Multislice CT, 4th Edition

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

https://medipdf.com/product/multislice-ct-4th-edition/

Click Download Now

Contents

Part I Techniques

Multi-slice CT: Current Technology and Future Developments . . 3

Stefan Ulzheimer, Malte Bongers, and Thomas Flohr
Radiation Dose Optimization in CT. . . . . . . . . . . . . . . . . . . . . . . . . 35
Shaunagh McDermott, Alexi Otrakji, and Mannudeep K. Kalra
Spectral CT/Dual-Energy CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Anushri Parakh, Manuel Patino, and Dushyant V. Sahani
Contrast Enhancement at CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Kazuo Awai, Toru Higaki, and Fuminari Tatsugami
Image Processing from 2D to 3D. . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Steven P. Rowe and Elliot K. Fishman
Perfusion CT: Technical Aspects. . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Vicky Goh and Davide Prezzi

Part II Neuro/ENT

Stroke/Cerebral Perfusion CT: Technique

and Clinical Applications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Adrienne Moraff, Jeremy Heit, and Max Wintermark
CT Diagnostics in Brain Tumors. . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Marco Essig
MDCT in Neurovascular Imaging. . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Giovanna Negrao de Figueiredo and Birgit Ertl-Wagner
Anatomy and Pathology of the Temporal Bone. . . . . . . . . . . . . . . . 207
Sabrina Kösling
Dental CT: Pathologic Findings in the Teeth and Jaws. . . . . . . . . . 217
Wolfgang Wuest
Anatomy and Corresponding Oncological Imaging
of the Head and Neck. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
Christian Czerny and Juergen Lutz

vii
viii Contents

Part III Chest

Interstitial Lung Diseases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261

Christina Mueller-Mang, Helmut Ringl, and Christian Herold
Pneumonia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
Sabine Dettmer and Jens Vogel-Claussen
CT of the Airways. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
Michael Trojan, Hans-Ulrich Kauczor, Claus Peter Heußel,
and Mark Oliver Wielpütz
Lung Cancer Screening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331
Oyunbileg von Stackelberg and Hans-Ulrich Kauczor
Chest Neoplasias. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345
Tina D. Tailor
CT of Pulmonary Embolism: Imaging Update . . . . . . . . . . . . . . . . 395
Antoine Hutt, Paul Felloni, Jacques Remy, and Martine
Remy-Jardin
COPD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409
Anna Rita Larici, Paola Franchi, Giuseppe Cicchetti,
and Lorenzo Bonomo

Part IV Abdomen

Focal Lesions in Non-cirrhotic Liver. . . . . . . . . . . . . . . . . . . . . . . . . 433

Christoph J. Zech
Cirrhotic Liver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447
Keitaro Sofue, Masakatsu Tsurusaki, and Takamichi Murakami
Pancreatic Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 491
Jeong Min Lee and Hyo-Jin Kang
Acute and Chronic Pancreatitis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 527
G. Zamboni, M. Chincarini, R. Negrelli,
and R. Pozzi Mucelli
Spleen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 555
Andre Euler and Sebastian T. Schindera
Imaging of the Stomach and Esophagus Using
CT and PET/CT Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 579
Ahmed Ba-Ssalamah, Sarah Poetter-Lang, Nina Bastati,
Jacqueline C. Hodge, Helmut Ringl, and Richard M. Gore
Small Bowel MDCT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 619
Marco Rengo, Simona Picchia, and Andrea Laghi
Imaging of Large Bowel with Multidetector Row CT. . . . . . . . . . . 641
Jay D. Patel, Heather I. Gale, and Kevin J. Chang
Contents ix

Peritoneal Surface Malignancy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 667

Davide Bellini, Paolo Sammartino, and Andrea Laghi
Multislice PET/CT in Neuroendocrine Tumors. . . . . . . . . . . . . . . . 675
Gabriele Pöpperl and Clemens Cyran
Adrenals. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 691
Christoph Schabel and Daniele Marin
Kidneys, Ureters, and Bladder. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 697
Christoph Schabel and Daniele Marin

Part V Cardiovascular

Technical Innovations and Concepts in Coronary CT . . . . . . . . . . 713

Nils Vogler, Mathias Meyer, and Thomas Henzler
Noninvasive Coronary Artery Imaging. . . . . . . . . . . . . . . . . . . . . . . 729
Manoj Mannil and Hatem Alkadhi
Pre- and Postinterventional/Surgical Evaluation by CT. . . . . . . . . 743
Harald Seifarth and David Maintz
Computed Tomography in the Management
of Electrophysiology Procedures. . . . . . . . . . . . . . . . . . . . . . . . . . . . 755
Joseph Negusei, Ian R. Drexler, Jim Cheung,
and Quynh A. Truong
Functional Cardiac CT Angiography. . . . . . . . . . . . . . . . . . . . . . . . 777
Domenico De Santis, Marwen Eid, Taylor M. Duguay,
U. Joseph Schoepf, and Carlo N. De Cecco
CT Angiography of the Peripheral Arteries. . . . . . . . . . . . . . . . . . . 805
Newton B. Neidert, Nikkole M. Weber, Jeffrey C. Hellinger,
and Eric E. Williamson
Acute Aortic Syndromes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 825
Christian Loewe
CT Venography. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 855
Simer Grewal, Patrick Sutphin, and Sanjeeva P. Kalva
Aortic Aneurysm and Stent Graft Assessment. . . . . . . . . . . . . . . . . 869
Ilya Livshits, Safet Lekperic, and Robert Lookstein

Part VI Interventions

CT-Guided Biopsy and Drainage. . . . . . . . . . . . . . . . . . . . . . . . . . . . 893

Giovanna Negrão de Figueiredo and Christoph G. Trumm
CT-Guided Spinal Interventions:
Vertebroplasty/Kyphoplasty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 925
Tobias F. Jakobs and Stefanie C. Surwald
x Contents

CT-Guided Tumor Ablation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 945

Ralf-Thorsten Hoffmann
Functional CT for Image-Guided Personalized
Tumor Interventions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 957
Horger Marius

Part VII Pediatrics

Dedicated CT Protocols for Children. . . . . . . . . . . . . . . . . . . . . . . . 969

Ilias Tsiflikas
Congenital Heart Disease in Children. . . . . . . . . . . . . . . . . . . . . . . . 987
Aurelio Secinaro and Davide Curione
Chest CT Imaging in Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1011
Sebastian Ley and Julia Ley-Zaporozhan
CT of the Pediatric Abdomen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1037
Michael Riccabona and Alexander Pilhatsch

Part VIII Miscellaneous Topics

Emergency CT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1051

Samad Shah, Sunil Jeph, and Savvas Nicolaou
Clinical Application of Musculoskeletal CT:
Trauma, Oncology, and Postsurgery. . . . . . . . . . . . . . . . . . . . . . . . . 1079
Pedro Augusto Gondim Teixeira and Alain Blum
Incidental Findings in Multislice CT of the Body . . . . . . . . . . . . . . 1107
Mikael Hellström
Correction to: Chest Neoplasias . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1139
Tina D. Tailor
Correction to: Imaging of the Stomach and Esophagus
Using CT and PET/CT Techniques. . . . . . . . . . . . . . . . . . . . . . . . . . 1141
Ahmed Ba-Ssalamah, Sarah Poetter-Lang, Nina Bastati,
Jacqueline C. Hodge, Helmut Ringl, and Richard M. Gore
 Part I
Techniques
 Multi-slice CT: Current Technology
and Future Developments

Stefan Ulzheimer, Malte Bongers,

and Thomas Flohr

Contents Abstract
1    Introduction 3 Since its introduction in the early seventies of
the past century, computed tomography (CT)
2    System Design 6
2.1 Gantry 6 has undergone tremendous improvements in
2.2 X-Ray Tube and Generator 7 terms of technology, performance, and clinical
2.3 MDCT Detector Design and Slice applications. Based on the historic evolution of
Collimation 8 CT and basic CT physics this chapter describes
2.4 Dual Source CT 10
the status quo of the technology and tries to
3    Measurement Techniques 11 anticipate future developments. Besides the
3.1 MDCT Sequential (Axial) Scanning 11
3.2 MDCT Spiral (Helical) Scanning 12
description of key components of CT systems,
3.3 ECG-Synchronized Cardiovascular CT 17 a special focus is laid on breakthrough devel-
3.4 Dual Energy Computed Tomography 21 opments such as multi-slice CT and dedicated
4    Radiation Dose Reduction 23 scan modes for cardiac imaging.
4.1 Anatomical Tube Current Modulation 23
4.2 Adaptation of the X-Ray Tube Voltage 24
4.3 Spectral Shaping 25
4.4 Iterative Reconstruction 26
1 Introduction
5    Future Developments 27 In 1972, the English engineer G.N. Hounsfield
References 30 built the first commercial medical X-ray com-
puted tomography (CT) scanner for the company
EMI Ltd. as a pure head scanner with an X-ray
S. Ulzheimer
Siemens Healthcare GmbH, Computed Tomography, tube and two detector elements moving incre-
Forchheim, Germany mentally around the patient. It was able to acquire
e-mail: [email protected] twelve slices with 13 mm slice thickness each
M. Bongers (*) and reconstruct the images with a matrix of
Institute for Diagnostic and Interventional Radiology, 80 × 80 pixels (Fig. 1a) in approximately 35 min.
Eberhard-Karls-University, Tübingen, Germany
Even though the performance of CT scanners
e-mail: [email protected]
increased dramatically over time there were no
T. Flohr
principally new developments in conventional
Siemens Healthcare GmbH, Computed Tomography,
Forchheim, Germany CT until 1989. By then the acquisition time for
one image decreased from 300 s in 1972 to 1–2 s,
Institute for Diagnostic and Interventional Radiology,
Eberhard-Karls-University, Tübingen, Germany thin slices of down to 1 mm became possible and
e-mail: [email protected] the in-plane resolution increased from 3 line pairs

Med Radiol Diagn Imaging (2018) 3

https://doi.org/10.1007/174_2018_187, © Springer International Publishing AG
Published Online: 07 July 2018
4 S. Ulzheimer et al.

Fig. 1 Development of computed tomography over time. sectional slices in the year 2007 (Image courtesy of Mayo
(a) Cross-sectional image of a brain in the year 1971 and Clinic Rochester)
(b) the whole brain with sagittal, coronal, and cross-­

per cm (lp/cm) to 15 lp/cm with typically 5122 overlapping image reconstruction could be used
matrices. to improve through-plane resolution. Volume
As acquisition times of mechanical CT scan- data became the very basis for applications such
ners were expected to be far too long for high as CT angiography (CTA) (Rubin et al. 1995),
quality cardiac imaging for the next years or even which has revolutionized noninvasive assessment
decades to come, a completely new technical of vascular disease. The ability to acquire volume
concept for a CT scanner without moving parts data was the prerequisite for the development of
for extremely fast data acquisition within 50 ms three-dimensional image processing techniques
was suggested and promoted as cardiovascular such as multiplanar reformations (MPR), maxi-
CT (CVCT) scanner. These scanners were also mum intensity projections (MIP), surface shaded
called “Ultrafast CT” scanners or “Electron displays (SSD), or volume rendering techniques
Beam CT” (EBT or EBCT) scanners. High cost (VRT), which have become a vital component of
and limited image quality combined with low medical imaging today.
volume coverage prevented the wide propagation Main drawbacks of single-slice spiral CT are
of the modality, and the production and distribu- either insufficient volume coverage within one
tion of these scanners were soon discontinued. breath-hold time of the patient or missing spatial
Based on the introduction of slip ring technol- resolution in z-axis due to wide collimation. With
ogy to get power to and data off the rotating gan- single-slice spiral CT the ideal isotropic resolu-
try, continuous rotation of the X-ray tube and the tion, i.e., of equal resolution in all three spatial
detector became possible. The ability of continu- axes, can only be achieved for very limited scan
ous rotation led to the development of spiral CT ranges (Kalender 1995).
scanners in the early nineties (Kalender et al. Larger volume coverage in shorter scan times
1990; Crawford and King 1990). Volume data and improved through-plane resolution became
could be acquired without the danger of mis- or feasible after the broad introduction of 4-slice CT
double-registration of anatomical details. Images systems by all major CT manufacturers in 1998
could be reconstructed at any position along the (Klingenbeck-Regn et al. 1999; Mccollough and
patient axis (through-plane axis, z-axis), and Zink 1999; Hu et al. 2000). The increased perfor-
Multi-slice CT: Current Technology and Future Developments 5

mance allowed for the optimization of a variety collimated slice width, and further reduced rota-
of clinically relevant scan parameters. tion times (down to 0.33 s). GE, Philips, and
Examination times for standard protocols could Toshiba aimed at an increase in volume coverage
be significantly reduced; alternatively, scan speed by using detectors with 64 rows instead of
ranges could be significantly extended. 16, thus providing 32–40 mm z-coverage.
Furthermore, a given anatomic volume could be Siemens used 32 physical detector rows in com-
scanned within a given scan time with substan- bination with double z-sampling, a refined
tially reduced slice width. This way, for many z-sampling technique enabled by a z-flying focal
clinical applications the goal of isotropic resolu- spot (see Sect. 3.2.4), to simultaneously acquire
tion was within reach with 4-slice CT systems. 64 overlapping 0.6 mm slices with the goal of
Multi-detector row CT (MDCT) also dramati- pitch-independent increase of through-plane res-
cally expanded into areas previously considered olution and reduction of spiral artifacts (Flohr
beyond the scope of third-generation CT scanners et al. 2004, 2005). With 64-slice CT systems, CT
based on the mechanical rotation of X-ray tube scans with submillimeter resolution became fea-
and detector, such as cardiac imaging with the sible even for extended anatomical ranges. The
addition of ECG gating capability enabled by improved temporal resolution due to faster gantry
gantry rotation times down to 0.5 s (Ohnesorge rotation increased clinical robustness of ECG-­
et al. 2000; Kachelriess et al. 2000). Despite all gated scanning, thereby facilitating the success-
these promising advances, clinical challenges and ful integration of CT coronary angiography into
limitations remained for 4-slice CT systems. True routine clinical algorithms (Leber et al. 2005;
isotropic had not yet been achieved for many rou- Leschka et al. 2005), although higher and irregu-
tine applications requiring extended scan ranges, lar heart rates were still problematic.
since wider collimated slices (4 × 2.5 mm or In 2007, one vendor introduced a MDCT sys-
4 × 3.75 mm) had to be chosen to complete the tem with 128 simultaneously acquired slices,
scan within a breath-hold time of the patient. For based on a 64-row detector with 0.6 mm colli-
ECG-gated coronary CTA, stents or severely cal- mated slice width (38.4 mm z-axis coverage) and
cified arteries constituted a diagnostic dilemma, double z-sampling by means of a z-flying focal
mainly due to partial volume artifacts as a conse- spot. Later, simultaneous acquisition of 256
quence of insufficient through-­ plane resolution slices became available with a CT system
(Nieman et al. 2001), and reliable imaging of equipped with a 128-row detector (0.625 mm
patients with higher heart rates was not possible collimated slice width, 80 mm z-axis coverage)
due to limited temporal resolution. and double z-sampling.
As a next step, the introduction of an 8-slice Clinical experience with 64-, 128-, or 256-­
CT system in 2000 enabled shorter scan times, slice CT indicated that adding even more detector
but did not yet provide improved through-plane rows would not by itself translate into increased
resolution (thinnest collimation 8 × 1.25 mm). clinical benefit. Instead, new CT concepts were
This was achieved with the introduction of introduced to solve remaining limitations of
16-slice CT (Flohr et al. 2002a, b), which made it MDCT.
possible to routinely acquire substantial anatomic One remaining challenge for MDCT is the
volumes with isotropic submillimeter spatial res- visualization of dynamic processes in extended
olution. ECG-gated cardiac scanning was anatomical ranges, e.g., to characterize the inflow
enhanced by both improved temporal resolution and outflow of contrast agent in the arterial and
achieved by gantry rotation time down to 0.375 s venous system in dynamic CTAs, or to determine
and improved spatial resolution (Nieman et al. the enhancement characteristics of the contrast
2002; Ropers et al. 2003). agent in volume perfusion studies. Dynamic CT
In 2004, all major CT manufacturers intro- examinations are enabled by CT systems with
duced MDCT systems with simultaneous acqui- area detectors large enough to cover entire
sition of 64-slices at 0.5, 0.6 mm, or 0.625 mm organs, such as the heart, the kidneys, or the
6 S. Ulzheimer et al.

brain, in one axial scan (requiring 120 mm vol- 2 System Design

ume coverage or more). Another way to acquire
dynamic volume data is the introduction of “shut- The overall performance of a MDCT system
tle modes” with periodic table movement between depends on several key components. These com-
two z-positions (e.g., Goetti et al. 2010). ponents include the gantry, X-ray source, a high-­
In 2007, a CT scanner with 16 cm z-axis cov- powered generator, detector and detector
erage at isocenter was introduced which has the electronics, data transmission systems (slip-­
potential to provide dynamic volume data with- rings), and the computer system for image recon-
out table movement. Currently, two CT scanners struction and manipulation.
following this design principle are commercially
available by different vendors, one with
320 × 0.5 mm collimation and 0.27 s gantry rota- 2.1 Gantry
tion time, and the other with 256 × 0.625 mm col-
limation and 0.28 s gantry rotation time. CT Third-generation CT scanners employ the so-­called
systems with 16 cm detector coverage can also “rotate/rotate” geometry, in which both X-ray tube
scan the entire heart in one axial scan, thereby and detector are mounted onto a rotating gantry
avoiding the typical stair-step artifacts in cardiac and rotate around the patient (Fig. 2). A CT detector
CT images which are a potential problem in CT has 700 or more individual detector elements in the
systems with smaller detectors (Rybicki et al. fan angle direction (see Fig. 2) which cover a scan
2008; Steigner et al. 2009; Dewey et al. 2009).
Motion artifacts due to insufficient temporal
resolution remain an important challenge for
cardiothoracic imaging and coronary CTA even
with the latest generation of MDCT. Image tem-
poral resolution of less than 100 ms at all heart
rates can be achieved with dual source CT
(DSCT) systems, i.e., CT scanners with two
X-ray tubes and two corresponding detectors
offset by 90° (Flohr et al. 2006). Meanwhile,
three generations of DSCT systems have been
commercially introduced, and clinical studies
have demonstrated the potential of DSCT to
reliably perform coronary CTA with little or no
dependence on the patient’s heart rate (Scheffel
et al. 2006; Matt et al. 2007; Ropers et al. 2007;
Weustink et al. 2009). DSCT scanners also show
promising properties for general radiology
applications. Both X-ray tubes can be operated
Fig. 2 Basic system components of a modern “third-­
simultaneously in standard acquisitions to pro-
generation” CT system. First-generation systems used a
vide high power reserves when necessary. collimated pencil beam and therefore required a transla-
Additionally, both X-ray tubes can be operated tion of the pencil beam and the single detector element
at different kV settings and/or different pre-fil- before each rotational step to scan the whole object.
Second-generation scanner used a small fan beam but still
trations, enabling dual energy CT. Since the
required translational and rotational movement patterns of
introduction of DSCT, tissue characterization, the X-ray source and the small detector array. The fan
calcium quantification, and quantification of the beam of third-generation scanners covers a SFOV of typi-
local blood volume in contrast-enhanced scans cally 50 cm diameter and allows for a pure rotational
motion of the tube and the detector around the patient.
have been investigated as potential dual energy
This was the key to reduce scan times per image from
applications (Johnson et al. 2007). minutes to less than a second. All medical CT scanners
today are third-generation scanners
Multi-slice CT: Current Technology and Future Developments 7

field of view (SFOV) of usually 50 cm diameter. In components on it, such as X-ray tube, tube colli-
a MDCT system, the detector comprises several mator, and data measurement system (DMS), have
detector rows in the z-axis direction (patient direc- to be designed to withstand the high gravitational
tion). The X-ray attenuation of the object is mea- forces associated with fast gantry rotation.
sured by the individual detector elements. All
measurement values acquired at the same angular
position of the measurement system form a “pro- 2.2 X-Ray Tube and Generator
jection” or “view.” About 1000 projections are
measured during each 360° rotation. State-of-the-art X-ray tube/generator combina-
Key requirement for the mechanical design of tions provide a peak power of 60–120 kW, usu-
the gantry is the stability of both focal spot and ally at various, user-selectable voltages, e.g.,
detector position during rotation, in particular with 70–140 kV in steps of 10 kV. In a conventional
regard to the rapidly increasing rotational speeds tube design, an anode plate of typically 160–
of modern CT systems (from 0.75 s in 1994 to 220 mm diameter rotates in a vacuum housing
0.25 s in 2017). Hence, the gantry as well as all the (Fig. 3). The heat storage capacity of anode

Anode
Cooling oil

X-rays

Anode
Deflection unit

e-beam
Cathode

X-rays

Fig. 3 Schematic drawings and pictures of a conven- envelope tube, the anode plate constitutes an outer wall of
tional X-ray tube (top) and a rotating envelope tube (bot- the tube housing and is in direct contact with the cooling
tom). The electrons emitted by the cathode are represented oil. Heat is more efficiently dissipated via thermal con-
by green lines, the X-rays generated in the anode are duction, and the cooling rate is significantly increased.
depicted as purple arrows. In a conventional X-ray tube Rotating envelope tubes have no moving parts and no
the anode plate rotates in a vacuum housing. Heat is bearings in the vacuum. (Images not to scale)
mainly dissipated via thermal radiation. In a rotating
8 S. Ulzheimer et al.

plate and tube housing—measured in Mega 2.3  DCT Detector Design

M
Heat Units (MHU)—determines the perfor- and Slice Collimation
mance level: the bigger the anode plate is, the
larger is the heat storage capacity, and the more All commercially available MDCT systems to-­
scan-seconds can be delivered until the anode date are equipped with solid-state scintillation
plate reaches its temperature limit. An alterna- detectors. Each detector element consists of a
tive design is the rotating envelope tube (Schardt radiation-sensitive solid-state material (such as
et al. 2004). The anode plate constitutes an outer cadmium tungstate, gadolinium-oxide, or gado-
wall of the rotating tube housing; it is therefore linium oxi-sulfide) with suitable dopings, which
in direct contact with the cooling oil and can be converts the absorbed X-rays into visible light.
efficiently cooled via thermal conduction The light is then detected by a Si photodiode
(Fig. 3). This way, a very high heat dissipation attached to the backside of the scintillator. The
rate and fast anode cooling is achieved, enabling resulting electrical current is amplified and con-
high power scans in rapid succession. Due to the verted into a digital signal. Key requirements
central rotating cathode, permanent electromag- for a suitable detector material are good detec-
netic deflection of the electron beam is needed tion efficiency, i.e., high atomic number, and
to position and shape the focal spot on the very short afterglow time to enable the high
anode. The electromagnetic deflection is also gantry rotation speeds that are essential for the
used for the double z-sampling technology of imaging of moving organs and for fast volume
high-end scanners of one vendor (Flohr et al. coverage.
2004, 2005). CT detectors must provide different slice
Different clinical applications require differ- widths to adjust the scan speed, through-plane
ent X-ray spectra and hence different kV settings resolution, and image noise for each application.
for optimum image quality and/or best possible With a single-slice CT detector, different colli-
signal-to-noise ratio at lowest radiation dose. mated slice widths are adjusted by pre-patient
Contrast-enhanced CT scans using iodinated collimation of the X-ray beam.1 For a 2-slice CT
contrast agent, in particular CT angiographic detector, different slice widths can be obtained by
examinations, benefit from low kV settings. pre-patient collimation if the detector is sepa-
Because of the increased iodine contrast at lower rated midway along the z-width of the X-ray
kV, the contrast-to-noise ratio (CNR) in contrast-­ beam.
enhanced images increases at low kV if the radia- To simultaneously acquire more than 2 slices
tion dose is kept constant (McCollough et al. at different slice widths, detectors with a larger
2009). Vice versa, lower radiation dose is suffi- number of detector rows than finally read-out
cient at low kV to maintain a desired CNR slices have to be used. The required total beam
(Schaller et al. 2001a; Wintersperger et al. 2005; width in the z-direction is adjusted by pre-
McCollough et al. 2009). Because of limitations patient collimation, and the signals of every two
of the X-ray tube current low kV protocols have (or more) detectors along the z-axis are elec-
so far been limited to small patients and children. tronically combined to thicker slices. The detec-
Recent progress in X-ray tube design has led to tor of a 16-slice CT (Siemens SOMATOM
the introduction of X-ray tubes capable of pro- Emotion 16) as an example consists of 16 cen-
viding high power reserves at 70, 80 and tral rows, each with 0.6 mm collimated slice
90 kV. They have the potential to enable contrast-­ width at isocenter, and 4 outer rows on either
enhanced low kV scans in adult and in obese side, each with 1.2 mm collimated slice width—
patients without compromising CNR. In coro- in total, 24 rows with a z-width of 19.2 mm at
nary CTA, as an example, a radiation dose reduc- isocenter (Fig. 4). In a 16 × 0.6 mm acquisition
tion by 49–68% even in obese patients has been
reported when using 70 and 80 kV protocols 1
Note that the slice width is always measured at the iso-
(Meinel et al. 2014). center of the CT system.
Multi-slice CT: Current Technology and Future Developments 9

Fig. 4 Principle of MDCT slice collimation. Example of combination of the signals of every 2 central rows—16
a 16-slice detector, which consists of 24 detector rows and collimated 1.2 mm slices (bottom)
provides either 16 collimated 0.6 mm slices (top) or—by

mode the X-ray beam width is adjusted such detector, in total 16 collimated 1.2 mm slices
that only the central 16 rows are irradiated (Fig 5, bottom). The 16-slice detectors of other
which are read-out individually (Fig. 4, top). To vendors are similarly designed, providing, e.g.,
obtain 16 collimated 1.2 mm slices, the pre- 16 collimated 0.625 mm slices or 16 collimated
patient collimator is opened. The full z-width of 1.25 mm slices.
the detector is irradiated, and the signals of MDCT detectors with 64 detector rows pro-
every 2 central rows are electronically com- vide 64 collimated 0.5, 0.6 mm, or 0.625 mm
bined. This results in 8 central 1.2 mm slices slices, depending on the manufacturer. They
plus 4 outer 1.2 mm slices on either side of the allow acquisition of thicker slices by electronic
10 S. Ulzheimer et al.

90° 95°

Detector A

Detector A
26 cm

33 cm
Detector B Detector B

Fig. 5 Dual source CT scanner with two independent angle of 90°. One detector (a) covers the entire SFOV
measurement systems. A scanner of this type provides with a diameter of 50 cm, while the other detector (b) is
temporal resolution equivalent to a quarter of the gantry restricted to a smaller, central field of view. Second gen-
rotation time, independent of the patient’s heart rate. First eration (right). To enlarge the SFOV of detector B, the
generation (center). The measurement systems are at an system angle was increased to 95°

combination of every two detector rows. This 0.6 mm slices per rotation. The shortest gantry
results in 32 collimated 1.0, 1.2 mm, or 1.25 mm rotation time is 0.33 s. Meanwhile, newer gen-
slices. One CT system has a detector with 128 erations of DSCT systems have been intro-
collimated 0.625 mm slices (z-width 80 mm at duced, which are equipped with 128- and
isocenter). Meanwhile, CT systems with 80 or 196-slice detectors, respectively, and provide
160 collimated 0.5 mm slices have been intro- gantry rotation times down to 0.25 s. To enlarge
duced. Third-generation DSCT systems provide the SFOV of detector B, the system angle was
96 collimated 0.6 mm slices with both detectors increased to 95°.
(z-width 57.6 mm). The widest commercially One key benefit of DSCT is improved tem-
available CT detectors cover 16 cm at isocenter, poral resolution for the examination of moving
either by acquiring 320 collimated 0.5 mm slices organs, such as the heart, the lung, and the tho-
or 256 collimated 0.625 mm slices, depending on racic vessels. The shortest data interval needed
the vendor. for image reconstruction at the isocenter is half
a rotation of scan data—a so-called half scan
sinogram. In a DSCT scanner, the halfscan sin-
2.4 Dual Source CT ogram can be split up into two quarter scan
sinograms which are simultaneously acquired
A dual source CT (DSCT) is a CT system with by the two measurement systems in the same
two X-ray tubes and two detectors, see Fig. 5. relative phase of the patient’s cardiac cycle and
Both measurement systems operate simultane- at the same anatomical level due to the 90°
ously and acquire CT scan data at the same ana- angle between both detectors. With this
tomical level of the patient (same z-position). approach, constant temporal resolution equiva-
In 2005, the first DSCT system was commer- lent to a quarter of the gantry rotation time trot/4
cially introduced (Flohr et al. 2006). The two is achieved in a centered region of the SFOV—
acquisition systems are mounted onto the rotat- 83 ms for the first-generation DSCT, 75 ms for
ing gantry with an angular offset of 90°. One the second-generation DSCT, and 66 ms for
detector (A) covers the entire SFOV with a the third-generation DSCT, independent of the
diameter of 50 cm, while the other detector (B) patient’s heart rate. DSCT is sufficiently accu-
is restricted to a smaller, central field of view rate to diagnose coronary artery disease in
because of space limitations on the gantry. patients with high and even irregular heart
Figure 5 illustrates the principle. Using the rates, and in difficult-to-image patients (e.g.,
z-­flying focal spot technique (Flohr et al. 2004, Sun et al. 2011; Lee et al. 2012; Paul et al.
2005), each detector acquires 64 overlapping 2013; Westwood et al. 2013). The good tempo-
Multi-slice CT: Current Technology and Future Developments 11

ral resolution is also beneficial to reduce

motion artifacts in cardiothoracic studies (e.g.,
Hutt et al. 2016).
In addition to improving temporal resolu-
tion, the dual source principle can be exploited
favorably in other clinical situations. If both
X-ray tubes are simultaneously operated at the
same tube potential (kV), up to 240 kW peak
power is available with the third-generation
DSCT. These power reserves are not only ben-
eficial for the examination of morbidly obese
patients, whose number is dramatically grow-
ing in western societies, but also to maintain
adequate X-ray photon flux when high volume
coverage speed is needed. Additionally, both
X-ray tubes can be operated at different kV and
mA settings, enabling the acquisition of dual
energy data. While dual energy CT was already
evaluated 30 years ago (Kalender et al. 1986;
Vetter et al. 1986), technical limitations of the
CT scanners at those times prevented the devel-
opment of routine clinical applications. On the
DSCT system dual energy data can be acquired
nearly simultaneously with sub-second gantry
rotation times and fast volume coverage. The
use of dual energy CT data can in principle add
functional information to the morphological
information based on X-ray attenuation coeffi-
cients that is usually obtained in a CT exami-
nation. Meanwhile, a variety of different
applications of dual energy CT scans have been Fig. 6 Case study illustrating the clinical performance of
evaluated, with some of them on their way to third-generation DSCT. ECG-triggered high-pitch spiral
clinical routine (see the reviews in Lu et al. scan of the aorta and the iliac arteries in a patient with
aortic dissection. Scan parameters: 0.25 s rotation time,
2012; Remy-Jardin et al. 2014; Marin et al.
pitch 3.2, scan speed 738 mm/s, 90 kV, DLP = 177 mGy cm.
2014; Agrawal et al. 2014; Patino et al. 2016). Total scan time 0.8 s. Note the clear visualization of the
By simultaneously operating both measure- right coronary artery (arrow). Courtesy of Klinikum
ment systems in a spiral scan mode, DSCT sys- Großhadern, Munich, Germany
tems provide very high scan speeds up to
737 mm/s, see the clinical example in Fig. 6. 3 Measurement Techniques
As a drawback of DSCT systems, the
SFOV of the second detector cross-scattered The two basic modes of MDCT data acquisition
radiation, i.e., scattered radiation originating are sequential (axial) and spiral (helical) scanning.
from tube A and detected by detector B (at an
angle of 90°) and vice versa, has to be care-
fully corrected for to avoid distortions of CT 3.1  DCT Sequential (Axial)
M
numbers by cupping or streaking artifacts. Scanning
This can be done either by measurement of
cross-scattered radiation or by model-based Using sequential (axial) scanning, the scan vol-
approaches (Petersilka et al. 2010). ume is covered by consecutive axial scans in a