of possible gray levels.
and brightness) to each of the grid points or pixels. In The probability of a gray value v is given by its relative
a digital image, the gray levels are integers. Although frequency in the image, that is,
brightness values are continuous in real life, in a digital
image we have only a limited number of gray levels number of pixels having gray value v
h(v) = . (1.1)
at our disposal. The conversion from analog sam- total number of pixels
ples to discrete-valued samples is called quantization.
Figure 1.3 shows the same image using two different
quantizations. When too few gray values are used, con- Image quality
touring appears. The image is reduced to an artificial The resolution of a digital image is sometimes wrongly
looking height map. How many gray values are needed defined as the linear pixel density (expressed in dots
to produce a continuous looking image? Assume that per inch). This is, however, only an upper bound for
n + 1 gray values are displayed with corresponding the resolution. Resolution is also determined by the
physical intensities I0 , I1 , . . . , In . I0 is the lowest attain- imaging process. The more blurring, the lower is the
able intensity and In the maximum intensity. The ratio resolution. Factors that contribute to the unsharpness
In /I0 is called the dynamic range. The human eye can- of an image are (1) the characteristics of the imag-
not distinguish subsequent intensities Ij and Ij+1 if ing system, such as the focal spot and the amount of
they differ less than 1%, i.e., if Ij+1 ≤ 1.01 Ij . In detector blur, (2) the scene characteristics and geom-
that case In ≤ 1.01n I0 and n ≥ log1.01 (In /I0 ). For etry, such as the shape of the subject, its position and
a dynamic range of 100 the required number of gray motion, and (3) the viewing conditions.
values is 463 and a dynamic range of 1000 requires Resolution can be defined as follows. When imag-
694 different gray values for a continuous looking ing a very small, bright point on a dark background,
brightness. Most digital medical images today use 4096 this dot will normally not appear as sharp in the image
Figure 1.3 The same image quantized

with (a) 8 bpp and (b) 4 bpp.
2 (a) (b)
Figure 1.4 (a) Sharp bright spot on a

dark background. (b) Typical image of (a).
The smoothed blob is called the point
spread function (PSF) of the imaging
system.
(a) (b)
as it actually is. It will be smoothed, and the obtained |OTF|

1
blob is called the point spread function (PSF) (see
Figure 1.4). An indicative measure of the resolution is
the full width at half maximum (FWHM) of the point
spread function. When two such blobs are placed at
this distance or shorter from each other, they will no lp/mm
longer be distinguishable as two separate objects. If the (a) (b)
resolution is the same in all directions, the line spread
Figure 1.5 (a) Point spread function (PSF). (b) Corresponding
function (LSF), i.e., the actual image of a thin line, may modulation transfer function (MTF). The MTF is the amplitude of
be more practical than the PSF. the optical transfer function (OTF), which is the Fourier transform
Instead of using the PSF or LSF it is also pos- (FT) of the PSF.
sible to use the optical transfer function (OTF) (see
Figure 1.5). The OTF expresses the relative amplitude As explained in Appendix A, the OTF is the Fourier
and phase shift of a sinusoidal target as a function of transform (FT) of the PSF or LSF.
frequency. The modulation transfer function (MTF) Contrast is the difference in intensity of adja-
is the amplitude (i.e. MTF = |OTF|) and the phase cent regions of the image. More accurately, it is the
transfer function (PTF) is the phase component of the amplitude of the Fourier transform of the image as a
OTF. For small amplitudes the lines may no longer function of spatial frequency. Using the Fourier trans-
be distinguishable. An indication of the resolution is form, the image is unraveled in sinusoidal patterns
the number of line pairs per millimeter (lp/mm) at a with corresponding amplitude and these amplitudes 3
specified small amplitude (e.g., 10%). represent the contrast at different spatial frequencies.
The contrast is defined by (1) the imaging process, physically impossible for the human eye to distinguish
such as the source intensity and the absorption effi- all these gray values at once in a single image. Conse-
ciency or sensitivity of the capturing device, (2) the quently, not all the diagnostic information encoded in
scene characteristics, such as the physical properties, the image may be perceived. Meaningful details must
size and shape of the object, and the use of contrast have a sufficiently high contrast to allow the clinician
agents, and (3) the viewing conditions, such as the to detect them easily.
room illumination and display equipment. Because The larger the number of gray values in the image,
the OTF drops off for larger frequencies, the con- the more important this issue becomes, as lower
trast of very small objects will be influenced by the contrast features may become available in the image
resolution as well. data. Therefore, image enhancement will not become
A third quality factor is image noise. The emission less important as the quality of digital image captur-
and detection of light and all other types of electromag- ing systems improves. On the contrary, it will gain
netic waves are stochastic processes. Because of the importance.
statistical nature of imaging, noise is always present.
It is the random component in the image. If the noise
level is high compared with the image intensity of an Gray level transformations
object, the meaningful information is lost in the noise. Given a digital image I that attributes a gray value
An important measure, obtained from signal theory, (i.e., brightness) to each of the pixels (i, j), a gray level
is therefore the signal-to-noise ratio (SNR or S/N). In transformation is a function g that transforms each
the terminology of images this is the contrast-to-noise gray level I (i, j) to another value I (i, j) independent
ratio (CNR). Both contrast and noise are frequency of the position (i, j). Hence, for all pixels (i, j)
dependent. An estimate of the noise can be obtained
by making a flat-field image, i.e., an image without I (i, j) = g (I (i, j)). (1.2)
an object between the source and the detector. The
noise amplitude as a function of spatial frequency can
be calculated from the square root of the so-called In practice, g is an increasing function. Instead of
Wiener spectrum, which is the Fourier transform of transforming gray values it is also possible to oper-
the autocorrelation of a flat-field image. ate on color (i.e., hue, saturation and brightness). In
Artifacts are artificial image features such as dust or that case three of these transformations are needed to
scratches in photographs. Examples in medical images transform colors to colors.
are metal streak artifacts in computed tomography Note that, in this textbook, the notation I is used
(CT) images and geometric distortions in magnetic not only for the physical intensity but also for the gray
resonance (MR) images. Artifacts may also be intro- value (or color), which are usually not identical. The
duced by digital image processing, such as edge gray value can represent brightness (logarithm of the
enhancement. Because artifacts may hamper the diag- intensity, see p. 1), relative signal intensity or any other
nosis or yield incorrect measurements, it is important derived quantity. Nevertheless the terms intensity and
to avoid them or at least understand their origin. intensity image are loosely used as synonyms for gray
In the following chapters, image resolution, noise, value and gray value image.
contrast, and artifacts will be discussed for each of the If pixel (i1 , j1 ) appears brighter than pixel (i2 , j2 ) in
imaging modalities. the original image, this relation holds after the gray
level transformation. The main use of such a gray
level transformation is to increase the contrast in some
Basic image operations regions of the image. The price to be paid is a decreased
In this section a number of basic mathematical opera- contrast in other parts of the image. Indeed, in a region
tions on images are described. They can be employed containing pixels with gray values in the range where
for image enhancement, analysis and visualization. the slope of g is larger than 1, the difference between
The aim of medical image enhancement is to allow these gray values increases. In regions with gray val-
the clinician to perceive better all the relevant diag- ues in the range with slope smaller than 1, gray values
nostic information present in the image. In digital come closer together and different values may even
4 radiography for example, 12-bit images with 4096 pos- become identical after the transformation. Figure 1.6
sible gray levels are available. As discussed above, it is shows an example of such a transformation.
white
4500
4000
3500
3000
2500
2000
1500
1000
500
black 0
0 500 1000 1500 2000 2500 3000 3500 4000 4500
black white
(a) (b) (c)
Figure 1.6 A gray level transformation that increases the contrast in dark areas and decreases the contrast in bright regions. It can be used
when the clinically relevant information is situated in the dark areas, such as the lungs in this example: (b) the original image, (c) the
transformed image.
4000 4000
Figure 1.7 (a) Window/leveling with
l = 1500, w = 1000. (b) Thresholding with
3000 3000 tr = 1000.
2000 2000
1000 1000
0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000
(a) (b)
A particular and popular transformation is the These operations can be very useful for images with a
window/level operation (see Figure 1.7(a)). In this oper- bimodal histogram (see Figure 1.8).
ation, an interval or window is selected, determined by
the window center or level l, and the window width w.
Explicitly
Multi-image operations
A simple operation is adding or subtracting images in
 w

 0 for t < l − a pixelwise way. For two images I1 and I2 , the sum I+

 2
 M w w w and the difference I− are defined as
gl,w (t ) = t −l + for l − ≤ t ≤ l +
w
 2 2 2
I+ (i, j) = I1 (i, j) + I2 (i, j)

 w (1.5)
M for t > l + ,
2 I− (i, j) = I1 (i, j) − I2 (i, j). (1.6)
(1.3)
where M is the maximal available gray value. Con- If these operations yield values outside the available
trast outside the window is lost completely, whereas gray value range, the resulting image can be brought
the portion of the range lying inside the window is back into that range by a linear transformation. The
stretched to the complete gray value range. average of n images is defined as
An even simpler operation is thresholding
1
(Figure 1.7(b)). Here all gray levels up to a certain Iav (i, j) = (I1 (i, j) + · · · + In (i, j)). (1.7)
threshold tr are set to zero, and all gray levels above n
the threshold equal the maximal gray value Averaging can be useful to decrease the noise in a
gtr (t ) = 0 for t ≤ tr sequence of images of a motionless object (Figure 1.9).
(1.4) The random noise averages out, whereas the object 5
gtr (t ) = M for t > tr.
remains unchanged (if the images match perfectly).
Figure 1.8 Original CT image (a) with

Lung Bone bimodal histogram (b). (c, d) Result of
window/leveling using a bone window
(dashed line in (b)) and lung window
(solid line in (b)), respectively.
white
histogram
black
(a) (b)
(c) (d)
Figure 1.9 (a) Magnetic resonance

image of a slice through the brain. This
image was obtained with a T1 -weighted
EPI sequence (see p. 82) and therefore
has a low SNR. (b) To increase the SNR, 16
subsequent images of the same slice
were acquired and averaged. (Courtesy of
Professor S. Sunaert, Department of
Radiology.)
(a) (b)
This method can also be used for color images by aver- made, one without a contrast agent and another with
aging the different channels independently like gray contrast agent injected in the blood vessels. Subtrac-
level images. Subtraction can be used to get rid of the tion of these two images yields a pure image of the
background in two similar images. For example, in blood vessels because the subtraction deletes the other
6 blood vessel imaging (angiography), two images are anatomical features. Figure 1.10 shows an example.
(a) (b) (c)
Figure 1.10 (a) Radiographic image after injection of a contrast agent. (b) Mask image, that is, the same exposure before contrast injection.
(c) Subtraction of (a) and (b), followed by contrast enhancement. (Courtesy of Professor G. Wilms, Department of Radiology.)
Geometric operations and rotation as special cases. Affine transforma-

It is often necessary to perform elementary geometric tions preserve parallelism of lines but generally not
operations on an image, such as scaling (zooming), lengths and angles. Angles and lengths are preserved
translation, rotation, and shear. Examples are the reg- by orthogonal transformations (e.g., rotations and
istration of images (see p. 173) and image-to-patient translations)
registration for image-guided surgery (see p. 211). A     
spatial or geometric transformation assigns each point x r11 r12 tx x
(x, y) to a new location (x , y ) = S(x, y). The most orthogonal y  = r21 r22 ty  y  , (1.9)
common two-dimensional (2D) transformations can 1 0 0 1 1
be written using homogeneous coordinates:
     where the 2 × 2 matrix R = rr11 21
r12
r22 is subject to the
x sx 0 0 x constraint R R = 1. T
scaling  y  =  0 s y 0  y  A pixel (x, y) = (i, j) of image I (i, j) will be
1 0 0 1 1 mapped onto (x , y ) and x and y are usually no
    
x 1 0 tx x longer integer values. To obtain a new image I (i , j )
translation y  = 0 1 ty  y  on a pixel grid, interpolation is used. For each (i , j )
1 0 0 1 1 the gray value I (i , j ) is then calculated by simple (e.g.,
     bilinear) interpolation between the gray values of the
x 1 ux 0 x
y  = uy 1 0 y  pixels of I lying closest to the inverse transformation
shear
of (i , j ), i.e., S −1 (i , j ).
1 0 0 1 1
     Today the majority of medical images are three
x cos θ −sin θ 0 x dimensional (3D). The above matrices can easily
rotation y  =  sin θ cos θ 0 y  be extended to three dimensions. For example, the
1 0 0 1 1 general affine 3D transformation can be written as
    
x a11 a12 tx x     
general affine y  = a21 a22 ty  y  . x a11 a12 a13 tx x
1 0 0 1 1 y  a21 a22 a23 ty  y 
general affine     
z  = a31 a32 a33 tz  z  .
(1.8)
1 0 0 0 1 1
Composition of two such transformations amounts to (1.10)
multiplying the corresponding matrices.
A general affine 2D transformation depends on six While most medical images are three dimensional, 7
parameters and includes scaling, translation, shear, interventional imaging is often still two dimensional.
To map the 3D image data onto the 2D image a pro- In practice, the flipped kernel h defined as h(i, j) =
jective transformation is needed. Assuming a pinhole f (−i, −j) is usually used. Hence, Eq. (1.13) can be
camera, such as an X-ray tube, any 3D point (x, y, z) rewritten as
is mapped onto its 2D projection point (u, v) by the
projective matrix (more details on p. 216) L(I )(i, j) = f ∗ I (i, j)
 
    x = f (k, l)I (i − k, j − l)
u fx κx u0 0  
 v  = κy fy v0 0 y  k,l
z 
w 0 0 1 0 = h(k, l)I (i + k, j + l)
1
   u  k,l
u w
v  =  v  . = h • I (i, j), (1.14)
(1.11)
w
1 1
where h •I is the cross-correlation of h and I . If the filter
Using homogeneous coordinates the above geo- is symmetric, which is often the case, cross-correlation
metric transformations can all be represented by and convolution are identical.
matrices. In some cases, however, it might be neces- A cross-correlation of an image I (i, j) with a ker-
sary to use more flexible transformations. For exam- nel h has the following physical meaning. The kernel
ple, the comparison of images at different moments, h is used as an image template or mask that is shifted
such as in follow-up studies, may be hampered due across the image. For every image pixel (i, j), the tem-
to patient movement, organ deformations, e.g., dif- plate pixel h(0, 0), which typically lies in the center of
ferences in bladder and rectum filling, or breathing. the mask, is superimposed onto this pixel (i, j), and
Another example is the geometric distortion of mag- the values of the template and image that correspond
netic resonance images resulting from undesired devi- to the same positions are multiplied. Next, all these
ations of the magnetic field (see p. 92). Geometric values are summed. A cross-correlation emphasizes
transformations are discussed further in Chapter 7. patterns in the image similar to the template.
Often local filters with only a few pixels in diam-
eter are used. A simple example is the 3 × 3 mask
Filters with values 1/9 at each position (Figure 1.11). This
Linear filters filter performs an averaging on the image, making it
From linear system theory (see Eq. (A.22)), we know smoother and removing some noise. The filter gives
that an image I (i, j) can be written as follows: the same weight to the center pixel as to its neighbors.
A softer way of smoothing the image is to give a high
I (i, j) = I (k, l)δ(i − k, j − l). (1.12) weight to the center pixel and less weight to pixels fur-
k,l
ther away from the central pixel. A suitable filter for
For a linear shift-invariant transformation L (see also
Eq. (A.31)),

L(I )(i, j) = I (k, l)L(δ)(i − k, j − l)
k,l
1/9 1/9 1/9
= I (k, l)f (i − k, j − l)
k,l 1/9 1/9 1/9
1/9 1/9 1/9
= f (k, l)I (i − k, j − l)
k,l
= f ∗ I (i, j), (1.13)

(a) (b)
where f is called the kernel or filter, and the linear
8 transformation on the digital image I is the discrete Figure 1.11 (a) 3 × 3 averaging filter. (b) The filter as floating image
convolution with its kernel f = L(δ). template or mask.
Figure 1.12 (a) Radiography of the skull.

(b) Low-pass filtered image with a
Gaussian filter (20 × 20 pixels, σ = 15).
(c) High-pass filtered image obtained by
subtracting (b) from (a).
(a) (b) (c)
this operation is the discretized Gaussian function Hence, the derivative is approximated by a convolu-
tion with a filter that is the sampled derivative of some
1
e(−r /2σ ) differentiable function f ( r ). This procedure can now
2 2
g (
r) = r = (i, j). (1.15)
2π σ 2
be used further to approximate the gradient and the
Small values are put to zero in order to produce a Laplacian of a digital image:
local filter. The Fourier transform of the Gaussian is
∇I = ∇f ∗ I
again Gaussian. In the Fourier domain, convolution
(1.17)
with a filter becomes multiplication. Taking this into ∇ 2I = ∇ 2f ∗ I ,
account, it is clear that a Gaussian filter attenuates
the high frequencies in the image. These averaging where it is understood that we use the discrete convo-
filters are therefore also called low-pass filters. In lution. If f is a Gaussian g , the following differential
contrast, filters that emphasize high frequencies are convolution operators are obtained:
called high-pass filters. A high-pass filter can be con-
structed simply from a low-pass one by subtracting 1
∇g (
r) = − g (
r ) · r
the low-pass filter g from the identity filter δ. A σ2
high-pass filter enhances small-scale variations in the (1.18)
2 1
image. It extracts edges and fine textures. An exam- r ) = 4 (r 2 − 2σ 2 ) · g (
∇ g ( r ).
ple of low-pass and high-pass filtering is shown in σ
Figure 1.12. For σ = 0.5, this procedure yields approximately the
Other types of linear filters are differential opera- following 3 × 3 filters (see Figure 1.13):
tors such as the gradient and the Laplacian. However,
these operations are not defined on discrete images. 0.01 0.08 0.01
Because derivatives are defined on differentiable func- Gaussian 0.08 0.64 0.08
tions, the computation is performed by first fitting a 0.01 0.08 0.01
differentiable function through the discrete data set.
This can be obtained by convolving the discrete image
0.05 0 −0.05
with a continuous function f . The derivative of this ∂
0.34 0 −0.34
result is evaluated at the points (i, j) of the origi- ∂x
0.05 0 −0.05
nal sampling grid. For the 1D partial derivative this
(1.19)
sequence of operations can be written as follows:
0.05 0.34 0.05
   ∂
0 0 0
∂ ∂ ∂y
I (i, j) ≈   I (k, l)f (x − k, y − l) −0.05 −0.34 −0.05
∂x ∂x
k,l x=i,y=j
  0.3 0.7 0.3
∂f
= (i − k, j − l)I (k, l) . (1.16) ∇2 0.7 −4 0.7
∂x 0.3 0.7 0.3 9
k,l
z Figure 1.13 (a) A Gaussian function.

(b) Derivative of the Gaussian in the x-direction.
(c) Derivative of the Gaussian in the y-direction.
(d) Laplacian of the Gaussian.
z
x x
y y
(a) (b)
z
x
y
z
x
y
(c) (d)
Note that integration of a Gaussian over the whole Note that, if we compute the convolution of an image
spatial domain must be 1, and for the gradient and with a filter, it is necessary to extend the image at
Laplacian this must be 0. To satisfy this condition, the its boundaries because pixels lying outside the image
numbers in the templates above, which are spatially will be addressed by the convolution algorithm. This
limited, were adapted. is best done in a smooth way, for example by repeat-
The Laplacian of a Gaussian is sometimes approx- ing the boundary pixels. If not, artifacts appear at the
imated by a difference of Gaussians with different boundaries after the convolution.
values of σ . This can be derived from Eq. (1.18). As an application of linear filtering, let us discuss
Rewriting it as edge enhancement using unsharp masking . Figure 1.14
2 shows an example. As already mentioned, a low-pass
r 2 4 filter g can be used to split an image I into two parts: a
+ g (
r ) − 2 g (
r) (1.20)
σ 4 σ 2 σ smooth part g ∗ I , and the remaining high-frequency
shows us that the second term is proportional to the part I − g ∗ I containing the edges in the image or
original Gaussian g , while the first term drops off more image details. Hence
slowly because of the r 2 and acts as if it were a Gaussian
I = g ∗ I + (I − g ∗ I ). (1.22)
with a larger value of σ (the 2/σ 2 added to the r 2 /σ 4
makes it a monotonically decreasing function in the
Note that I − g ∗ I is a crude approximation of the
radial direction).
Laplacian of I . Unsharp masking enhances the image
Popular derivative filters are the Sobel operator
details by emphasizing the high-frequency part and
for the first derivative, and the average - δ for the
assigning it a higher weight. For some α > 0, the
Laplacian, which use integer filter elements:
output image I is then given by
1 0 −1
Sobel 2 0 −2 I = g ∗ I + (1 + α)(I − g ∗ I )
1 0 −1 = I + α(I − g ∗ I )
(1.21)
= (1 + α)I − α g ∗ I . (1.23)
1 1 1
average - δ 1 −8 1 The parameter α controls the strength of the enhance-
10 1 1 1 ment, and the parameter σ is responsible for the size
Figure 1.14 Radiography of a hand. (a)

Original image I. (b) Smoothed image g ∗ I
with g a 3 × 3 averaging filter. (c) Edges
I − g ∗ I of the image. (d) Unsharp masked
image (α = 5).
(a) (b)
(c) (d)
(a) (b) (c)
Figure 1.15 (a) Original karyotype (chromosome image). (b) Image smoothed with a Gaussian filter. (c) Image filtered with a median filter.
of the frequency band that is enhanced. The smaller in the image. The output image is, however, much
the value of σ , the more unsharp masking focuses on smoother than the input image. In particular, edges
the finest details. are smeared out and may even disappear. To avoid
smoothing, it can therefore be better to calculate the
Nonlinear filters median instead of the mean value in a small window
Not every goal can be achieved by using linear filters. around each pixel. This procedure better preserves
Many problems are better solved with nonlinear meth- the edges (check this with paper and pencil on a step
ods. Consider, for example, the denoising problem. edge). Figure 1.15 shows an example on a chromosome 11
As explained above, the averaging filter removes noise image.
Figure 1.16 The effect of the filter size

in unsharp masking. (a) Original image
(1024 × 1248 pixels). Unsharp masking
with filter size (b) 10, (c) 60, and (d) 125.
Image (b) shows enhanced fine details
but an overall reduction of the contrast.
In image (d), large-scale variations, which
correspond to the lungs and the
mediastinum, are enhanced, and most of
the small details are suppressed. Image
(c) shows a case somewhere between (b)
and (d).
(a) (b)
(c) (d)
Multiscale image processing fixed size of the mask, which is a parameter that must
In the previous sections a number of basic image be chosen. Figure 1.16 shows the effect of the filter size
operations have been described that can be employed for unsharp masking. Using a low-pass filter, the image
for image enhancement and analysis (see for example is split into a low-pass and a remaining high-pass part.
Figures 1.6 and 1.14). Next, the high-pass part is emphasized, both parts are
Gray value transformations (Figure 1.6), such added again (Eq. (1.23)), and the result is normalized
as the widespread window/level operation, increase to the available gray value range. If the filter size is
the contrast in a subpart of the gray value scale. small, this procedure emphasizes small-scale features
They are quite useful for low-contrast objects situ- and suppresses gray value variations that extend over
ated in the enhanced gray value band. Unfortunately, larger areas in the image. With a large-size filter, large
features outside this gray value interval are attenu- image features are enhanced at the expense of the small
ated instead of enhanced. In addition, gray value details. With this method, the following problems are
transformations do not make use of the spatial rela- encountered.
tionship among object pixels and therefore equally • The image operation is tuned to a particular fre-
enhance meaningful and meaningless features such as quency band that is predetermined by the choice
noise. of the filter size. However, diagnostic information
Spatial operations overcome this problem. Dif- is available at all scales in the image and is not
ferential operations, such as unsharp masking limited to a particular frequency band.
(Figure 1.14), enhance gray value variations or edges, • Gray value variations in the selected frequency
whereas other operations, such as spatial averaging
band are intensified equally. This is desired for low-
and median filtering, reduce the noise. However, they
contrast features but unnecessary for high-contrast
focus on features of a particular size because of the
features that are easily perceivable.
12
It is clear that a method is needed that is inde- this textbook. More about multiscale image analysis
pendent of the spatial extent or scale of the image can be found, for example, in [1].
features and emphasizes the amplitude of only the
low-contrast features. Multiscale image processing has
been studied extensively, not only by computer scien-
tists but also by neurophysiologists. It is well known [1] B. M. ter Haar Romeny. Front-End Vision and Multi-Scale Image
Analysis: Multi-Scale Computer Vision Theory and Applications writ-
that the human visual system makes use of a multiscale ten in Mathematica, Volume 27 of Computational Imaging and
approach. However, this theory is beyond the scope of Vision. Springer, 2003.
13
Chapter
2 Radiography
Introduction consequently, the corresponding photon energies are

X-rays were discovered by Wilhelm Konrad Röntgen on the order of keV (1 eV = 1.602 × 10−19 J).
in 1895 while he was experimenting with cathode X-rays are generated in an X-ray tube, which con-
tubes. In these experiments, he used fluorescent sists of a vacuum tube with a cathode and an anode
screens, which start glowing when struck by light emit- (Figure 2.2(a)). The cathode current J releases elec-
ted from the tube. To Röntgen’s surprise, this effect trons at the cathode by thermal excitation. These
persisted even when the tube was placed in a carton electrons are accelerated toward the anode by a voltage
box. He soon realized that the tube was emitting not U between the cathode and the anode. The elec-
only light, but also a new kind of radiation, which he trons hit the anode and release their energy, partly
called X-rays because of their mysterious nature. This in the form of X-rays, i.e., as bremsstrahlung and
new kind of radiation could not only travel through characteristic radiation. Bremsstrahlung yields a con-
the box. Röntgen found out that it was attenuated tinuous X-ray spectrum while characteristic radiation
in a different way by various kinds of materials and yields characteristic peaks superimposed onto the
that it could, like light, be captured on a photographic continuous spectrum (Figure 2.2(b)).
plate. This opened up the way for its use in medicine.
Brehmsstrahlung
The first “Röntgen picture” of a hand was made soon
after the discovery of X-rays. No more than a few The energy (expressed in eV) and wavelength of the
months later, radiographs were already used in clinical bremsstrahlung photons are bounded by
practice. The nature of X-rays as short-wave electro-
hc
magnetic radiation was established by Max von Laue E ≤ Emax = qU , λ ≥ λmin = , (2.2)
in 1912. qU
where q is the electric charge of an electron. For

X-rays example, if U = 100 kV, then Emax = 100 keV.
X-rays are electromagnetic waves. Electromagnetic
radiation consists of photons. The energy E of a Characteristic radiation
photon with frequency f and wavelength λ is The energy of the electrons at the cathode can release
an orbital electron from a shell (e.g., the K-shell), leav-
hc ing a hole. This hole can be refilled when an electron
E = hf = , (2.1) of higher energy (e.g., from the L-shell or the M-shell)
λ
drops into the hole while emitting photons of a very
where h is Planck’s constant and c is the speed of light specific energy. The energy of the photon is the differ-
in vacuum; hc = 1.2397 × 10−6 eV m. The electro- ence between the energies of the two electron states;
magnetic spectrum (see Figure 2.1) can be divided for example, when an electron from the L-shell (with
into several bands, starting with very long radio energy EL ) drops into the K-shell (getting energy EK )
waves, used in magnetic resonance imaging (MRI) (see a photon of energy
Chapter 4), extending over microwaves, infrared, vis-
ible and ultraviolet light, X-rays, used in radiography, E = EL − EK (2.3)
up to the ultrashort-wave, high energetic γ-rays, used
in nuclear imaging (see Chapter 5). The wavelength is emitted. Such transitions therefore yield character-
for X-rays is on the order of Angstrøms (10−10 m) and, istic peaks in the X-ray spectrum.
Chapter 2: Radiography
Visible light
Radio Waves Micro waves Infrared light Ultraviolet light X-rays γ-rays
2 –1 –2 –3 –4 –5 –6 –7 –8 –9 –10 –11 –12 –13

10 10 1 10 10 10 10 10 10 10 10 10 10 10 10 10
400nm
Wavelength (m)
700nm
MRI
endoscopy radiography nuclear
CT imaging
Photon energy (eV)
–8 –7 –6 –5 –4 –3 –2 –1 2 3 4 5 6 7
10 10 10 10 10 10 10 10 1 10 10 10 10 10 10 10
Figure 2.1 The electromagnetic spectrum.
X-ray beam intensity

6
Characteristic
radiation
5 Ka
25 kV
4
20 kV Continuous
3 radiation
lead shield absorbing most X-rays Kb

15 kV
vacuum 2
cooling
cathode e– anode
10 kV
1
focus shield
x-rays
l min 5 kV
0
0 0.5 1 1.5 2 2.5 3
30–100 kV
(a) (b) Wavelength l, Å
Figure 2.2 (a) Scheme of an X-ray tube. (b) Intensity distribution in the Röntgen spectrum of molybdenum for different voltages. The
excitation potential of the K-series is 20.1 kV. This series appears as characteristic peaks in the 25 kV curve. The peaks Kα and Kβ are due to 15
L-shell and M-shell drops respectively.
The important parameters of an X-ray source are the incoming photon was traveling. This mecha-
the following. nism is called photoelectric absorption.
• A second possibility is that the photon transfers
• The amount of electrons hitting the anode and, only part of its energy to eject an electron with a
consequently, the amount of emitted photons con- certain kinetic energy. In that case, a photon of the
trolled by the cathode current multiplied by the remaining lower energy is emitted and its direction
time the current is on (typically expressed in mA s). deviates from the direction of the incoming pho-
Typical values range from 1 to 100 mA s. ton. The electron then escapes in another direction.
• The energy of the electrons hitting the anodes This process is called Compton scattering.
and, consequently, the energy of the emitted pho- • A third mechanism is pair production. If the energy
tons (typically expressed in keV), controlled by of a photon is at least 1.02 MeV, the photon can
the voltage between cathode and anode (typically be transformed into an electron and a positron
expressed in kV). For most examinations the val- (electron–positron pair). A positron is the antipar-
ues vary from 50 to 125 kV. For mammography ticle of an electron, with equal mass but opposite
the voltage is 22–34 kV. The energy of the atom charge. Soon after its formation, however, the
defines the upper limit of the photon energy. positron will meet another electron, and they will
• The total incident energy (typically expressed in annihilate each other while creating two photons
joules, 1 J = 1 kV mA s) at the anode, defined by of energy 511 keV that fly off in opposite direc-
the product of the voltage, the cathode current and tions. This process finds its application in nuclear
the time the current is on. Note that almost all of medicine.
this energy is degraded to heat within the tube. Less • At still higher energies, photons may cause nuclear
than 1% is transmitted into X-rays. reactions. These interactions are not used for
medical applications.
Interaction with matter Interaction of an X-ray beam with tissue

Interaction of photons with matter Consider an X-ray beam and a material of thickness
X-rays and γ-rays are ionizing waves. Such photons are d = xout − xin (see Figure 2.3(a)). Inside the mate-
able to ionize an atom, i.e., to release an electron from rial, the beam is attenuated by the different types of
the atom. Photons with energy less than 13.6 eV are interaction explained above. Although the individual
nonionizing. These photons cannot eject an electron interactions are of statistical nature, the macroscopic
from its atom, but are only able to raise it to a higher intensity of the beam follows a deterministic expo-
energy shell, a process called excitation. Ionizing nential law: The intensity of the outgoing beam Iout is
photons can interact with matter in different ways. related to the intensity of the incoming beam Iin by
• The energy of X-ray photons can be absorbed by an Iout = Iin e−µ d , (2.4)
atom and immediately released again in the form where µ is called the linear attenuation coefficient (typ-
of a new photon with the same energy but traveling ically expressed in cm−1 ). This simple law is only valid
in a different direction. This nonionizing process when the material is homogeneous and the beam con-
is called Rayleigh scattering or coherent scattering sists of photons of a single energy. Actually, µ is a
and occurs mainly at low energies (<30 keV). The function of both the photon energy and the material,
lower the energy the higher is the scattering angle. that is, µ = µ(E, material), for example:
In most radiological examinations it does not play
a major role because the voltage used is typically µ(10 keV, H2 O) = 5 cm−1
in the range from 50 to 125 kV. For mammogra-
µ(100 keV, H2 O) = 0.17 cm−1
phy, however, the voltage is lower (22–34 kV) and (2.5)
Rayleigh scatter cannot be neglected. µ(10 keV, Ca) = 144 cm−1
• A photon can be absorbed by an atom while µ(100 keV, Ca) = 0.40 cm−1 .
16 its energy excites an electron. The electron then
escapes from its nucleus in the same direction as Equation (2.4) can be generalized as follows.
(a) (b) µ, cm–1

100
K-edge
I in I out
10
Pb I
PbII
1 Pb
Al PbIII
AlII
0.1
x in x out Al I
AlIII
0.01
0.01 0.1 1 10 100
Photon energy, MeV
Figure 2.3 (a) X-ray beam traveling through a slab of material. (b) Linear attenuation coefficient for photons in aluminum and lead. The solid
curves represent the total linear attenuation coefficient. The dashed lines show the partial linear attenuation coefficient for each of the three
effects: I for photoelectric absorption, dominant at low energies; II for Compton scattering, dominant at higher energies; III for pair production,
dominant at very high energies.
• When a beam of single-energy photons trav- atomic number Z . With increasing Z , photoelectric
els through a nonhomogeneous medium, Iout is absorption increases more rapidly than Compton
related to Iin by scattering.
Often the mass attenuation coefficient (µm ) is used
xout
− µ(x) dx instead of the linear attenuation coefficient:
Iout = Iin e xin . (2.6)
µm = µ/ρ, (2.8)
• A real X-ray beam does not contain a single photon
energy but a whole spectrum of energies. Making
the intensity distribution of the incoming where ρ is the mass density of the attenuating medium.
∞ beam a For example, for water ρ = 1 g/cm3 and for calcium
function of the energy, that is, Iin = 0 σ (E) dE,
the intensity of the outgoing beam is equal to ρ = 1.55 g/cm3 . Hence,
∞ xout µm (10 keV, H2 O) = 5 cm2 /g

− µ(E,x) dx
Iout = σ (E) e xin dE. (2.7)
0 µm (100 keV, H2 O) = 0.17 cm2 /g
(2.9)
µm (10 keV, Ca) = 93 cm2 /g
Figure 2.3(b) shows that at low energies photoelectric
absorption is most prominent while at intermediate µm (100 keV, Ca) = 0.258 cm2 /g.
energies the Compton scattering dominates. Pair pro-
duction exists only at very high energies. Photoelectric
absorption occurs at photon energies higher than the X-ray detectors
binding energy of K-shell electrons. Hence, the atten- To produce an image from the attenuated X-ray beam,
uation coefficient suddenly increases at this energy, the X-rays need to be captured and converted to
known as the K-edge. Figure 2.3(b) also shows that with image information. Some detectors for digital radio-
increasing energy, photoelectric absorption decreases graphy are relatively recent developments. Older but
more rapidly than Compton scattering. Note also that still in use are the screen–film detector and the image 17
the linear attenuation coefficient increases with the intensifier.
Screen–film detector stopped. If the delay to reach peak emission is

Screen longer than 10−8 seconds or if the material con-
tinues to emit light after this period, it is said to
Photographic film is very inefficient for capturing
phosphoresce. Phosphorescence in screens is an
X-rays. Only 2% of the incoming X-ray photons con-
undesirable effect, because it causes ghost images
tribute to the output image on a film. This percentage
and occasionally film fogging.
of contributing photons corresponds to the probabil-
ity that an X-ray photon (quantum) is absorbed by
the detector. It is known as the absorption efficiency. Film
The low sensitivity of film for X-rays would yield The film contains an emulsion with silver halide crys-
prohibitively large patient doses. Therefore, an inten- tals (e.g., AgBr). When exposed to light, the silver
sifying screen is used in front of the film. This type of halide grains absorb optical energy and undergo a
screen contains a heavy chemical element that absorbs complex physical change. Each grain that absorbs
most of the X-ray photons. When an X-ray photon a sufficient amount of photons contains dark, tiny
is absorbed, the kinetic energy of the released elec- patches of metallic silver called development centers.
tron raises many other electrons to a higher energy It is important to note that the amount of photons
state. When returning to their initial state they pro- required is independent of the grain size. When the
duce a flash of visible light, called a scintillation. Note film is developed, the development centers precipi-
that these light photons are scattered in all directions. tate the change of the entire grain to metallic silver.
Consequently, two intensifying screens can be used, The more light reaching a given area of the film, the
i.e., one in front and one behind the film, to increase more grains are involved and the darker the area after
the absorption efficiency further. The portion of the development. In this way a negative is formed. After
light that is directed toward the film contributes to the development, the film is fixed by chemically removing
exposure of the film. In this way, the absorption effi- the remaining silver halide crystals.
ciency can be increased to more than 50% instead of In radiography, the negative image is the final
the 2% for film. Because the light is emitted in all direc- output image. In photography, the same procedure
tions, a smooth light spot (the PSF, see p. 3) instead of has to be repeated to produce a positive image. The
a sharp peak hits the film and causes image blurring. negative is then projected onto a sensitive paper car-
X-ray intensifying screens consist of scintillating rying silver halide emulsion similar to that used in the
substances that exhibit luminescence. Luminescence is photographic film.
the ability of a material to emit light after excitation, Typical characteristics of a film are its graininess,
either immediately or delayed. speed, and contrast.
• Fluorescence is the prompt emission of light when • Graininess The image derived from the silver crys-
excited by X-rays and is used in intensifying tals is not continuous but grainy. This effect is
screens. A material is said to fluoresce when light most prominent in fast films. Indeed, because the
emission begins simultaneously with the exciting amount of photons needed to change a grain into
radiation and light emission stops immediately metallic silver upon development is independent
after the exciting radiation has stopped. Initially, of the grain size, the larger the grains, the faster the
calcium tungstate (CaWO4 ) was most commonly film becomes dark.
used for intensifying screens. Advances in tech- • Speed The speed of a film is inversely propor-
nology have now resulted in the use of rare tional to the amount of light needed to produce
earth compounds, such as gadolinium oxysulfide a given amount of metallic silver on develop-
(Gd2 O2 S). A more recent scintillator material is ment. The speed is mainly determined by the
thallium-doped cesium iodide (CsI:Tl), which has silver halide grain size. The larger the grain size
not only an excellent absorption efficiency but also the higher the speed because the number of pho-
a good resolution because of the needle-shaped or tons needed to change the grain into metallic silver
pillarlike crystal structure, which limits lateral light upon development is independent of the grain size.
diffusion. Speed is expressed in ASA (American Standards
• Phosphorescence or afterglow is the continuation Association) or in ISO (International Standards
18
of light emission after the exciting radiation has Organization). These units are the same.
For X-ray imaging with a screen–film combina- D

4
tion, it makes more sense to speak about the speed
of the screen–film combination: how many X-ray 3.5
photons are needed to produce a certain density
3
on the film. The speed then depends on the prop-
erties of the intensifying screen and the film, but 2.5
also on the quality of film–screen contact, and on a
good match between the emission spectrum of the 2
screen and the spectral sensitivity of the film used. 1.5

Because light is emitted in all directions, a sig-
nificant proportion, about 50%, of that light is 1
not directed toward the film. A reflective layer 0.5
behind the screen–film–screen redirects it toward
the film, ensuring that it contributes to expo- 0
log E
sure. This has the advantage of increasing the
speed of the screen–film–screen combination with Figure 2.4 Typical sensitometric curve for radiographic film. D is
a corresponding reduction in patient dose. the optical density and E the exposure.
• Contrast The most widely used description of the

photosensitive properties of a film is the plot of
the optical density D versus the logarithm of the
Image intensifier
An image intensifier works as follows (see Figure 2.5).
exposure E. This curve is called the sensitometric
A fluorescent screen converts the X-rays into visible
curve. The exposure is the product of incident light
light. The emitted light hits a photocathode, and the
intensity and its duration. The optical density is
energy of the photons releases electrons from this cath-
defined by
ode. A large potential difference between the cathode
and the output accelerates the ejected electrons. The
Iin resulting electron beam is directed onto a small flu-
D = log , (2.10)
Iout orescent screen by electrostatic or magnetic focusing
and converted to light photons again. This focusing
where Iin and Iout are the incoming and outgoing makes the system suitable to be coupled to a camera
light intensity when exposing the developed film without any loss of light. The main advantage of an
with a light source. Note that Iin and Iout are image intensifier system is that it is capable of pro-
different from the incident light intensity in the ducing dynamic image sequences in real time at video
definition of the exposure E, in which it refers to rate, a process known as fluoroscopy. However, when
the light emitted by the intensifying screen during compared with film–screen systems, the images are
X-ray irradiation. degraded in three ways.
Figure 2.4 shows a typical sensitometric curve. • The spatial resolution will generally be less than
It is S-shaped. In low- and high-density areas, con- that of a film–screen system because of the limited
trast is low and there is little information. Only the camera resolution.
linear part is really useful and its slope character-
• Because of the additional conversions (light →
izes the film contrast. The maximal slope of the
electrons → light), the noise increases slightly.
curve is known as the gamma of the film. Note that
• Geometric distortion occurs, called pin-cushion
a larger slope implies a higher contrast at the cost
of a smaller useful exposure range.∗ distortion, particularly toward the borders of the
image.
∗ Double contrast films also exist. Their sensitometric curve con- Detectors for digital radiography
tains two linear parts with a different slope, i.e., a high-contrast part Storage phosphors
as usual, continued by a low-contrast part at high optical densities.
This increases the perceptibility in hyperdense regions, such as in A special case of phosphorescence is when part of 19
mammography at the border of the breast. the absorbed energy is not released immediately in
Input Figure 2.5 Scheme of an image

Phosphor intensifier. The camera is placed against
the output screen to minimize light loss.
Input (Reprinted with permission of Kieran
Photocathode Evacuated Tube Maher.)
Screen
X-rays Output
Screen
light
Electron lenses
Output
Phosphor
High Voltage Power Supply
the form of light. The temporarily stored energy

can be released upon stimulation by other forms of
energy such as laser light. This phenomenon is called
photostimulated luminescence and is used in digital
radiography. This type of scintillator is called a storage
phosphor or photostimulable phosphor. The screen–
film combination is then replaced by a screen coated
with such a scintillator. When X-rays are absorbed
by the phosphor, electrons are pumped up from the
valence band to the conduction band. In a classical
scintillator plate such an electron falls back to the
valence band while releasing its energy in the form of
a light photon. In a storage phosphor, however, these
electrons are trapped by electron traps, which are impu-
rities in the scintillator. In this way, the incident X-ray
energy is converted into stored energy. After expo-
sure a latent image is trapped in the scintillator. The
latent image can be stored in the phosphor plate for a
considerable period after exposure. It takes 8 hours to
decrease the stored energy by about 25%. The stored
energy can be extracted by pixelwise scanning with a
laser beam. This way the trapped electrons receive a
new energy shot that allows them to escape from their
trap and fall back into the valence band. The latent Figure 2.6 This system scans the latent image with a laser beam
and erases the residual image on the storage phosphor after which
image information is thereby released as visible light, the screen can be reused for new X-ray exposure.
which is captured by an optic array and transmitted
to a photomultiplier. The photomultiplier converts
the detected light into an analog electrical signal. This exposure. As soon as the radiologic technician puts the
analog signal is then converted in an A/D converter to cassette into the scanner (Figure 2.6), this whole laser
a digital bit stream. The residual information on the scanning and cleaning process is done automatically.
20 scintillator screen is erased by a strong light source, Storage phosphor screens provide a much wider
after which the screen can be reused for new X-ray useful exposure range than conventional film–screen
systems. Moreover, the storage phosphor is a linear light photons negatively influences the PSF because of
detector. This means there is no contrast reduction the light distribution in different directions. A more
in the low- and high-density areas of the image, as is recent technique eliminates the need for a scintillator
the case with the S-shaped sensitometric curve. Con- by using a photoconductor, such as amorphous sele-
sequently, the system is much more tolerant to over- nium (a-Se) or cadmium telluride (CdTe), instead of
exposure and underexposure, and retakes caused by a phosphor. When exposed to radiation, the photo-
suboptimal exposure settings (mA s, kV) are reduced. conductor converts the energy of the X-ray photons
In theory, a reduction of the radiation dose per image directly into an electrical conductivity proportional to
is also possible because of the available contrast at low the intensity of the radiation. To scan this latent image,
exposure. However, dose reduction adversely affects the photoconductor layer is placed upon an active
the SNR of the resulting image. Therefore, reduc- matrix array that consists of a 2D array of capacitors
ing the dose per examination must be considered in (instead of photodiodes) deposited onto the amor-
relationship to the diagnostic information required. phous substrate. These capacitors store the electric
Often, the greed for diagnostic detail slightly increases charge produced by detected X-ray photons until it
the dose rather than reducing it. is read out by the electronic circuit of the active matrix
A second advantage of digital radiography is that array. This technology is known as direct radiography
the image is available for computer postprocessing as against the indirect approach where light is pro-
such as image enhancement and quantification. More- duced by a scintillator as an intermediate step in the
over, the image can easily be stored and transported transformation of X-rays to a measurable signal.
in digital form, making the images more accessible Active matrix flat panel detectors have become an
and making large film archives unnecessary. Today, accepted technology for mammography because of
digital picture archiving and communication systems their overall performance (see p. 24 below on DQE).∗
(PACS) are part of hospital information systems, mak-
ing the medical images immediately available through
the digital network in the same way as the other patient Dual-energy imaging
information. By taking two radiographic images, each capturing
a different energy spectrum, the image of substances
with a high atomic number (e.g., bone, calcifications,
Active matrix flat panel detectors stents) can be separated from that of the soft tissue
Newer detector technologies for digital radiography by proper image processing. This way two different
are flat panel detectors with fast-imaging capability. selective images are obtained, for example, a soft-
These systems produce nearly real time images, as tissue image and a bone image. Several methods have
opposed to storage phosphor systems which require a been proposed to calculate tissue selective images.
readout scan on the order of a minute and a workflow The method explained here is also used in computed
similar to that for screen–film systems. tomography (p. 48).
Traditional electronic capturing devices, including Two system configurations have been used. The
CCDs (charge-coupled devices), are almost exclusively first captures two radiographic images in a short time
based upon Si-crystal technology, and for manufac- interval (e.g., 200 ms) and at different X-ray tube volt-
turing reasons this restricts the devices to small areas. ages (e.g., peaks at 110–150 kV and at 60–80 kV). The
This is because it is difficult and expensive to create a second configuration contains two layers of scintillator
large defect-free semiconductor crystal. A flat, large- detectors and acquires the images in a single expo-
area integrated circuit, called an active matrix array, sure. The top layer detects and filters most low-energy
can easily be made by depositing a 2D array of identical photons, while the bottom layer detects primarily
semiconductor elements onto an amorphous material, high-energy photons. A third configuration is promis-
such as hydrogenated amorphous silicon (a-Si:H). ing but immature. It uses photon counting detectors
A light-sensitive active matrix array can be pro-
duced by depositing an array of photodiodes onto the
a-Si:H substrate. By coupling it to a fluorescent plate it ∗ Commercial mammography systems exist that are able to count
functions as a large and fast flat panel X-ray detector. the individual X-ray photons with a very high absorption effi-
ciency. To obtain their unsurpassed DQE they make use of
In spite of the technological progress in scintilla- crystalline silicon strip detectors in combination with a slit-scanning 21
tor materials, the conversion of X-ray radiation into technology.
3.00
cm–1 be written as a linear combination of the attenuation
Calcium
coefficient of two selected materials, provided that
2.00
the attenuation properties of both basis materials are
1.50
sufficiently different (e.g., bone and soft tissue)
1.00
0.70 µS (E) = a1 · µ1 (E) + a2 · µ2 (E). (2.13)

Iodine
0.50
Substituting Eq. (2.13) in Eq. (2.7) for a spectrum with
0.30 energy range [Emin , Emax ] yields
0.20 Water
Emax
− µ(E,s) ds
0.15 I (x, y) = σ (E) e Lx,y dE
20 40 60 80 100 120 140 keV Emin
Figure 2.7 Linear attenuation coefficient as a function of the
Emax
− Lx,y (a1 (s)·µ1 (E)+a2 (s)·µ2 (E)) ds
energy for calcium (Ca), water and the contrast agent iodine (I) in = σ (E) e dE,
water (10 mg/ml). Note the K-edge discontinuity of I at 33.2 keV. Emin
(2.14)
that count and measure the energy of the photons. where Lx,y is the projection line arriving at pixel (x, y)
This multi-energy technique has an improved spectral of the radiographic image. When taking images, the
sensitivity, needs only one radiographic image and is tissue-dependent coefficients a1 and a2 are unknown.
insensitive to patient motion. Defining
Dual-energy imaging relies on the dependence
of the attenuation coefficient µ on the energy E A1 (x, y) = a1 (s) ds
(Figure 2.7). In the absence of K-edge discontinu- Lx,y
(2.15)
ities in the used energy range [Emin , Emax ] the linear
A2 (x, y) = a2 (s) ds,
attenuation coefficient can be approximated as Lx,y
µ(E) ≈ µp (E) + µC (E) Eq. (2.14) can be written as

1
≈ ap + aC fKN (E). (2.11) I (x, y)
Em Emax
The two components express the attenuation due to = σ (E) e−(A1 (x,y)·µ1 (E)+A2 (x,y)·µ2 (E)) dE.
photoelectric interaction and Compton scatter respec- Emin
tively. The exponent m is an empirically defined (2.16)

parameter (e.g., m = 3 [2]). fKN (E) is the so-called A1 (x, y) and A2 (x, y) represent the equivalent thick-
Klein–Nishina function. The tissue-dependent coeffi- ness of the basis materials along ray Lx,y . In this
cients ap and aC are related to the physical material equation A1 (x, y) and A2 (x, y) are unknown, but they
properties: can be retrieved. Indeed, if two radiographic images
ρ are acquired, each at a different energy with corre-
ap ≈ Kp Z n , n≈4 sponding spectra σLE and σHE , the following system
A
ρ (2.12) of two nonlinear equations must be solved to calculate
aC ≈ KC Z A1 (x, y) and A2 (x, y) in pixel (x, y)
A
where Kp and KC are constants, ρ is the mass density, IHE (x, y)
A the mass number and Z the atomic number of the Emax
attenuating medium [2]. = σLE (E) e−(A1 (x,y)·µ1 (E)+A2 (x,y)·µ2 (E)) ds dE
Using Eq. (2.11) it can easily be shown that the Emin
attenuation coefficient of an arbitrary substance S can ILE (x, y)

Emax
[2] R. E. Alvarez and A. Macovski. Energy-selective reconstructions = σHE (E) e−(A1 (x,y)·µ1 (E)+A2 (x,y)·µ2 (E)) ds dE.
22 in x-ray computerized tomography. Physics in Medicine and Biology, Emin
21(5): 733–744, 1976. (2.17)
In case of a single exposure with spectrum σLE and two • The size of the focal spot. The anode tip should
detector layers, the second spectrum σHE is defined as make a large angle with the electron beam to
produce a nicely focused X-ray beam.
σHE (E) = σLE (E) e−µf (E) tf , (2.18) • The patient. Thicker patients cause more X-
ray scattering, deteriorating the image resolution.
where µf (E) and tf are the known attenuation coef- Patient scatter can be reduced by placing a collima-
ficient and thickness of the filtering top detector tor grid in front of the screen (see p. 24). The grid
layer. allows only the photons with low incidence angle
Various approaches exist to solve Eqs. (2.17). For to reach the screen.
• The light scattering properties of the fluorescent
example, the inverse relationship can be modeled by a
second- or third-order polynomial. If more than two screen.
measurements and corresponding equations are avail- • The film resolution, which is mainly determined
able, an optimization strategy is required to solve the by its grain size.
overdetermined system. This is for example the case • For image intensifier systems and digital radiogra-
when photon-counting detectors can be used. More phy, the sampling step at the end of the imaging
information on numerical optimization can be found chain is an important factor.
in [3].
Using the obtained values of A1 (x, y) and A2 (x, y) The resolving power (i.e., the frequency where
the original radiographic image can be separated the MTF is 10%) of clinical screen–film combinations
into two material equivalent images (e.g., bone and varies from 5 up to 15 lp/mm. In most cases, spatial
soft tissue). Note that the above theory needs some resolution is not a limiting factor in reader perfor-
modification in the presence of a substance with an mance with film. For images with storage phosphors,
observable K-edge in the energy range [Emin , Emax ]. In a resolving power of 2.5 up to 5 lp/mm (at 10% con-
that case Eqs. (2.11) and (2.13) have to be extended trast) is obtained. This corresponds to a pixel size of
with a third component and corresponding coeffi- 200 to 100 µm, which is mostly sufficient except for
cients aK and a3 respectively. This yields a third mammography, for which more recent detector tech-
unknown A3 in Eq. (2.16) and, hence, requires a multi- nology (see active matrix flat panel detectors, p. 21) is
energy approach with at least three different measure- needed. Depending on the size of the object, it is clear
ments [4]. The original image is then separated into that images with 2000 by 2000 pixels and even more
three instead of two basis images, the third being an are needed to obtain an acceptable resolution.
image of the substance with K-edge. The strength of
K-edge imaging is that the energy dependence of a
material with K-edge is very different around its K- Contrast
edge, resulting in a high sensitivity for multi-energy The contrast is the intensity difference in adjacent
imaging. K-edge imaging is immature but offers regions of the image. According to Eq. (2.7) the
opportunities for target-specific contrast agents and image intensity depends on the attenuation coeffi-
drugs, particularly in multi-energy CT (see p. 48). cients µ(E, x) and thicknesses of the different tissue
layers encountered along the projection line. Because
the attenuation coefficient depends on the energy of
Image quality the X-rays, the spectrum of the beam has an important
Resolution influence on the contrast. Soft radiation, as used in
mammography, yields a higher contrast than hard
The image resolution of a radiographic system radiation.
depends on several factors. Another important factor that influences the con-
trast is the absorption efficiency of the detector, which
[3] J. Nocedal and S. Wright. Numerical Optimization, Volume XXII is the fraction of the total radiation hitting the detec-
of Springer Series in Operations Research and Financial Engineering. tor that is actually absorbed by it. A higher absorption
Springer, second edition, 2006. efficiency yields a higher contrast.
[4] E. Roessl and R. Proksa. K-edge imaging in x-ray computed
tomography using multi-bin photon counting detectors. Physics in In systems with film, the contrast is strongly deter- 23
Medicine and Biology, 52: 4679–4696, 2007. mined by the contrast of the photographic film. The
higher the contrast, the lower the useful exposure influence the DQE, particularly the absorption effi-
range. In digital radiography, contrast can be adapted ciency of the detector, the point spread function of
after the image formation by using a suitable gray value the detector and the noise introduced by the detector.
transformation (see p. 4). Note however that such a Figure 2.8 shows an example of the DQE as a function
transformation also influences the noise, thus keeping of frequency for three different detector technologies.
the CNR unchanged.
Noise Artifacts
Quantum noise, which is due to the statistical nature Although other modalities suffer more from severe
of X-rays, is typically the dominant noise factor. A artifacts than radiography, X-ray images are generally
photon-detecting process is essentially a Poisson pro- not artifact free. Scratches in the detector, dead pixels,
cess (the variance is equal to the mean). Therefore, unread scan lines, inhomogeneous X-ray beam inten-
the noise amplitude (standard deviation) is propor- sity (heel effect), afterglow, etc., are not uncommon
tional to the square root of the signal amplitude, and and deteriorate the image quality.
the SNR also behaves as the square root of the sig-
nal amplitude. This explains why the dose cannot be Equipment
decreased unpunished. Doing so would reduce the
Let us now take a look at the complete radiographic
SNR to an unacceptable level. Further conversions
imaging chain, which is illustrated schematically in
during the imaging process, such as photon–electron
Figure 2.9. It consists of the following elements.
conversions, will add noise and further decrease
the SNR. • The X-ray source.
To quantify the quality of an image detector the • An aluminum filter, often complemented by a cop-
measure detective quantum efficiency (DQE) is often per filter. This filter removes low-energy photons,
used. The image detector is one element in the imag- thus increasing the mean energy of the photon
ing chain and to quantify its contribution to the SNR, beam. Low-energy photons deliver doses to the
the DQE is used, which expresses the signal-to-noise patient but are useless for the imaging process
transfer through the detector. The DQE can be cal- because they do not have enough energy to travel
culated by taking the ratio of the squared SNR at the through the patient and never reach the detector.
detector output to the squared SNR of the input signal Because low-energy photons are called soft radia-
as a function of spatial frequency: tion and high-energy photons hard radiation, this
removal of low-energy photons from the beam is
SNR 2out (f ) called beam hardening.
DQE = . (2.19)
SNR 2in (f ) • A collimator to limit the patient area to be
irradiated.
It is a measure of how the available signal-to-noise
ratio is degraded by the detector. Several factors
low-energy
collimating
absorbing
scatter grid
filter
Figure 2.8 DQE of four digital X-ray detection systems, obtained X-ray source collimator detector
under standardized measurement conditions. The DQE curve is cut
24 off at a frequency close to the Nyquist frequency, i.e., half of the Figure 2.9 Schematic representation of the radiographic imaging
sampling frequency. (Courtesy of Agfa HealthCare.) chain.
• The patient, who attenuates the X-ray beam and storage phosphor cassette, as well as an image inten-
produces scatter. sifier beneath the table. More recent X-ray systems
• A collimating scatter grid. This is a collimator contain an active matrix flat panel detector with fast-
that absorbs scatter photons. It stops photons with imaging capability, which replaces the cassette and
large incidence angle, whereas photons with small image intensifier (Figure 2.11).
incidence angle can pass right through the grid. Figure 2.12 shows a 3D rotational angiography sys-
The grid can be made of lead, for example. Note tem (3DRA). Images of the blood vessels can be made
that a scatter grid is not always used in paediatrics from any orientation by rotating the C-arm on which
because in small children the scatter is limited. the X-ray tube and image detector are mounted at
•
both ends. By continuously rotating the C-arm over
The detector. This can be a screen–film combina-
a large angle (180◦ and more), sufficient projection
tion in which a film is sandwiched between two
images are obtained to reconstruct the blood vessels
screens (see p. 18), an image intensifier coupled to
in three dimensions (3D) (Figure 2.13). The mathe-
a camera (see p. 19), a cassette containing a storage-
matical procedure used to calculate a 3D image from
phosphor plate (see p. 19), or an active matrix flat
its projections is also used in computed tomography
panel detector (see p. 21) or dual-layer detector
(CT) and is explained in Chapter 3.
(see p. 21).
Figure 2.10 shows a general purpose radiographic Clinical use

room. The table can be tilted in any orientation, from Today, the majority of the radiographic examinations
the horizontal to the vertical position. The X-ray sys- in a modern hospital are performed digitally. X-ray
tem contains a tray for a conventional film-based or a images can be static or dynamic. Static or still images
(a) (b)
Figure 2.10 Multipurpose radiographic room. The table can be tilted in any orientation. Both an image intensifier and a storage phosphor 25
are available.
Radiographic images
These are made of all parts of the human body. They
are still responsible for the majority of radiologi-
cal examinations. The most common investigations
include the following.
• skeletal X-rays (see Figure 2.14),
• chest images (radiographs of the thoracic cavity
and heart, see Figure 2.15),
• mammography (images of the breasts, see
Figure 2.16),
• dental X-rays (images of the teeth and jaw).
Fluoroscopic images
These are image sequences produced in real time.
Consequently, their application field focuses on inves-
tigations in which motion or the instant availability of
the images, or both, are crucial. This application field
is obviously narrower than that of radiographic exam-
inations, which explains why the number of fluoro-
scopic guided examinations is an order of magnitude
lower. The most typical applications, in decreasing
order of occurrence, include the following.
• Interventional fluoroscopy (see Figure 2.17). This
application is responsible for the majority of fluo-
roscopic sequences. Typically, the images are used
Figure 2.11 In more recent X-ray systems the cassette and image to guide and quickly verify surgical actions, partic-
intensifier are replaced by an active matrix flat panel detector. This ularly in bone surgery, such as for osteosynthesis
picture shows a Siemens system with large-area amorphous silicon
detector coupled to a CsI scintillator plate.
(traumatology, orthopedics).
• Angiography (see Figure 2.18), which takes images
of blood vessels through the injection of an iodine-
containing fluid into the arteries or veins. Usually,
are made with a film–screen combination or with digi- subtraction images are made by mathematically
tal radiography, whereas dynamic images are obtained subtracting postcontrast and precontrast images
with an image intensifier or an active matrix flat panel followed by a simple gray level transformation to
detector and viewed in real time on a TV monitor increase the image contrast of the vessels. The
or computer screen. Dynamic image sequences are result is an image in which the blood vessels appear
commonly known as fluoroscopic images as against as contrasting line patterns on a homogeneous
radiographic images, which refer to static images. background. Obviously, it is essential that the
In X-ray images, the attenuation differences of patient does not move during the imaging pro-
various nonbony matter are usually too small to distin- cedure, to avoid motion blurring and subtraction
guish them. A contrast agent or dye (i.e., a substance artifacts in the images. Traditionally, angiogra-
with a high attenuation coefficient) may overcome phy has been used for diagnosis of conditions
this problem. It is especially useful for intravascu- such as heart ischemiae caused by plaque buildup.
lar (blood vessels, heart cavities) and intracavitary However, today radiologists, cardiologists, and
(kidney, bladder, etc.) purposes. vascular surgeons also use the X-ray angiography
Following are a number of typical examples of fre- procedure to guide minimally invasive interven-
26 quently used examinations. They are subdivided into tions of the blood vessels, such as for vascular
radiographic images and fluoroscopic images. repermeabilization (Figure 2.12).
(a) (b)
Figure 2.12 3D rotational angiography (3DRA). (a) C-arm with X-ray tube and image intensifier at both ends. (b) More recent system in which
the image intensifier has been replaced by an active matrix flat panel detector with an acquisition frame rate of up to six 2048 × 2048 images
(12 bbp) per second. By rotating the C-arm on a circular arc (e.g., 240◦ in 4 s) around the patient, a series of projection images are acquired
that can be used to compute a 3D image of the blood vessels. (Courtesy of Professor G. Wilms, Department of Radiology.)
• Barium fluoroscopy of the gastrointestinal tract

after the patient swallows barium contrast solu-
tion and/or where the contrast is instilled via the
rectum (see Figure 2.19).
• Urography (image of the kidneys and bladder)
using an iodine-containing contrast fluid.
Biologic effects and safety

Even at very low X-ray doses the energy deposited by
ionizing radiation, such as X-rays, may be sufficient
to damage or destroy cells. Although this generally
has no negative consequence, the probability always
exists that modifications in single cells could lead to
malignancy (cancer) or genetic changes. There is no
evidence of a threshold dose below which the proba-
bility would be zero. If the X-ray dose increases, the
frequency of cell damage and the occurrence of cancer
Figure 2.13 3D image of the cerebral blood vessels reconstructed increases, but not the severity of the cancer.
from a series of 2D projection images around the patient, obtained Malignant disease and heritable effects, for which
with the 3DRA system shown in Figure 2.12(b). Selective injection of the probability but not the severity is proportional to
the right internal carotid artery in a patient with a subarachnoid
hemorrhage showing an aneurysm of the anterior communicating the dose, without any threshold, are stochastic effects of 27
artery. (Courtesy of Professor G. Wilms, Department of Radiology.) radiation. Deterministic effects of radiation also exist.
(a) (b)
Figure 2.14 (a) Double mandibular fracture with strong displacement to the left. (b) Solitary humeral bone cyst known as ‘‘fallen leaf sign’’.
(Courtesy of Dr. L. Lateur, Department of Radiology.)
(a) (b)
Figure 2.15 Radiographic chest image showing multiple lung metastases. (Courtesy of Professor J. Verschakelen, Department of Radiology.)
28
(a) (b)
Figure 2.16 (a) Dense opacity with spicular border in the cranial part of the right breast; histological proven invasive ductal carcinoma.
(b) Cluster of irregular microcalcifications suggesting a low differentiated carcinoma. (Courtesy of Dr. Van Ongeval, Department. of Radiology.)
They are injuries to a large population of cells where

repair mechanisms fail and the complete tissue is
damaged. Deterministic effects are characterized by
a threshold dose and an increase in the severity of the
tissue reaction with increasing dose.
The SI unit of absorbed dose, D, is the gray (Gy).
One Gy is one joule per kilogram of irradiated mate-
rial. If the average absorbed dose, DT , in organ or
tissue T is known, it is, for example, possible to predict
the onset of deterministic effects. Tables of threshold
values can be found in the literature. For example, a
dose of 5 Gy in a single exposure at the level of the eye
lens can cause visual impairment due to cataract. In
clinical practice deterministic effects are rare.
The probability of stochastic effects from radiation
depends heavily on the type of radiation. Some types
Figure 2.17 Postoperative fluoroscopic control of bone fixation of radiation are more detrimental per unit of absorbed
with plate and screws after a complete fracture of the humerus. dose than others. To assess the risk of the stochastic
(Courtesy of Dr. L. Lateur, Department of Radiology.)
effects of radiation in a particular organ or tissue T ,
29
To assess the overall radiation detriment † from

stochastic effects, the effective dose,†† also expressed
in sieverts (Sv), is used. The effective dose is the tis-
sue weighted sum of equivalent doses in all irradiated
tissues or organs of the body, that is,
E = T (wT · HT ), (2.21)
where HT is the equivalent dose in tissue or organ

T and wT the tissue weighting factor. The weights wT
represent the (rounded) relative radiation detriments
of the individual organs and tissues. The sum of all
weights equals 1:
T wT = 1. (2.22)
For example, the nominal risk coefficient, expressed in

cases per 10 000 persons per sievert, for the liver is 30.
The detriment, i.e., the radiation detriment adjusted
nominal risk coefficient, is 26.6. Given a total detri-
Figure 2.18 Cerebral angiogram showing an aneurysm or saccular
dilation of a cerebral artery. (Courtesy of Professor G. Wilms,
ment for all organs and tissues of 574, the relative
Department of Radiology.) detriment is 0.046 (rounded: wliver = 0.04).
Tissue weighting factors can be found in the “2007
Recommendations of the International Commission
on Radiological Protection” (Annals of the ICRP, pub-
the equivalent dose, HT , is used:
lication 103). They are averaged over all ages and both
sexes and therefore do not apply to particular indi-
HT = R (wR · DT ,R ), (2.20) viduals. Red bone marrow, colon, lung, stomach and
breast have a tissue weighting factor of 0.12. Gonads
where DT ,R is the average absorbed dose from radi- have 0.08; bladder, oesophagus, liver and thyroid have
ation of type R in tissue or organ T and wR is the 0.04; bone surface, brain, salivary glands and skin have
radiation weighted factor, which is a measure of its 0.01. Thirteen remainder tissues have been defined,
relative biological impact per unit of absorbed dose. i.e., adrenals, extrathoracic region, gall bladder, heart,
The SI unit of equivalent dose is the sievert (Sv). The kidneys, lymphatic nodes, muscle, oral mucosa, pan-
radiation weighting factor for X-rays, electrons and creas, prostate (male), small intestine, spleen, thymus
muons is 1, for protons and charged pions it is 2, and and uterus/cervix (female). Because the sum of all
for heavier particles 20. For neutrons there is no sin-
gle value but the weighting factor is a function of the † “Radiation detriment is a concept used to quantify the harmful
neutron energy. In most medical imaging applications effects of radiation exposure in different parts of the body. It is
determined from nominal risk coefficients, taking into account the
only X-rays are involved and wR is simply 1. In the lit- severity of the disease in terms of lethality and years of life lost.
erature, factors can be found that relate the equivalent Total detriment is the sum of the detriment for each part of the
organ or tissue dose to the risk of stochastic effects. For body (tissues and/or organs)” (Annals of the ICRP, publication 103,
2007).
example, lung cancer occurs on average in 114 cases †† “The concept of “effective dose” associated with a given expo-
per 10 000 persons per sievert, yielding a so-called sure involves weighting individual organs and tissues of interest
nominal risk coefficient ∗ of lung cancer induction of by the relative detriments for these parts of the body. In such a
system, the weighted sum of the tissue-specific dose equivalents,
1.14%/Sv. called the effective dose, should be proportional to the total esti-
mated detriment from the exposure, whatever the distribution of
equivalent dose within the body. The components of detriment are
∗ “Nominal risk coefficients are derived by averaging sex and age- essentially the same for cancer and heritable disease and, if desired,
30 at-exposure lifetime risk estimates in representative populations” these detriments may be combined” (Annals of the ICRP, publication
(Annals of the ICRP, publication 103, 2007). 103, 2007).
(a) (b)
Figure 2.19 (a) Double contrast (barium + gas insufflation) enema with multiple diverticula in the sigmoid colon (arrows). (b) Polypoid mass
proliferating intraluminally (arrowhead on the spotview). (Courtesy of Professor E. Ponette, Department of Radiology.)
weights equals 1, the weight for the remainder tissues According to the International Commission on
is 0.12. It must be applied to the arithmetic mean dose Radiological Protection (ICRP) the relative radiation
of the 13 organs and tissues. detriment adjusted nominal risk coefficient for cancer
Effective dose is a valuable measure to compare dif- is 5.5%/Sv and for heritable effects up to the second
ferent examinations. Examples of effective doses for generation is 0.2%/Sv. For adults (18 to 64 years), these
some typical radiographic examinations are: dental risk factors are a little lower, i.e., 4.1%/Sv and 0.1%/Sv
0.005–0.02 mSv; chest 0.01–0.05 mSv; skull 0.1– respectively.
0.2 mSv; pelvis 0.7–1.4 mSv; lumbar spine 0.5–1.5 mSv. Because of the potential risk of medical irradia-
Note that many examinations require more than one tion, the ICRP recommends keeping the magnitude
or a continuous X-ray exposure, which increases of individual examination doses as low as reason-
the dose. The use of fluoroscopy for diagnostic and ably achievable (ALARA principle). There are no dose
therapeutic reasons may yield doses around 5 mSv. limits for patients, but every exposure should be jus-
Examples are intravenous urography (3 mSv), barium tified. This is, to a large extent, a medical decision.
enema (8 mSv) and endoscopic retrograde cholan- The physician should have as much knowledge as
giopancreatography (4 mSv). Interventional proce- possible about previous examinations of the patient
dures, such as performed in the angiography room and about the patient’s condition. Pregnancy, for
or in the catheterization lab, may have much higher example, is a state where risks are increased. Most
doses, and occasionally even skin doses that reach countries have now introduced diagnostic reference
the thresholds for deterministic effects. Relatively low levels and can verify in this way whether the X-ray
doses are seen with cerebral angiography (5 mSv) doses for typical examinations in medical centers are
and much higher doses for transjugular intrahep- too high or too low. Particular attention is given to
atic portosystemic shunt procedures (TIPS) (70 mSv). screening examinations because they are performed
Compare this to the dose equivalent due to natural on asymptomatic people. In this regard, there is a lot of
31
sources, which is 2–3 mSv per year. experience in breast cancer screening programs, where
European Guidelines are widely applied. Special atten- Flat panel detectors for a large field of view and
tion should also be given to children and to high-dose with a fast readout capability will become available
imaging, such as interventional radiology. for 3D imaging. Hence, 2D projective imaging will
Furthermore, the ICRP recommends limiting all further be augmented by 3D volumetric imaging (see
exposed workers from regulated radiation practices to also Chapter 3).
20 mSv per year when averaged over five years and It can also be expected that the DQE of the
the public to 1 mSv per year. In particular, physicians detectors will continue to improve, yielding reduced
may receive a significant exposure when doing proce- radiation doses or images with enhanced contrast-to-
dures under fluoroscopy, but they too must not exceed noise ratio. Furthermore, photon counting detectors,
20 mSv per year. There are strict protection protocols which count the number of photons and measure
they have to follow, among which is the protection of their energy, will become commercially available by
the body and the thyroid gland with a lead apron and employing direct radiography with very fast readout
collar. A dosimeter, which is a small device clipped capability.
to the personnel’s clothing, measures the cumulative Currently all medical images can be fully integrated
absorbed dose. into the hospital information system. The images can
be interpreted on screen by the radiologist and elec-
tronically transmitted to the referring physician. It
Future expectations can be expected that manual interventions during the
Today, other imaging modalities, such as ultra- image acquisition process, such as cassette handling
sound, CT, and MRI, have largely replaced a number and parameter setting, will be further reduced. This
of X-ray examinations. Examples are arthrography will have a strong impact on the work flow in a med-
(joints), myelography (spinal cord), cholangiogra- ical imaging department. Furthermore, the computer
phy (bile ducts), cholecystography (gall bladder), and will behave as an intelligent assistant for the radiolo-
pyelography (urinary tract). Although radiography gist and will improve his/her performance. Computer
will remain an important imaging modality, this aided diagnosis (CAD) is discussed in more detail in
evolution can be expected to continue. Chapter 7.
32
Chapter
3 X-ray computed tomography
Introduction production, and interactions with tissue have already

X-ray computed tomography or CT (Figure 3.1) is been discussed in Chapter 2.
an imaging modality that produces cross-sectional The history of CT began in 1895, when Wilhelm
images representing the X-ray attenuation properties Konrad Röntgen reported the discovery of what he
of the body. The word tomography originates from the called “a new kind of rays.” Röntgen received the
Greek words τ oµoς (slice) and γραφιν (to write). first Nobel Prize in Physics in 1901. Reconstruction
Image formation of a cross-section is based on the fol- of a function from its projections was first formu-
lowing procedure. X-rays are produced by an X-ray lated by Johann Radon in 1917. Before the invention
tube, attenuated by the patient and measured by an X- of computed tomography, other kinds of tomography
ray detector. Using thin X-ray beams, a set of lines is existed.
scanned covering the entire field of view (Figure 3.2(a)
• Linear tomography (Figure 3.3(a)) The X-ray source
shows a parallel-beam geometry and Figure 3.2(b)
shows a fan-beam geometry). This process is repeated and film move at constant speed in opposite direc-
for a large number of angles (Figure 3.2(c) and (d)), tions. Under these circumstances, one section of
yielding line attenuation measurements for all pos- the patient (plane P1 –P2 ) is always projected at
sible angles and for all possible distances from the the same position on the film, whereas the rest
center. Based on all these measurements, the actual of the body is averaged out. In addition to linear
attenuation at each point of the scanned slice can tube and detector movement, curved paths (circu-
be reconstructed. Although the imaging modalities of lar, elliptical, hypocycloidal, . . .) have been used as
Chapters 4 and 5 (MR, PET, and SPECT) also rep- well.
resent a kind of computed tomography, the term CT • Axial transverse tomography (Figure 3.3(b)) The film
(originally CAT) is allocated for X-ray comput(eriz)ed is positioned horizontally in front of the patient
(axial) tomography. The physics of X-rays, their and slightly below the focal plane. Both the patient
(a) (b)
Figure 3.1 (a) Schematic representation, and (b) photograph of a CT scanner. (Courtesy of GE Healthcare.)
Chapter 3: X-ray computed tomography
and the film rotate at the same fixed speed around Robert S. Ledley in 1974. Since the introduction of
a vertical axis while the X-ray source remains sta- helical and multi-slice CT (respectively in 1989 and
tionary. Under these circumstances, the focal plane 1998), CT has opened the way to 3D images of the heart
in the patient remains in sharp focus throughout and has brought dynamic (4D) studies within reach.
the rotation, whereas all other planes are averaged In modern CT scanners, the images consist of
out. 512 × 512 pixels representing the CT number, which
is expressed in Hounsfield units (HU). The CT number
The first CT scanner (the EMI scanner) was devel- is defined as
oped by Godfrey N. Hounsfield in 1972. His work
was based on mathematical and experimental meth- µ − µH2 O
CT number (in HU) = · 1000, (3.1)
ods developed by A. M. Cormack a decade earlier. µH2 O
Hounsfield and Cormack shared the Nobel Prize in
Physiology or Medicine in 1979. The first whole-body where µ is the linear attenuation coefficient. With this
CT scanner (the ACTA scanner) was developed by definition, air and water have a CT number of, respec-
tively, −1000 HU and 0 HU. Bone falls on the positive
side of the scale, but has no unique CT number. This
value ranges from several hundreds to over 1000 HU.
The reason is that µ of bone (and all other tissues)
depends on its composition and structure, e.g., cortical
or trabecular, as well as on the energy of the absorbed
X-rays (see Figure 2.3).
Some clinical applications look at air–tissue or
tissue–bone contrasts on the order of 1000 HU, but
(a) (b) other clinical exams focus on small soft tissue contrasts
of a few HU. An optimal perception requires a suitable
gray level transformation. In clinical practice, this is
done by a real time window/level operation. The win-
dow and level respectively define the width and center
of the displayed gray level interval. Figure 3.4 shows an
example of a CT image of the chest with two different
window/level settings, the first to visualize the lungs
(a), and the second to emphasize the soft tissues (b).
X-ray detectors in CT
Energy integrating detectors
(c) (d)
Figure 3.2 Basic scanning procedure in CT. A set of lines is

scanned covering the entire field of view: (a) parallel-beam Most recent commercial CT detectors consist of a
geometry and (b) fan-beam geometry. This process is repeated for a scintillator crystal (CdWO4 , Y2 O3 , CsI, Gd2 O2 S)
large number of angles (c and d). in combination with a photodiode. The scintillator
Film Figure 3.3 (a) Linear tomography. The

X-ray source and film move at constant
speed in opposite directions. (b) Axial
transverse tomography. The film is
positioned horizontally in front of the
P1 P2 patient and slightly below the focal
plane. Both the patient and the film
rotate at the same fixed speed around a
vertical axis while the X-ray source
remains stationary.
x-ray tube
34
(a) (b)
Figure 3.4 CT image of the chest with

different window/level settings: (a) for the
lungs (window 1500 and level −500), and (b)
for the soft tissues (window 350 and level 50).
(a) (b)
material converts X-rays into visible light (scintilla- photodiode combination and the electronic noise no
tions), which then hits the photodiode, causing it longer dominates the signal from individual X-rays.
to produce an electric current. Individual scintilla- This difference allows an electronic circuit to detect
tor pieces are assembled into a reflector matrix in these charge packages and count the number of pho-
order to define the detector cells. The scintillators are tons. The fact that these detectors count the number of
produced with high optical quality so that the few photons instead of integrating their energy improves
millimeters thickness necessary to have a very high the CNR by 10 to 20%. First, carefully defining a
absorption efficiency (96%) also has good transfer of detection threshold eliminates the impact of electronic
light to the photodiode. Recent scintillators also offer noise. Second, the difference in attenuation between
a very fast response time (on the order of microsec- two tissues is generally larger for low-energy X-rays.
onds). Due to the finite thickness of the septa in Hence, the total CNR can be increased by assign-
the antiscatter grid, the absorption efficiency of the ing a higher weight to the detected low-energy X-ray
detector is limited by the area fill fraction, typically photons.
on the order of 80%. The multichannel readout elec- Yet, the main reason to consider photon count-
tronics or data acquisition system (DAS) connects to ing detectors is their ability to measure the amount
the photodiode. The DAS integrates the photocurrent of charge, and hence the energy, of the correspond-
from the diode and converts the electric charge sig- ing X-ray. The energy resolution can be much better
nal to voltage using a transimpedance amplifier. The than for example that of dual layer scintillator detec-
DAS also performs the analog to digital conversion tors, or even dual kV methods for dual-energy imaging
with typical sample rates on the order of a couple (see p. 21). Remaining challenges for commercial
of kilohertz. One limitation of these detectors is the introduction of direct conversion detectors for CT
susceptibility to electronic noise introduced by the applications include stability and the count rate limits,
transimpedance amplifier. For detectors using scin- and therefore it will be several years before scintillator-
tillator/photodiodes, electronic noise dominates the based detectors will be replaced on commercial CT
quantum noise at low signal levels, leading to noise scanners.
streaks (see p. 50) in the images.
Photon counting detectors Imaging

Recently photon counting detectors are receiving Data acquisition
increased attention for CT. They are based on direct Projection and Radon transform
conversion (see direct radiography, p. 21). A direct con- Consider the 2D parallel-beam geometry in Figure
version material such as cadmium telluride (CdTe) 3.5(a) in which µ(x, y) represents the distribution of
or cadmium-zinc-telluride (CZT) converts an X-ray the linear attenuation coefficient in the xy-plane. It
photon into a certain electronic charge proportional to is assumed that the patient lies along the z-axis and 35
its energy. The charge produced in direct conversion that µ(x, y) is zero outside a circular field of view
is about ten times that produced by the scintillator/ with diameter FOV. The X-ray beams make an angle
Iu(r) Figure 3.5 (a) Parallel-beam geometry

with coordinate systems. The X-ray
y beams make an angle θ with the y-axis
s u I0
␮(x,y) and are at distance r from the origin. (b)
An intensity profile Iθ (r) is measured for
every view (defined by an angle θ ). I0 is
r the unattenuated intensity. (c) The
I0 attenuation profiles pθ (r), obtained by
log-converting the intensity profiles Iθ (r),
r are the projections of the function µ(x, y)
u
x along the angle θ .
FOV
(b)
pu(r)
Iu(r)
(a) (c)
θ with the y-axis. The unattenuated intensity of the However, in practice it is typically assumed that the
X-ray beams is I0 . A new coordinate system (r, s) is X-rays are monochromatic, and Eq. (3.3) is used as an
defined by rotating (x, y) over the angle θ. This gives approximation.∗
the following transformation formulas: Each intensity profile is transformed into an atten-
uation profile:

r cos θ sin θ x
= Iθ (r)
s −sin θ cos θ y pθ (r) = −ln
(3.2) I0
x cos θ −sin θ r
= . = µ(r · cos θ − s · sin θ, r · sin θ + s · cos θ ) ds,
y sin θ cos θ s Lr,θ
(3.5)
For a fixed angle θ, the measured intensity profile as a
function of r is shown in Figure 3.5(b) and is given by where pθ (r) is the projection of the function µ(x, y)
along the angle θ (Figure 3.5(c)). Note that pθ (r) is
− µ(x,y) ds zero for |r| ≥ FOV/2.
Iθ (r) = I0 · e Lr,θ
pθ (r) can be measured for θ ranging from 0 to

− µ(r·cos θ−s·sin θ ,r·sin θ +s·cos θ) ds
= I0 · e Lr,θ
, 2π . Because concurrent beams coming from oppo-
(3.3) site sides theoretically yield identical measurements,
attenuation profiles acquired at opposite sides contain
redundant information. Therefore, as far as parallel-
where Lr,θ is the line that makes an angle θ with the
beam geometry is concerned, it is sufficient to measure
y-axis at distance r from the origin. Actually, the spec-
pθ (r) for θ ranging from 0 to π .
trum of the X-ray tube and the attenuation depend on
Stacking all these projections pθ (r) results in a
the energy, yielding (see Eq. (2.7))
2D dataset p(r, θ) called a sinogram (see Figure 3.6).
Assume a distribution µ(x, y) containing a single
Iθ (r) dot, as in Figure 3.7(a) and (b). The corresponding
∞
− µ(E,r·cos θ−s·sin θ,r·sin θ+s·cos θ) ds
= σ (E) · e Lr,θ
dE.
36 ∗ Dual-energy CT is a recent development and is introduced
0
(3.4) on p. 48.
Exploring the Variety of Random
Documents with Different Content
Taking his arm, Julie guided him to his feet.
"Look, dear," she said, "couldn't you drive back to the country with
me? A few days vacation wouldn't hurt too much, surely. You'd like
to, wouldn't you?"
"I'd love to," Marc said suddenly. He took her hand in his. "Let's go."
"You poor dear," Julie murmured. "I wonder how you stood it, with
everyone saying such awful things about you when you really hadn't
done anything at all."
Together, they left the court and started down the walk toward the
convertible.
As they left the city and started into the country, Marc pulled the car
over to the side of the highway and gave his attention to the drama
of the brilliant sunset.
"Well," he sighed, "there it goes, the first day of spring."
"Thank heavens," Julie said. "Now we can relax and enjoy it."
But there was still a question nagging at the back of Marc's mind.
"I was just thinking, dear," he said, "about your birthday...."
"Birthday!" Julie said. "But that's months away yet!"
"But, still," Marc said, "I was wondering what you'd like for a gift. I
thought maybe some nice pink lace underwear...."
"Pink lace underwear!" Julie said. She began to laugh.
"What's so funny?" Marc asked suspiciously.
"Darling," Julie said, "don't you remember the pink lace underwear
mother gave me for Christmas and how I loathed it? Well, I brought
it to the country where it wouldn't matter just so I could wear it out
and get rid of it."
Marc's relief came to the surface in a smile. "Then pink lace is out,
huh?"
"Definitely," Julie said. "But if you insist on lingerie, get me
something wicked and black. No true siren would ever dream of
letting herself be caught in pink."
Marc reached across the seat and drew her close to him. "In the
spring time," he said, "a young man's likely to get fancy."
The sun, on the horizon, slid conveniently out of sight and was gone.
As it did, a breeze blew lightly through the car and somewhere, it
seemed, there was laughter.
*** END OF THE PROJECT GUTENBERG EBOOK THE LAUGHTER OF
TOFFEE ***
Updated editions will replace the previous one—the old editions will
be renamed.
Creating the works from print editions not protected by U.S.

copyright law means that no one owns a United States copyright in
these works, so the Foundation (and you!) can copy and distribute it
in the United States without permission and without paying
copyright royalties. Special rules, set forth in the General Terms of
Use part of this license, apply to copying and distributing Project
Gutenberg™ electronic works to protect the PROJECT GUTENBERG™
concept and trademark. Project Gutenberg is a registered trademark,
and may not be used if you charge for an eBook, except by following
the terms of the trademark license, including paying royalties for use
of the Project Gutenberg trademark. If you do not charge anything
for copies of this eBook, complying with the trademark license is
very easy. You may use this eBook for nearly any purpose such as
creation of derivative works, reports, performances and research.
Project Gutenberg eBooks may be modified and printed and given
away—you may do practically ANYTHING in the United States with
eBooks not protected by U.S. copyright law. Redistribution is subject
to the trademark license, especially commercial redistribution.
START: FULL LICENSE

THE FULL PROJECT GUTENBERG LICENSE
PLEASE READ THIS BEFORE YOU DISTRIBUTE OR USE THIS WORK
To protect the Project Gutenberg™ mission of promoting the free

distribution of electronic works, by using or distributing this work (or
any other work associated in any way with the phrase “Project
Gutenberg”), you agree to comply with all the terms of the Full
Project Gutenberg™ License available with this file or online at
www.gutenberg.org/license.
Section 1. General Terms of Use and

Redistributing Project Gutenberg™
electronic works
1.A. By reading or using any part of this Project Gutenberg™
electronic work, you indicate that you have read, understand, agree
to and accept all the terms of this license and intellectual property
(trademark/copyright) agreement. If you do not agree to abide by all
the terms of this agreement, you must cease using and return or
destroy all copies of Project Gutenberg™ electronic works in your
possession. If you paid a fee for obtaining a copy of or access to a
Project Gutenberg™ electronic work and you do not agree to be
bound by the terms of this agreement, you may obtain a refund
from the person or entity to whom you paid the fee as set forth in
paragraph 1.E.8.
1.B. “Project Gutenberg” is a registered trademark. It may only be

used on or associated in any way with an electronic work by people
who agree to be bound by the terms of this agreement. There are a
few things that you can do with most Project Gutenberg™ electronic
works even without complying with the full terms of this agreement.
See paragraph 1.C below. There are a lot of things you can do with
Project Gutenberg™ electronic works if you follow the terms of this
agreement and help preserve free future access to Project
Gutenberg™ electronic works. See paragraph 1.E below.
1.C. The Project Gutenberg Literary Archive Foundation (“the
Foundation” or PGLAF), owns a compilation copyright in the
collection of Project Gutenberg™ electronic works. Nearly all the
individual works in the collection are in the public domain in the
United States. If an individual work is unprotected by copyright law
in the United States and you are located in the United States, we do
not claim a right to prevent you from copying, distributing,
performing, displaying or creating derivative works based on the
work as long as all references to Project Gutenberg are removed. Of
course, we hope that you will support the Project Gutenberg™
mission of promoting free access to electronic works by freely
sharing Project Gutenberg™ works in compliance with the terms of
this agreement for keeping the Project Gutenberg™ name associated
with the work. You can easily comply with the terms of this
agreement by keeping this work in the same format with its attached
full Project Gutenberg™ License when you share it without charge
with others.
1.D. The copyright laws of the place where you are located also
govern what you can do with this work. Copyright laws in most
countries are in a constant state of change. If you are outside the
United States, check the laws of your country in addition to the
terms of this agreement before downloading, copying, displaying,
performing, distributing or creating derivative works based on this
work or any other Project Gutenberg™ work. The Foundation makes
no representations concerning the copyright status of any work in
any country other than the United States.
1.E. Unless you have removed all references to Project Gutenberg:
1.E.1. The following sentence, with active links to, or other

immediate access to, the full Project Gutenberg™ License must
appear prominently whenever any copy of a Project Gutenberg™
work (any work on which the phrase “Project Gutenberg” appears,
or with which the phrase “Project Gutenberg” is associated) is
accessed, displayed, performed, viewed, copied or distributed:
This eBook is for the use of anyone anywhere in the United
States and most other parts of the world at no cost and with
almost no restrictions whatsoever. You may copy it, give it away
or re-use it under the terms of the Project Gutenberg License
included with this eBook or online at www.gutenberg.org. If you
are not located in the United States, you will have to check the
laws of the country where you are located before using this
eBook.
1.E.2. If an individual Project Gutenberg™ electronic work is derived

from texts not protected by U.S. copyright law (does not contain a
notice indicating that it is posted with permission of the copyright
holder), the work can be copied and distributed to anyone in the
United States without paying any fees or charges. If you are
redistributing or providing access to a work with the phrase “Project
Gutenberg” associated with or appearing on the work, you must
comply either with the requirements of paragraphs 1.E.1 through
1.E.7 or obtain permission for the use of the work and the Project
Gutenberg™ trademark as set forth in paragraphs 1.E.8 or 1.E.9.
1.E.3. If an individual Project Gutenberg™ electronic work is posted

with the permission of the copyright holder, your use and distribution
must comply with both paragraphs 1.E.1 through 1.E.7 and any
additional terms imposed by the copyright holder. Additional terms
will be linked to the Project Gutenberg™ License for all works posted
with the permission of the copyright holder found at the beginning
of this work.
1.E.4. Do not unlink or detach or remove the full Project

Gutenberg™ License terms from this work, or any files containing a
part of this work or any other work associated with Project
Gutenberg™.
1.E.5. Do not copy, display, perform, distribute or redistribute this

electronic work, or any part of this electronic work, without
prominently displaying the sentence set forth in paragraph 1.E.1
with active links or immediate access to the full terms of the Project
Gutenberg™ License.
1.E.6. You may convert to and distribute this work in any binary,
compressed, marked up, nonproprietary or proprietary form,
including any word processing or hypertext form. However, if you
provide access to or distribute copies of a Project Gutenberg™ work
in a format other than “Plain Vanilla ASCII” or other format used in
the official version posted on the official Project Gutenberg™ website
(www.gutenberg.org), you must, at no additional cost, fee or
expense to the user, provide a copy, a means of exporting a copy, or
a means of obtaining a copy upon request, of the work in its original
“Plain Vanilla ASCII” or other form. Any alternate format must
include the full Project Gutenberg™ License as specified in
paragraph 1.E.1.
1.E.7. Do not charge a fee for access to, viewing, displaying,

performing, copying or distributing any Project Gutenberg™ works
unless you comply with paragraph 1.E.8 or 1.E.9.
1.E.8. You may charge a reasonable fee for copies of or providing

access to or distributing Project Gutenberg™ electronic works
provided that:
• You pay a royalty fee of 20% of the gross profits you derive
from the use of Project Gutenberg™ works calculated using the
method you already use to calculate your applicable taxes. The
fee is owed to the owner of the Project Gutenberg™ trademark,
but he has agreed to donate royalties under this paragraph to
the Project Gutenberg Literary Archive Foundation. Royalty
payments must be paid within 60 days following each date on
which you prepare (or are legally required to prepare) your
periodic tax returns. Royalty payments should be clearly marked
as such and sent to the Project Gutenberg Literary Archive
Foundation at the address specified in Section 4, “Information
about donations to the Project Gutenberg Literary Archive
Foundation.”
• You provide a full refund of any money paid by a user who

notifies you in writing (or by e-mail) within 30 days of receipt
that s/he does not agree to the terms of the full Project
Gutenberg™ License. You must require such a user to return or
destroy all copies of the works possessed in a physical medium
and discontinue all use of and all access to other copies of
Project Gutenberg™ works.
• You provide, in accordance with paragraph 1.F.3, a full refund of

any money paid for a work or a replacement copy, if a defect in
the electronic work is discovered and reported to you within 90
days of receipt of the work.
• You comply with all other terms of this agreement for free
distribution of Project Gutenberg™ works.
1.E.9. If you wish to charge a fee or distribute a Project Gutenberg™

electronic work or group of works on different terms than are set
forth in this agreement, you must obtain permission in writing from
the Project Gutenberg Literary Archive Foundation, the manager of
the Project Gutenberg™ trademark. Contact the Foundation as set
forth in Section 3 below.
1.F.
1.F.1. Project Gutenberg volunteers and employees expend

considerable effort to identify, do copyright research on, transcribe
and proofread works not protected by U.S. copyright law in creating
the Project Gutenberg™ collection. Despite these efforts, Project
Gutenberg™ electronic works, and the medium on which they may
be stored, may contain “Defects,” such as, but not limited to,
incomplete, inaccurate or corrupt data, transcription errors, a
copyright or other intellectual property infringement, a defective or
damaged disk or other medium, a computer virus, or computer
codes that damage or cannot be read by your equipment.
1.F.2. LIMITED WARRANTY, DISCLAIMER OF DAMAGES - Except for

the “Right of Replacement or Refund” described in paragraph 1.F.3,
the Project Gutenberg Literary Archive Foundation, the owner of the
Project Gutenberg™ trademark, and any other party distributing a
Project Gutenberg™ electronic work under this agreement, disclaim
all liability to you for damages, costs and expenses, including legal
fees. YOU AGREE THAT YOU HAVE NO REMEDIES FOR
NEGLIGENCE, STRICT LIABILITY, BREACH OF WARRANTY OR
BREACH OF CONTRACT EXCEPT THOSE PROVIDED IN PARAGRAPH
1.F.3. YOU AGREE THAT THE FOUNDATION, THE TRADEMARK
OWNER, AND ANY DISTRIBUTOR UNDER THIS AGREEMENT WILL
NOT BE LIABLE TO YOU FOR ACTUAL, DIRECT, INDIRECT,
CONSEQUENTIAL, PUNITIVE OR INCIDENTAL DAMAGES EVEN IF
YOU GIVE NOTICE OF THE POSSIBILITY OF SUCH DAMAGE.
1.F.3. LIMITED RIGHT OF REPLACEMENT OR REFUND - If you

discover a defect in this electronic work within 90 days of receiving
it, you can receive a refund of the money (if any) you paid for it by
sending a written explanation to the person you received the work
from. If you received the work on a physical medium, you must
return the medium with your written explanation. The person or
entity that provided you with the defective work may elect to provide
a replacement copy in lieu of a refund. If you received the work
electronically, the person or entity providing it to you may choose to
give you a second opportunity to receive the work electronically in
lieu of a refund. If the second copy is also defective, you may
demand a refund in writing without further opportunities to fix the
problem.
1.F.4. Except for the limited right of replacement or refund set forth
in paragraph 1.F.3, this work is provided to you ‘AS-IS’, WITH NO
OTHER WARRANTIES OF ANY KIND, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR ANY PURPOSE.
1.F.5. Some states do not allow disclaimers of certain implied

warranties or the exclusion or limitation of certain types of damages.
If any disclaimer or limitation set forth in this agreement violates the
law of the state applicable to this agreement, the agreement shall be
interpreted to make the maximum disclaimer or limitation permitted
by the applicable state law. The invalidity or unenforceability of any
provision of this agreement shall not void the remaining provisions.
1.F.6. INDEMNITY - You agree to indemnify and hold the Foundation,

the trademark owner, any agent or employee of the Foundation,
anyone providing copies of Project Gutenberg™ electronic works in
accordance with this agreement, and any volunteers associated with
the production, promotion and distribution of Project Gutenberg™
electronic works, harmless from all liability, costs and expenses,
including legal fees, that arise directly or indirectly from any of the
following which you do or cause to occur: (a) distribution of this or
any Project Gutenberg™ work, (b) alteration, modification, or
additions or deletions to any Project Gutenberg™ work, and (c) any
Defect you cause.
Section 2. Information about the Mission

of Project Gutenberg™
Project Gutenberg™ is synonymous with the free distribution of
electronic works in formats readable by the widest variety of
computers including obsolete, old, middle-aged and new computers.
It exists because of the efforts of hundreds of volunteers and
donations from people in all walks of life.
Volunteers and financial support to provide volunteers with the

assistance they need are critical to reaching Project Gutenberg™’s
goals and ensuring that the Project Gutenberg™ collection will
remain freely available for generations to come. In 2001, the Project
Gutenberg Literary Archive Foundation was created to provide a
secure and permanent future for Project Gutenberg™ and future
generations. To learn more about the Project Gutenberg Literary
Archive Foundation and how your efforts and donations can help,
see Sections 3 and 4 and the Foundation information page at
www.gutenberg.org.
Section 3. Information about the Project

Gutenberg Literary Archive Foundation
The Project Gutenberg Literary Archive Foundation is a non-profit
501(c)(3) educational corporation organized under the laws of the
state of Mississippi and granted tax exempt status by the Internal
Revenue Service. The Foundation’s EIN or federal tax identification
number is 64-6221541. Contributions to the Project Gutenberg
Literary Archive Foundation are tax deductible to the full extent
permitted by U.S. federal laws and your state’s laws.
The Foundation’s business office is located at 809 North 1500 West,

Salt Lake City, UT 84116, (801) 596-1887. Email contact links and up
to date contact information can be found at the Foundation’s website
and official page at www.gutenberg.org/contact
Section 4. Information about Donations to

the Project Gutenberg Literary Archive
Foundation
Project Gutenberg™ depends upon and cannot survive without
widespread public support and donations to carry out its mission of
increasing the number of public domain and licensed works that can
be freely distributed in machine-readable form accessible by the
widest array of equipment including outdated equipment. Many
small donations ($1 to $5,000) are particularly important to
maintaining tax exempt status with the IRS.
The Foundation is committed to complying with the laws regulating

charities and charitable donations in all 50 states of the United
States. Compliance requirements are not uniform and it takes a
considerable effort, much paperwork and many fees to meet and
keep up with these requirements. We do not solicit donations in
locations where we have not received written confirmation of
compliance. To SEND DONATIONS or determine the status of
compliance for any particular state visit www.gutenberg.org/donate.
While we cannot and do not solicit contributions from states where

we have not met the solicitation requirements, we know of no
prohibition against accepting unsolicited donations from donors in
such states who approach us with offers to donate.
International donations are gratefully accepted, but we cannot make

any statements concerning tax treatment of donations received from
outside the United States. U.S. laws alone swamp our small staff.
Please check the Project Gutenberg web pages for current donation
methods and addresses. Donations are accepted in a number of
other ways including checks, online payments and credit card
donations. To donate, please visit: www.gutenberg.org/donate.
Section 5. General Information About

Project Gutenberg™ electronic works
Professor Michael S. Hart was the originator of the Project
Gutenberg™ concept of a library of electronic works that could be
freely shared with anyone. For forty years, he produced and
distributed Project Gutenberg™ eBooks with only a loose network of
volunteer support.
Project Gutenberg™ eBooks are often created from several printed
editions, all of which are confirmed as not protected by copyright in
the U.S. unless a copyright notice is included. Thus, we do not
necessarily keep eBooks in compliance with any particular paper
edition.
Most people start at our website which has the main PG search
facility: www.gutenberg.org.
This website includes information about Project Gutenberg™,

including how to make donations to the Project Gutenberg Literary
Archive Foundation, how to help produce our new eBooks, and how
to subscribe to our email newsletter to hear about new eBooks.

51955

Uploaded by

Copyright:

Available Formats

51955

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

51955

Uploaded by

Copyright:

Available Formats

Download the full version of the ebook at

Fundamentals of Medical Imaging 2nd Edition

Explore and download more ebook at https://ebookultra.com

Fundamentals of Digital Imaging H. J. Trussell

Imaging Dopamine 1st Edition Paul Cumming

Image Radiation Biology of Medical Imaging 1st Edition

Fundamentals of Musculoskeletal Imaging 3rd Edition

Imaging systems for medical diagnostics fundamentals

Ultrasound Imaging and Therapy Imaging in Medical

Fundamentals of Mathematical Analysis 1st Edition Paul J.

Medical Imaging Based on Magnetic Fields and Ultrasounds

Cambridge University Press

ISBN-13 978-0-511-59640-7 eBook (NetLibrary)

ISBN-13 978-0-521-51915-1 Hardback

Cambridge University Press has no responsibility for the persistence or accuracy

Preface page vii Image quality 90

Geometric model matching using a Virtual reality 205

8 Visualization for diagnosis

1 Introduction to digital image processing

Figure 1.3 The same image quantized

Figure 1.4 (a) Sharp bright spot on a

as it actually is. It will be smoothed, and the obtained |OTF|

Figure 1.8 Original CT image (a) with

Figure 1.9 (a) Magnetic resonance

(a) (b) (c)

Geometric operations and rotation as special cases. Afﬁne transforma-

= f ∗ I (i, j), (1.13)

Figure 1.12 (a) Radiography of the skull.

(a) (b) (c)

z Figure 1.13 (a) A Gaussian function.

Figure 1.14 Radiography of a hand. (a)

(a) (b) (c)

Figure 1.16 The effect of the filter size

Introduction consequently, the corresponding photon energies are

where q is the electric charge of an electron. For

2 –1 –2 –3 –4 –5 –6 –7 –8 –9 –10 –11 –12 –13

Photon energy (eV)

Figure 2.1 The electromagnetic spectrum.

X-ray beam intensity

lead shield absorbing most X-rays Kb

Interaction with matter Interaction of an X-ray beam with tissue

(a) (b) µ, cm–1

∞ xout µm (10 keV, H2 O) = 5 cm2 /g

Screen–ﬁlm detector stopped. If the delay to reach peak emission is

For X-ray imaging with a screen–ﬁlm combina- D

screen and the spectral sensitivity of the ﬁlm used. 1.5

• Contrast The most widely used description of the

Input Figure 2.5 Scheme of an image

High Voltage Power Supply

the form of light. The temporarily stored energy

0.70 µS (E) = a1 · µ1 (E) + a2 · µ2 (E). (2.13)

µ(E) ≈ µp (E) + µC (E) Eq. (2.14) can be written as

tively. The exponent m is an empirically deﬁned (2.16)

attenuation coefﬁcient of an arbitrary substance S can ILE (x, y)

Figure 2.10 shows a general purpose radiographic Clinical use

• Barium ﬂuoroscopy of the gastrointestinal tract

Biologic effects and safety

They are injuries to a large population of cells where

To assess the overall radiation detriment † from

where HT is the equivalent dose in tissue or organ