3D Ultrasound in Prenatal Diagnosis: A Practical Approach 9783110497359, 9783110496512

Citation preview

Rabih Chaoui, Kai-Sven Heling 3D Ultrasound in Prenatal Diagnosis

Rabih Chaoui, Kai-Sven Heling

3D Ultrasound in Prenatal Diagnosis A Practical Approach

Prof. Dr. med. Rabih Chaoui PD Dr. med. Kai-Sven Heling Prenatal Diagnosis Clinic Friedrichstraße 147 10117 Berlin, Germany

This is a translation of the original book in German: 3D-Sonografie in der pränatalen Diagnostik Translated by Rabih Chaoui.

ISBN: 978-3-11-049651-2 e-ISBN (PDF): 978-3-11-049735-9 e-ISBN (EPUB): 978-3-11-049400-6 Library of Congress Cataloging-in-Publication Data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.dnb.de. The publisher, together with the authors and editors, has taken great pains to ensure that all information presented in this work (programs, applications, amounts, dosages, etc.) reflects the standard of knowledge at the time of publication. Despite careful manuscript preparation and proof correction, errors can nevertheless occur. Authors, editors and publisher disclaim all responsibility and for any errors or omissions or liability for the results obtained from use of the information, or parts thereof, contained in this work. The citation of registered names, trade names, trademarks, etc. in this work does not imply, even in the absence of a specific statement, that such names are exempt from laws and regulations protecting trademarks etc. and therefore free for general use.

© 2016 Walter de Gruyter GmbH, Berlin/Boston. Typesetting: LVD GmbH, Berlin Printing and Binding: Hubert & Co. GmbH & Co. KG, Göttingen Cover image: © Rabih Chaoui, Kai-Sven Heling ♾ Printed on acid-free paper Printed in Germany www.degruyter.com

For Kathleen, Amin and Ella Chaoui. For Rajae, Anais, Reem and Anna Heling.

Preface The first three-dimensional (3D) ultrasound demonstration of a fetal face was performed in 1989, a momentous event that was considered as the birth of 3D ultrasound. More than 7 years later, the first major scientific event on 3D ultrasound occurred in 1997 when Professor Merz organized the first world congress on this topic. The introduction of rapid computer processing around the year 2000 enabled the widespread use of 3D ultrasound equipment. Indeed, more than half of the obstetrical clinics and offices are currently using ultrasound equipment with 3D capabilities. Despite this rapid expansion in 3D ultrasound equipment and a large body of scientific literature on the use of 3D ultrasound in obstetrical imaging, very few textbooks are available on this topic. This book is intended to fill this void, to serve as a guide on this subject with a primary focus on the technical aspects of the 3D technology. We have dedicated a significant part of our daily clinical practice on 3D ultrasound over the past decade and have organized and participated in numerous educational activities on 3D ultrasound. We have also been very active in research and discovery on this topic and have contributed significantly to the current state-of-the art of 3D ultrasound. This book is the culmination of our life work on this topic. A successful 3D ultrasound examination has two important parts: the acquisition of the 3D volume and the post-processing manipulation of the volume data set. In this book the acquisition and manipulation of 3D volumes is explained in a step-by-step practical approach. The book is divided into three main sections: the first section provides details on how to acquire the optimum volume, the second section describes various volume rendering modes and the third section explains the organ-specific application of 3D techniques. With more than 500 figures, the book provides an exemplary approach to 3D ultrasound in prenatal diagnosis. We owe several people a debt of gratitude for their significant contribution to our 3D ultrasound journey. First and foremost, Dr. Bernard Benoit, a giant in the field of ultrasound imaging, who has been and continues to be a great source of inspiration to us. Many of the 3D ultrasound tools could not have been developed without his tremendous technical and artistic experience. We also would like to thank the engineering and management teams at Kretztechnik (General Electric-Healthcare) in Zipf, Austria for their close cooperation, and their tireless support over the years. We thank our patients who contributed to all the images in this book and who continues to motivate us to push the limit of this technology forward. This book will not have been realized without the professional team at DeGruyter publishing, especially Mrs. Simone Witzel, Dr. Bettina Noto and Mrs. Anne Hirschelmann for their committed and unwavering support in this effort. We have delayed production of this book on several occasions. The time has now come to present you this book on the most recent available 3D ultrasound in obstetrics. Berlin, December 2015 R. Chaoui, K. S. Heling

VIII 

 Preface

Technical ultrasound words All 3D examinations and experience in this book is based on Voluson ultrasound equipment produced by the company General Electric, GE Healthcare. The images in this book were generated with Voluson e8 and e10 machines and most tools presented in this book as VCI®, TUI®, Magicut®, Glass-body mode®, HD-Live®, Sono-AVC®, VOCAL® and others are protected names. To facilitate reading we decided to omit the ® sign in all the book. Some abbreviations are listet below: Abbreviations 3D 4D GA HD Sono-AVC® TUI® VCI® VOCAL®

Three-dimensional Ultrasound Four Dimensional Ultrasound Gestational Age High-Definition Sono Automatic Volume Calculation Tomographic Ultrasound Imaging Volume Contrast Imaging Virtual Organ Computer-aided AnaLysis (VOCAL)

Contents Preface 

 VII

Part I: Basics of 3D Ultrasound 1 1.1 1.2 1.3 1.4

 3 Basics of 3D and 4D Volume Acquisition   3 Introduction   3 Preparing the volume acquisition      Types of volume acquisition 10  13 Conclusions 

2 2.1 2.2 2.3 2.4 2.5 2.6

Orientation and Navigation within a Volume   15 Introduction  Storing and exporting volume data sets  Orientation in the three orthogonal planes  Navigation within the orthogonal planes   23 Artifacts in the multiplanar mode      Conclusions 25

 15  15  16  17

Part II: Methods of 3D Rendering

3.6 3.7 3.8 3.9

 29 3D Rendering of a Volume      Introduction 29  29 The render box and the orientation within a 3D volume   30 Artifacts in 3D rendering   34 Different rendering modes and the mixing of modes  Special effects in 3D: dynamic depth 3D rendering  39 and light source   41 Threshold, transparency, brightness and color scales      Magicut, the electronic scalpel 43  46 Multiple light sources and „HD-live studio“   48 Conclusions 

4 4.1 4.2 4.3 4.4 4.5 4.6

 49 Volume Contrast Imaging (VCI)      Introduction 49  49 Principle of VCI   53 Static VCI   56 4D with VCI-Omniview      4D with VCI-A 58  61 Conclusions 

3 3.1 3.2 3.3 3.4 3.5

X 

5 5.1 5.2 5.3 5.4 5.5 5.6

 Contents

Multiplanar Display I – Orthogonal Mode and Omniview Planes   62 Principle  Multiplanar reconstruction and different ways  62 of displaying cross-sectional images     63 Practical approach in orthogonal mode Practical approach in getting an „anyplane“ using  64 Omniview tool   67 Typical applications of Omniview planes      Conclusions 74

6 6.1 6.2 6.3 6.4

 75 Multiplanar Display II: Tomographic Mode      Principle 75  75 Practical approach   81 Typical applications in tomographic mode   88 Conclusions 

7 7.1 7.2 7.3 7.4

Surface Mode Rendering and HD-Live   93 Principle   93 Practical approach 

8 8.1 8.2 8.3 8.4

 106 Maximum Mode Rendering      Principle 106  107 Practical approach  Typical applications of maximum mode   116 Conclusions 

9 9.1 9.2 9.3 9.4

 117 The Minimum Mode   117 Principle   117 Practical approach  Typical applications of minimum mode   124 Conclusions 

10 10.1 10.2 10.3 10.4

 125 The Inversion Mode      Introduction 125  125 Practical approach  Typical applications of inversion mode   132 Conclusions 

11 11.1

The Silhouette Tool   133 Principle 

 93

Typical applications of surface mode   105 Conclusions 

 133

 98

 112

 119

 127

 62

Contents 

11.2 11.3 11.4

 133 Practical application  Typical applications of silhouette tool   142 Conclusions 

 137

12 12.1 12.2 12.3 12.4 12.5 12.6

 143 The Glass-Body Mode and HD-Live Flow   143 Principle   144 Practical approach   148 Glass-body mode with HD-live flow function     148 Typical applications in the glass-body mode  153 HD-live flow using the color silhouette tool   155 Conclusions 

13 13.1 12.2 13.3 13.4

 156 The B-Flow Mode   156 Principle   158 Practical approach 

14 14.1 14.2 14.3 14.4

Biplane Display using the Electronic Matrix Transducer   162 Principle   162 Practical approach   164 Typical applications of biplane mode   177 Conclusions 

15 15.1 15.2 15.3 15.4

 178 Calculation of 3D Volumes   178 Principle   178 Practical approach 

Typical applications of the B-flow mode   161 Conclusions 

 158

Clinical application of volume calculation   184 Conclusions 

 162

 184

Part III: Clinical Applications of Prenatal Diagnosis 16 16.1 16.2 16.3 16.4 16.5 16.6 16.7

 187 3D Fetal Neurosonography      Introduction 187 Fetal neurosonography with 3D ultrasound  3D visualization of specific brain structures 

 187  192

 196 Reconstruction of fetal brain structures in 3D rendering      The intracranial vascular system in color Doppler 196  200 Fetal neurosonography before 14 weeks of gestation   205 Conclusions 

 XI

XII 

 Contents

17 17.1 17.2 17.3 17.4 17.5

 206 3D of the Fetal Skeleton  Limitations in the assessment of the fetal skeleton  206 using 2D ultrasound   206 3D of fetal spine and ribs      3D of the fetal limbs 213  216 3D of the facial and cranial bones   218 Conclusions 

18 18.1 18.2 18.3 18.4 18.5 18.6

 219 3D of the Fetal Face  The sonographic examination of the face in 2D and 3D ultrasound   220 The face in multiplanar display   223 The normal face in 3D/4D surface mode      The abnormal face in 3D/4D 229  234 The facial bones in 3D/4D   235 Conclusions 

19 19.1 19.2 19.3 19.4

3D Intrathoracic and Intraabdominal Organs   236 Introduction   236 Intrathoracic organs     242 Intraabdominal organs     Conclusions 254

20 20.1 20.2 20.3 20.4 20.5

 255 STIC and 3D/4D Fetal Echocardiography  The sonographic assessment of the heart in two-dimensional  255 ultrasound   255 Acquiring cardiac volumes  Fetal echocardiography in 3D/4D multiplanar reconstruction   258 Fetal heart in 3D/4D volume rendering      Conclusions 268

21 21.1 21.2 21.3 21.4

 269 3D in Early Pregnancy      Background 269  269 3D volume rendering in early gestation   278 Multiplanar display in early gestation   282 Conclusions 

Further literature references and sources  Index 

 287

 283

 219

 236

 257

Part I: Basics of 3D Ultrasound

1 Basics of 3D and 4D Volume Acquisition 1.1 Introduction The current technology of 3D ultrasound is based on advanced mechanical or electronic transducers with the ability to acquire a volume or sequence of volumes. The pictorial information acquired in a 3D volume can then be displayed on the screen in different ways: either as one or multiple multiplanar 2D images (See Chapters 4, 5 and 6) or as a spatial volume, which projects external or internal anatomic features of a volume (see Chapter 3). It is generally acknowledged that acquisition, display, and manipulation of 3D volumes are techniques that entail a steep learning curve. The “quality of a volume” to provide valuable information or a perfect 3D picture depends not only on the skill of the examiner in the post-processing manipulation, but also on the adjustment of the 2D image prior to volume acquisition. In this chapter, we discuss some aspects of image optimization as well as some basics of volume acquisition.

1.2 Preparing the volume acquisition Five important steps should be considered during the preparation of a 3D volume acquisition. These steps are: 1. Optimization of the 2D image before volume acquisition 2. Choice of the best reference or starting plane with anticipation of the result expected 3. The box of acquisition or volume box 4. Acquisition angle 5. Volume quality and resolution

1.2.1 Optimization of the 2D image before volume acquisition Before a 3D-, 4D- or STIC-volume is acquired, the 2D image optimization is mandatory for obtaining optimal results. The term “reference plane” or “acquisition plane” is then used to designate the starting 2D plane for a 3D acquisition. A 3D volume is a collection of adjacent 2D images and the resolution in the complete volume improves with the resolution of each single plane. Apart from the choice of line density and image frame rate, optimization of the image also includes correct positioning of the “region of interest” in the volume box and optimal positioning of the focus zone. In this case, the choice of both the angle size of the box and the angle depth (acquisition angle) are important. If the volume is acquired with color Doppler, the examiner should also consider optimizing color resolution, color persistence and frame rate. Figures 1.1 to 1.3 are examples of optimization of images prior to volume acquisition.

4 

 1 Basics of 3D and 4D Volume Acquisition

Fig. 1.1: Left: During the preparation of a 3D acquisition of a head and brain, part of the head will be absent in this case. In the figure to the right, the image is more centered and thus optimal for a 3D acquisition.

Fig. 1.2: Left. The image is not optimized and appears too “bright” with low contrast for a 3D acquisition in surface mode. Right: After image optimization, amniotic fluid appears black and transparent with the surface contours well defined.

1.2.2 Choice of the best starting plane at volume acquisition In 3D ultrasound, the best image quality within a volume is found in the reference plane and in the planes parallel to the reference plane, whereas the reconstructed orthogonal or planes oblique to the reference plane have a reduced image quality. Even if it is not always the case, it is still better if the operator knows what the volume is going to be used for before the acquisition.

1.2 Preparing the volume acquisition 

 5

Fig. 1.3: Left: The preset of this image is not well adjusted neither for a static nor for a STIC acquisition. The color Doppler is too large and the velocity scale is too low. The right figure presents an optimized image prior to acquisition.

1.2.3 The acquisition box or volume box The volume acquisition box or volume box determines two parameters of a 3D volume in the 2D image, namely the height and width, corresponding to the X- and Y-axis respectively (Fig. 1.5). It is recommended that the operator adjusts the box size to include all anatomic components of a target volume. During an acquisition in 4D, the box borders are selected close to the anatomic region of interest, and can be corrected directly during the 4D display, however, for a static 3D recording, we recommend choosing a large box in order to avoid excluding some structures adjacent to the anatomic regions from the volume.

Fig. 1.4: The volume box has three dimensions. Image height and width are selected in 2D mode while the depth of the volume box is selected by choosing the acquisition angle f.ex. 50°, 70° etc. (compare with next figure).

6 

 1 Basics of 3D and 4D Volume Acquisition

Fig. 1.5: A volume box size consists of height, width and depth. The acquisition angle is the total volume angle, which is during the acquisition half the angle in front and half the angle behind the reference plane. The reference plane is the image the examiner sees on the screen while activating a 3D acquisition.

1.2.4 Acquisition angle The angle of acquisition refers to the depth of a volume, corresponding to the Z-axis, and is the sweep angle of the elements within the probe during acquisition (Fig. 1.5). The angle of acquisition is adjusted by the operator with the choice of the size of the volume box prior to the 3D volume acquisition. There is no gold standard for the best angle of acquisition but the choice depends mainly on the anatomy of the target organ and the type of acquisition. The angle of acquisition is then the total angle of the volume, but during the acquisition half of the angle is present behind and the other half in front of the reference or acquisition plane (Fig. 1.5). Depending on the organ examined, the size and shape of the box will differ. Figures 1.6 and 1.7 present different types of acquisition boxes. For instance, the volume box of a fetal spine sets the box wide but the acquisition angle narrow (Fig. 1.6), while for the heart, width and depth are almost equal (Fig. 1.7).

1.2.5 Acquisition quality The 3D volume quality is selected by the user with the choice of the volume acquisition duration. The examiner should keep in mind that, within a volume box with the same

1.2 Preparing the volume acquisition 

 7

Fig. 1.6: The shape of a volume box is generally defined by the organ examined. In the case of spine and ribs the box is large with a rather narrow depth. The size of the volume box is displayed on the screen with B for width 66° and V for Volume depth 40°.

Fig. 1.7: The choice of the volume box shape is often determined by the examined region: Left up: typical volume box of a spine in a longitudinal view. Right up is an example of a box for a fetal face in 3D. The box in the left bottom image has a narrow volume depth in a STIC acquisition and the large box in the right bottom image is for the acquisition of a large body part as the head, or abdomen and thorax or a complete fetus in early gestation.

8 

 1 Basics of 3D and 4D Volume Acquisition

Fig. 1.8: With the same volume angle the choice of “high quality” acquisition (as maximum, high2, high1) leads to the acquisition of many images, with the result of a high resolution image, while the acquisition of few images leads to a low or middle quality volume.

Fig. 1.9: 3D volume of a fetus in low quality (upper images) and maximum quality (lower images) acquisition with corresponding different resolutions in the acquired images.

1.2 Preparing the volume acquisition 

 9

acquisition angle, a slower sweep will allow the acquisition of more images and provides better resolution, while a quick sweep acquires fewer images, resulting in an image of reduced resolution (Fig. 1.8). Moreover, if in a volume box more images are available for 3D calculation, this results in better quality images of the reconstructed B and C planes in the multiplanar display. In Figs. 1.9 and 1.10, compare the top and bottom images. It should also be kept in mind that the best quality is not always achieved by choosing the maximum resolution, and the examiner should find his or her own preferred presets. Figure 1.11 reveals a fetus after a 3D acquisition in low resolution (left), in middle resolution (middle) and in highest resolution (right). In our opinion, the middle image has the best quality with a smooth face, while the right image has many details close to artifacts. In 3D static and 4D acquisition, the quality of acquisition is referred to as low, medium, high and maximum, whereas in STIC acquisition, the quality of acquisition is reflected in the duration of acquisition: 7.5, 10, 12.5, or 15 seconds. Figure 1.12 reveals the same fetus after a 3D static (left image) and after a 4D acquisition (right image), both with high resolution.

Fig. 1.10: STIC volumes with an acquisition of short duration and corresponding low quality (upper images) and acquisition with a long duration with a better resolution (lower images). The resolution in both cases is different due to the different acquisition time.

10 

 1 Basics of 3D and 4D Volume Acquisition

Fig. 1.11: Fetal face after a 3D static acquisition in different qualities in “low”, “mid2” and “maximum”. The middle image appears to be the best and demonstrates that the best result is not always achieved by choosing the highest resolution.

Fig. 1.12: Acquisition of a fetal face with static 3D (left) and in 4D (right). Details recognition and resolution are generally better with static 3D.

1.3 Types of volume acquisition Four types of acquisition of a volume are used, namely 1. Static 3D 2. Real-time 4D with a mechanical 3D transducer (4D) 3. Spatial and temporal image correlation (STIC) 4. Real-time 4D with an electronic matrix transducer (4D)

1.3 Types of volume acquisition 

 11

1.3.1 3D Static acquisition Principle: 3D Static refers to a single 3D volume acquired which contains an infinite number of adjacent 2D still ultrasound planes with no regard to temporal or spatial motion. Currently, this is the most common mode of volume acquisition in obstetrics and gynecology. Potential: This type of acquisition is easy to learn and can be rapidly performed, allowing the examiner to acquire several volumes and to store them for later evaluation. The static 3D acquisition is usually acquired with 2D preset but can also be combined with color Doppler, power Doppler or B flow for vascular evaluation of volume content. The post-acquisition rendering makes numerous displays possible, which are discussed in greater detail in the following chapters. Limitations: The main limitation of static 3D acquisition is its inability to assess events related to movement, especially at the level of the heart in grayscale and in combination with color or power Doppler. Valvular movements, myocardial contractility and flow events cannot be reliably assessed with static 3D. Another limitation is the common occurrence of movement artifacts when movements occur during the acquisition, as observed on fetal face, limbs, spine or others (see chapter 2,3).

1.3.2 Real-time 4D with a mechanical 3D transducer (4D) Principle: Most 4D volume acquisitions are achieved today using a mechanical transducer with an integrated rotation motor. The principle is similar to the static 3D technology with the difference that the motor rotates continuously, thereby acquiring a series of volumes to be displayed almost as one movement. The combination of a series of 3D volumes within a time interval is then called 4D. Different terminologies are used to describe this method, including real-time 3D, real-time 4D or 4D; in this book we will only use the term 4D. Potential: The major advantage of 4D acquisition is its ability to display instantaneously on the ultrasound screen real-time 4D volumes as they are acquired. This is impressive especially when visualizing the face, hands and feet of a moving fetus. Opening eyes, yawning or other movements make the fetus feel much more real and human to the parents. This technique is ideal for beginners and many examiners since the 3D image appears directly on the screen and can be adapted accordingly. Limitations: The main limitation in this type of acquisition is the challenge of finding a balance between a good quality 4D image on one hand and the speed of the rotation within the motor on the other, to allow an almost live impression. On a routine good

12 

 1 Basics of 3D and 4D Volume Acquisition

resolution image of a fetal face in live 4D, 4 images per second are displayed; this is nowhere near the 15 images per second or more needed to produce a “live” impression of images. Therefore the image flow often appears not smooth, unless the fetal movements are slow. Slow movements of fetal arms, legs, or facial grimace, yawning or eye opening can be followed well using this technique.

1.3.3 Spatio-Temporal Image Correlation (STIC) acquisition Principle: STIC acquisition is similar to a slow 3D acquisition of a duration between 7.5 and 15 seconds, and is used mainly for the acquisition of images of a beating heart or vessels with pulsation. The software makes calculation of heart rate based on the tissue excursion concurrent with cardiac motion possible. The acquired volume is processed internally, where the systolic peaks are used to calculate the fetal heart rate and the volume images are then rearranged according to their temporal events within the heart cycle, thus creating a cine-like loop of a single cardiac cycle. Potential: The advantages of STIC volume acquisition include the ability to assess myocardial wall motion and valve excursion. The 4D information is available within seconds from the volume acquisition. Once the reference plane is optimized, the STIC acquisition can be easily achieved. STIC acquisition can be obtained from 2D grayscale imaging combined with other imaging modalities, such as color, power, or high-definition Doppler and B-flow. If the 2D and color Doppler scanning conditions are good, the STIC can then be used for offline reconstruction of planes and off-line assessment. This potential of a virtual examination of the heart is one of the big potentials of this technique. Its clinical use is discussed in Chapter 15. Limitations: Disadvantages of STIC acquisition include a fairly delayed acquisition time, which can be hampered by fetal movements or maternal breathing movements, thus introducing artifact into the volume. Another limitation is the fact that a single heart cycle is displayed as a cine loop, which makes this technique inaccurate in the assessment of arrhythmias, in particular ectopic beats.

1.3.4 Real-time 4D with an electronic matrix transducer (4D) Principle: The usual type of 3D mechanical transducer consists of one row of crystals used to generate the 2D image and a mechanical motor that sweeps the ultrasound beam so as to generate multiple 2D planes which are then stacked together to produce the 3D volume. Recent electronic matrix transducers are designed with a rectangular area of crystals (around 8000) arranged in rows. For a 2D examination few rows are activated while for a real-time 4D examination almost all can be activated, if needed. The sweep in 3D or 4D is achieved electronically by activating the neighboring crystal rows which provides images two to four times faster than with mechanical 3D transducers.

1.3 Types of volume acquisition 

 13

Potential: In addition to the advantage of a 4D in almost real-time as explained above, the advantage of the matrix probe is the speed of image acquisition, which is ideal for following fetal movements and in fetal cardiology. It can also be expected that the forthcoming years will see further improvements with the advent of faster processors. Limitations: The main limitations today are the challenges associated with cramming such major technology into a small transducer, in particular trying to reduce its weight and the heat it produces. Another limitation is the speed of calculation of information before displaying the images in a real time flow.

1.4 Conclusions These days, the acquisition of 3D volumes can be achieved either with a mechanical 3D transducer or, more recently, with an electronic matrix transducer (Fig. 1.13). Before commencing the acquisition, the examiner should decide on the target and how it should be displayed with 3D. After 2D image optimization, a selection box is chosen and height, width and depth of the volume box adapted according to the requirements. The volume box is centered, the volume quality is selected and the acquisition type started as 3D static, STIC or 4D (Fig. 1.13). The volume data can be displayed on the screen either as planes (known as multiplanar reconstruction or display) or as a three-dimensional volume image in one of the numerous volume rendering modes (Fig. 1.14). The next chapters will discuss the different ways of displaying and manipulating 3D volume data in depth.

Mechanical transducer

Electronic matrix transducer

Biplane

Static 3D volume

STIC volume

4D volumes

Fig. 1.13: Scheme presenting the possibilities of volumes acquisition. Using either a mechanical or an electronic transducer the acquisition can be performed with the choice of static 3D, STIC volume or 4D volumes. Moreover, the electronic matrix transducer enables the acquisition of biplane-images, which, however, cannot be further manipulated, as is the case in 3D volume data sets.

14 

 1 Basics of 3D and 4D Volume Acquisition

Static 3D volume

STIC

4D volumes

Multiplanar reconstruction

Volume rendering

Sectional planes: – Single plane – Orthogonal mode – Tomography mode – Omniview planes

Volume images: – Surface mode – Transparency mode – Volume calculation

Fig. 1.14: Overview showing the different possibilities of volume rendering and display on the screen after the acquisition of a volume either as static 3D, as STIC volume or as 4D volumes. A volume data set can be displayed either as “planes” what is called multiplanar reconstruction or as a spatial volume called “volume rendering”. The different rendering modes listed are discussed in the next chapters.

2 Orientation and Navigation within a Volume 2.1 Introduction In the previous chapter we discussed how a 3D volume acquisition is prepared and a 3D volume data set acquired. In this chapter we explain how to display the result of volume acquisition on the screen and how 2D and 3D images are extracted after volume manipulation. Many examiners store the volume during the examination, to edit it at the end of the examination or later. In order to achieve an ideal image from a volume the examiner must know how to use the different tools in the 3D software and how to navigate through the volume. In other words, 3D volume manipulation represents a pure application of digital software that must be comprehensively learned. Such expertise can be gained only with a lot of hands-on use of this software, in combination with reading monographs and attending special courses in 3D ultrasound. The aim of this chapter is to provide the user with some helpful tips and hidden features that will help to gain an orientation in the volume and achieve a good image. The 3D rendering of a volume with a spatial 3D result is discussed in the next chapter.

2.2 Storing and exporting volume data sets Occasionally the acquired volume is directly manipulated by the sonographer during the examination. This carries a risk of losing the volume if an incorrect button is pressed. For this reason, we recommend directly storing a good volume on the ultrasound machine hard drive prior to volume manipulation. When the data set is being saved, care should be taken to choose the correct file format. This is generally achieved by adjusting the configuration of the “storing buttons” when the ultrasound equipment is delivered to an ultrasound center. A volume can be (wrongly) saved as an image (Bitmap, TIFF, JPEG) or correctly as a volume (3D) dataset. Acquired STIC or 4D volumes should be saved as a “volume cineloop” and not as 3D. Saving a volume in the wrong format makes subsequent manipulation impossible. In order to determine whether a volume or image is saved accurately on the machine, it is best to acquire different images and volumes and to open and manipulate them at the end of the examination. A simple trick is to check the size of the selected image: a figure has around 1MB, while volumes are more than 5MB. STIC and 4D Volumes have an additional time-line symbol that serve to illustrate a series of volumes. When working with a volume, the function “export” makes exporting as an image (for example, the figures in this book) possible, for example as video clips (e.g., for use with patients or in scientific talks) or as a digital data set. In order to export one volume or a collection of volumes from one patient on an external drive, it is recom-

16 

 2 Orientation and Navigation within a Volume

mended that the data be exported as “uncompressed volume data” and in format “.4dv”. Saving in this format makes it easier for the data to be selected for reimporting into an ultrasound system of the same series or to be used on a remote computer with the PC-Software 4D-view®.

2.3 Orientation in the three orthogonal planes After a volume acquisition, in most cases the 3D-display on the screen is presented in a multiplanar mode, mostly in the three orthogonal planes (Fig. 2.1). These planes are labeled A, B and C, respectively. Plane A is shown in the upper left of the image and refers to the reference plane during volume acquisition (see Chapter 1). Planes B and C are digitally reconstructed planes orthogonal to plane A. Plane B is the 90° rotation, and C corresponds to the horizontal plane. The acquisition angle corresponds to the

Fig. 2.1: In the orthogonal mode the volume data set is displayed as three planes perpendicular to each other. In the upper left the reference plane A is displayed, in the upper right the 90° vertical rotation plane and in the bottom left the 90° horizontal rotation plane C. In the B-plane the acquisition angle is recognized and when present movements artifacts can be identified in this plane (see Figs. 2.13, 2.14).

2.4 Navigation within the orthogonal planes 

 17

aperture of plane B, while the width of the box is recognized in image A. The values are displayed on the side of the image. The image in plane A is usually of best quality, because it was directly visualized, while the images in planes B and C are of lower resolution since they were calculated from the digital information. The display of a 3D volume dataset can, however, be saved differently by the user such that a 3D rendering or a tomographic image or others appears directly on the screen after volume acquisition.

2.4 Navigation within the orthogonal planes Navigation within a volume enables the generation of new planes and can thus simulate an ultrasound examination (Figs. 2.2–2.6). The planes seen on the screen are interrelated and any change within one plane affects the others. As starting plane the so-called active plane is selected and can be recognized by the calipers on the border of the image (Fig. 2.2 upper right). When the navigation is performed in the active plane, the images change in the two other orthogonal planes. The examiner can switch to another plane to continue navigation, which then becomes the active plane. In general, navigation within a 3D volume can be achieved in three ways:

Fig. 2.2: This and next images illustrate how the intersection dot can be used for the navigation within the volume. This dot always points to the same position displayed in the planes A, B and C. In A it is displayed in yellow, in B in orange and in C in cyan. In this example all three planes intersect in the liver, where the dot is seen. In the plane B the stomach is recognized. The examiner now moves the point in plane B (arrow) placing it on the stomach and the images in plane A and B yield the result as shown in Fig. 2.3.

18 

 2 Orientation and Navigation within a Volume

Fig. 2.3: (see previous Fig. 2.2). After the intersection dot has been moved in plane B to be placed in the stomach the images in A and C have changed to display new images, where the stomach is seen as well. The dot always is pointed at the same place in all three planes. Now the examiner wants to visualize the descending aorta. In plane B the point is moved to the descending aorta (arrow) and two new planes A and C emerge.

Fig. 2.4: (see also Figs. 2.2 and 2.3). In this figure, the intersection point now lies in the plane B in the aorta, which is also seen in planes A and C. Using this approach, the examiner can continue to navigate within the volume. Often, the resulting image should be adjusted by slightly rotating the volume, which is shown in the next figures.

2.4 Navigation within the orthogonal planes 

 19

Fig. 2.5: The 3D volume of a fetal face in orthogonal display mode. In the plane A, one has the impression to recognize a good profile, but planes B and C demonstrate that the plane is oblique. To adjust the volume, the intersection dot is moved in plane B to be placed on the nose (1, short arrow) and the image rotated around this point (2, curved arrow), resulting in both eyes being positioned horizontally (see result in Fig. 2.6), and in plane A the profile is now exactly seen in the midline. This step of manipulation is called rotation.

1. 2. 3.

By moving the intersection dot in one plane (called navigation), By rotating the axes (called rotation), or By scrolling through the volume and getting parallel images (called translation). Planes are called A, B or C (Fig. 2.1), while axes are labeled X, Y and Z and displayed in different colors (Figs. 2.7, 2.8).

Navigation with the intersection dot: In the orthogonal mode display, the three planes A,B and C are perpendicular to one other and the intersection of all three planes is the intersection dot (Fig. 2.3). This dot can be actively clicked by the examiner and moved from its position, which results in a change in the two other planes (Figs. 2.3, 2.4). Since the dot always indicates all three planes to the same structure, it can be placed and changed in any plane depending on the region of interest. Such navigation can always be achieved in any of the A, B or C planes. Figures 2.2 to 2.4 illustrate a step-by-step navigation using the intersection dot. Rotation: Selecting one of the X-, Y- or Z-axes makes rotation of the image along this axis possible (Figs. 2.5, 2.6). The axes can be rotated either by using one of the three knobs on the machine or by selecting one of the lines. Instead of trying to work out

20 

 2 Orientation and Navigation within a Volume

Fig. 2.6: The 3D volume in Fig. 2.5 was adjusted so that both eyes in plane B are positioned horizontally. In next step, plane C was adjusted to align the face axis and get in plane A the profile exactly in the midline.

Fig. 2.7: This image is part of a volume in orthogonal display mode and this plane A illustrates the three axes X, Y and Z resp. as horizontal line, vertical line and as a dot. In Fig. 2.8 these lines were drawn for a better understanding of the rotation steps.

2.4 Navigation within the orthogonal planes 

 21

Fig. 2.8.: Orthogonal display mode with the lines X, Y and Z, which were drawn for a better understanding. The arrows show the rotation directions, which result when the buttons X , Y or Z are rotated in both directions.

Fig. 2.9: In a volume, here displayed as tomography mode, the examiner can also scroll through different parallel planes. The resulting images are then shown, image-by-image. Scrolling can be used in any plane in the volume to display parallel planes starting from the plane of interest.

22 

 2 Orientation and Navigation within a Volume

Fig. 2.10: When scrolling through a volume, the images displayed are parallel planes to the starting plane and scrolling corresponds to translation movements, which is a sliding along a horizontal axis. In addition to navigating with the dot and rotating along axes, translation is the third way of navigating within a volume.

Fig. 2.11: During navigation through a volume with axis rotation and translation the orientation was lost as shown in this case. The use of the INIT-button makes it possible to return to the initial image of the volume at the stage of volume acquisition, as shown in Fig. 2.12.

2.5 Artifacts in the multiplanar mode 

 23

Fig. 2.12: The figure presents the image of Fig. 2.11 after activating the INIT-Button. The lateral view of the face can now be viewed as the original image.

which knob leads to which rotation, most beginners will use trial and error, turning one knob and seeing what happens on the screen. Translation: After selecting an active plane on the screen, the activation of the knob “translation” will lead to scrolling through parallel planes to the active plane (Figs. 2.9, 2.10). This scrolling resembles a sliding movement with the transducer during a live examination. “INIT”, the initial position and starting point: Occasionally, after turning different knobs and moving the intersection dot, the examiner may lose the orientation (Fig. 2.11). The easiest way to recover is to press the button “INIT” (for initial position), which will then return the volume display to its initial position (Figs. 2.11, 2.12) when it was acquired and stored.

2.5 Artifacts in the multiplanar mode Artifacts occur more commonly in 3D than in 2D sonography. They occur during the 3D volume acquisition and are either due to maternal movements such as breathing, laughing, etc., or more commonly due to fetal movements. Artifacts arising during

24 

 2 Orientation and Navigation within a Volume

Fig. 2.13: Images in plane A is directly recorded during volume acquisition while images in planes B or C are digitally reconstructed images from adjacent images of plane A and can thus reflect movement artifacts. Artifacts during volume acquisition are therefore best recognized in planes B and C.

Fig. 2.14: Artifacts due to fetal movements during acquisition are rarely seen in plane A, but almost always in planes B and C (see explanation in Fig. 2.13)

2.5 Artifacts in the multiplanar mode 

 25

volume acquisition are best recognized in the B-Plane in the orthogonal mode (Figs. 2.13, 2.14). While significant movements are easily recognized, small movement artifacts lead to only slight distortion of the image, which may escape detection. Decent artifacts during volume acquisition of regions like brain, heart, abdominal organs or skeleton remain often hidden. The examiner should therefore always bear in mind that a 3D examination is a reconstructed examination of acquired planes, which can become important when measurements are performed. In the next chapter we will discuss the impact of artifacts upon 3D rendering of a volume data set.

2.6 Conclusions The post processing working on a volume is a prerequisite for understanding 3D volume ultrasound. The two important steps are the orientation and the navigation within a volume. The best orientation is achieved in the three orthogonal planes, called A-, B- or C-planes, where the intersection dot is directed to the same point in the three planes. The navigation within a volume is best achieved in multiplanar than in volume display. The intersection dot can be used to navigate within the single planes, while the volume axes X, Y and Z are used to rotate the images of the volume and the planes, as in tomography, in order to scroll (translation) from image to image. These basic steps allow generating planes out of the volume, even if these are not ideally visualized during live scan and opens thus a new field in imaging. The navigation additionally makes it possible to simulate an examination out of volumes.

Part II: Methods of 3D Rendering

3 3D Rendering of a Volume 3.1 Introduction For many users, the spatial reconstruction of a volume with the display of a 3D image on the screen, especially an image of the face, has been synonymous with 3D ultrasound. In design and 3D terminology this spatial reconstruction is usually called “rendering”. The 3D rendering of an ultrasound volume data set is performed according to some principles and standards that will be explained in this chapter. Understanding some basics of rendering and manipulation can be very helpful in the achievement of good quality images in the different rendering modes. These modes are described separately in chapters 7–13.

3.2 The render box and the orientation within a 3D volume In the multiplanar mode, 3D volume rendering can be selected by activating the “Rendering” button. A rectangle will then appear in the 3 planes (A, B and C) and an additional fourth 3D calculated image is displayed in the right lower corner (Fig. 3.1). This volume-rendering box, hereinafter referred to as the “render box” in this book, can be modified in its height, width and depth. The render box allows the user to select the information to be included in the 3D calculation (see Figs. 3.2 to 3.6). The result can be recognized immediately in the 3D rendered image. All sides of the box are white with the exception of one side that is displayed in green in two planes (Figs. 3.2 to 3.6). This is the “projection line” or “green line” (similar to a camera) from which the 3D image perspective is seen. To facilitate orientation, the box has two orientation points, a rectangle and a rhombus, that are also displayed in the 3D box (Fig. 3.6). With more experience, orientation in the 3D image becomes easier and the green box with the marks can be removed from the 3D image (Figs. 3.3–3.5). The perspective from which the image is seen in 3D can also be modified (Figs. 3.3–3.5). In order to visualize the face, the line is often placed directly in the amniotic fluid in front of the face (Fig. 3.2). Figures 3.3–3.5 illustrate examples of how changing the line of projection influences the result. Under certain anatomic conditions (e.g., imaging of the heart), it may be necessary to change the line into a curve (Fig. 3.4). This can be achieved by modifying the position of a point to obtain a curved line. Once placed within the volume at its final position including the required information, the render box can be “fixed” for further manipulation. Using this selection, the orientation lines disappear (Fig. 3.7). In other words, from the entire volume acquired, only the information placed within the render box is then available for further 3D volume manipulation; the adjacent information is no longer displayed in the 3D image. Following this step, the electronic Magicut scalpel can be used to

30 

 3 3D Rendering of a Volume

Fig. 3.1: By activating the button “Rendering”, the examiner can switch from the orthogonal display mode to the volume-rendering mode. In planes A, B and C a render box appears and in the lower right panel the calculated 3D image is displayed. Size of the box can be changed by changing the position of one of the six borders of the box, defining thus the ultrasound information to be displayed in the volume (see next figure). The “green” projection line shows the perspective of view into the volume.

remove parts of the image, the image can be rotated or the information in the box displayed in different modes. These actions are known as “manipulation of the volume”.

3.3 Artifacts in 3D rendering Artifacts in 3D are often the result of fetal movements during volume acquisition and rarely due to maternal movements. These artifacts can easily be identified during 3D rendering directly on the displayed image (Fig. 3.8). While large movements cause obvious artifacts that make the image of no value for further interpretation, some minor fetal motions lead to slight image distortions that may escape detection. Small

3.3 Artifacts in 3D rendering 

 31

Fig. 3.2: In the 3D rendering mode of a fetal face the “green” projection line is placed on the top in front of the face (arrows).

Fig. 3.3: In this example the volume was rotated and a vertical line (arrows) was selected, to visualize the face. We do not recommend such an approach, since the orientation in planes A, B and C gets easily lost. Generally it is better to keep the position of the planes only slightly unchanged as shown in the previous figure.

32 

 3 3D Rendering of a Volume

Fig. 3.4: In this STIC volume the projection line (arrows) is placed inside the thorax directly within the heart just under the aortic root (plane B). This enables the demonstration of the four-chamberview in surface mode (also refer to Chapter 20).

Fig. 3.5: Upper panel: The projection line is placed in the amniotic fluid in front of the face (arrows). In the lower images, the projection line is placed behind the face and the so-called reverse-face view is displayed.

3.3 Artifacts in 3D rendering 

 33

Fig. 3.6: The 3D image (lower right panel) only displays the information included in the render box. Here the upper part of the head is out of the box and therefore not seen in 3D. For a better orientation in the render box two marks are displayed on the box and in the corresponding images, namely a square and a rhombus.

Fig. 3.7: In this case, the render box was “fixed” or “frozen”, which means that the 3D image information can be rotated, magnified and manipulated without a change in the information included in it. The green box can still be seen in the 3D figure but with increasing experience, the box can be removed from the image, as is seen in most figures in this book.

34 

 3 3D Rendering of a Volume

Fig. 3.8: During acquisition, these fetuses moved and the 3D images show the corresponding movement artifacts. In the lower right image, the artifact generated a typical Pinocchio nose in this fetus.

artifacts on the face are often recognized immediately, while in other regions, small artifacts may escape detection. In 4D the examiner switches immediately to the image without artifacts, while in 3D the user has to repeat the volume acquisition. Figure 3.8 presents some 3D motion artifacts.

3.4 Different rendering modes and the mixing of modes The render box offers the possibility to display images from the acquired volume by using different modes. The rendered 3D image appears then as a 2D projection on the 2D monitor with the impression of a 3D effect (like all 3D images in this book). The render box often includes information from different fetal structures, which have different ultrasound properties: fluid is anechoic, bony structures hyperechoic and tissue hypoechoic. When the render box and the projection line have been selected, the ultrasound system assesses all signals in the depth of the box seen from the projection line and the selected mode displays the required information. Generally there are two algorithms for 3D rendering with different types of visualization: either surface rendering or transparent rendering.

3.4 Different rendering modes and the mixing of modes 

 35

3.4.1 Surface modes rendering In surface mode rendering (Figs. 3.9 top, 3.10) the ultrasound signals that are analyzed are mainly those directly behind the projection line. In general, the projection line is placed in the amniotic fluid in such a way that the fetal skin becomes visible. In Chapter 7, the different applications of the surface mode are discussed. Different display algorithms are provided in surface mode rendering and are discussed in this section. Their selection depends on the object to be visualized and also for “aesthetic” aspects. The following calculations and display modes are available: Surface smooth, surface texture: In these modes, only the surface next to the projection line is displayed (Fig. 3.9 top, 3.10). In surface texture, the exact grayscale information present in the images is displayed and for surface smooth, the grayscale information is slightly blurred with a filter and displayed smoothly.

Fig. 3.9: Once a 3D rendering image is displayed on the screen the examiner can choose different modes of rendering. Here we see images for the same fetus, presented in surface smooth, surface texture, maximum and light mode.

36 

 3 3D Rendering of a Volume

Light mode: Dark and light are displayed predominantly here so that structures near the projection line are displayed as light and those deeper are displayed as dark (Fig. 3.9, bottom right). The light mode is almost never used, only occasionally with inversion mode. Gradient light mode: In this mode the surface is displayed as if illuminated by a light source with a depth-effect (Fig. 3.10, top left). Structures, which are perpendicular to insonation, are shown brighter than the other insonated regions. With gradient light, the best results are achieved when there is adequate fluid around the structure. HD-live mode: High-definition (HD)-live mode was introduced a few years ago to improve the surface image and deliver a realistic skin-like image (Fig. 3.10d, 3.11). A new transparency function was recently added to HD-live mode that highlights the contours and is called “silhouette” function. Silhouette enables gradual transparency display within the whole volume. Chapter 11 discusses the use of the silhouette function.

Fig. 3.10: Generally speaking, a mixture of two display modes is applied in a 3D image. The figure shows a fetal face in “gradient light” (a) and “surface texture” (b) and a better result in c) in a mixture of 70/30 %. The figure in d) is a result of a combination of High-definition (HD-) live surface and smooth of 50/50. There is no perfect combination, since each user has his personal preference.

3.4 Different rendering modes and the mixing of modes 

 37

Fig. 3.11: These figures show the step-by-step manipulation of a fetal face volume using HD-Live. The panel at left reveals that surface mode rendering with gradient light has been selected after 3D acquisition. The middle figure presents the result after the switch to HD-live mode with 50/50 ratio of “texture” and “smooth”. The final image (right) is the result after increasing HD-live to 100 %, increasing shadowing and transparency and changing the position of light source.

3.4.2 Transparency mode rendering While surface mode displays only the first layer, in the transparency mode different details can be highlighted within the render box. Depending on the object of interest, all signals included in the render box are analyzed and demonstrated accordingly. Maximum mode is a transparency mode in which all hyperechoic information in the render box is calculated and projected (Fig. 3.12, top left) (also refer to Chapter 8). This render mode is used to visualize bones and is ideal in the examination of the fetal skeletal system (see Chapter 17). Minimum mode is a transparency mode in which all anechoic information in the entire volume is calculated and projected (Fig. 3.12, top left) (also refer to Chapter 9). This approach is ideal to visualize fluid-filled organs as well as the heart and large vessels. Inversion mode inverts (as the name implies) the echogenicity of volume components and is thus the inversion of the information displayed with the minimum mode. Signals from neighboring structures are suppressed (Fig. 3.12, bottom left) (see Chapter 10).

38 

 3 3D Rendering of a Volume

Fig. 3.12: Demonstration of different organs and regions by using different transparent modes as maximum mode, minimum mode, inversion mode and X-Ray contrast mode. Please refer to the text and corresponding chapters regarding different modes.

X-Ray mode is a transparent contrast mode used for the visualization of hypoechoic tissue and is calculated as a mixture of minimum and maximum transparency modes. The ideal regions of interest for the use of this mode are the lungs, abdominal organs, brain (Fig. 3.12), and other regions. The X-Ray mode is most often combined with a thin slice such as that used in volume contrast imaging (VCI) (see Chapter 4). Silhouette mode is as previously mentioned used to visualize the contours of internal structures. This mode is combined with HD-live mode and gradual transparency can be selected (see Chapter 11). With increasing experience the examiner comes to realize that all of these modes provide the best results when used in combination. A button can be used to adjust the distribution between the two modes used. For the face of the fetus, for example, a 70 % gradient light and 30 % surface texture can be selected. Minimum and X-Ray modes is another good combination. When using HD-live, the image becomes smoother when HD-live smooth is increased (Fig. 3.11).

3.5 Special effects in 3D: dynamic depth 3D rendering and light source 

 39

In the different chapters, other combinations are discussed, such as color Doppler with the glass-body mode, B-Flow with static 3D and STIC or the new HD-live with the silhouette tool.

3.5 Special effects in 3D: dynamic depth 3D rendering and light source The 3D visualization on the screen, whether on the ultrasound system or on the computer, is in the end a projection of a 3D image onto a 2D surface and does not need (as used in consumer electronics today) stereoscopic glasses. For this reason, in recent years additional image enhancements have been introduced for 3D imaging that aim to highlight the spatial impression. Two functions in particular are of importance: 3D dynamic depth rendering: This software displays structures that are deep in the volume visualized with colors blue, gray or black, and with the color switch between sepia and blue a depth rendering can be appreciated. Often this is amniotic fluid that appears nicely blue. The level of depth effect can, however, be adjusted. These colors can then be shaded based on the depth of the regions examined: nearby areas are shown lighter and deeper areas darker. Figure 3.13 presents an example without (a) and with depth rendering in gray (b) and blue (c). In early pregnancy, the entire fetus with the amniotic cavity can be easily visualized and highlighted very well with this deep rendering (Fig. 3.14).

Fig. 3.13: The effect of depth can be improved by using the tool “dynamic depth rendering“, which adds a color blue or black to the structures that are deep in the volume, making the amniotic fluid blue in this case. The image to the left is the raw image and the images in the middle and to the right and are the result after adding black and blue respectively. The level of color can be adjusted according to the depth information in the image (see Fig. 3.14).

40 

 3 3D Rendering of a Volume

Fig. 3.14: The 3D effect by coloring the surrounding liquid (see previous figure), can be ideally applied to early gestation where a fetus or an embryo are well surrounded by amniotic fluid.

Fig. 3.15: With the recent software the 3D effect can be improved by using a light source. Similar to a torch the lighting can be placed in different positions with a shadow recognized behind the structures. Ideally for a fetal face we prefer to place the light source in the upper part of the image (see also Fig. 3.16).

3.6 Threshold, transparency, brightness and color scales 

 41

Fig. 3.16: The light source can be used for special effects in early gestation. It can be positioned in the upper part of the image, from the side or even as shown in the lower right panel from behind. The light source is seen on the screen to the lower right of the image.

Light source function: A few years ago, a new option was added to the modes already discussed that enables the illumination of the 3D image with a light source. The 3D image usually appears as if light is projected directly from the front onto the image. The new software allows the user to move a light source around a sphere so as to illuminate the image from different perspectives, even from behind (Fig. 3.15, 3.16). This effect is particularly impressive when used with HD-live with its skin-like tone (see Figs. 3.11, 3.15) and this lighting effect provides good results, particularly in early gestation (Fig. 3.16). The new multiple light source function is discussed at the end of the chapter (see Fig. 3.8).

3.6 Threshold, transparency, brightness and color scales The quality of a rendered 3D image depends mainly on the 2D image prior to volume acquisition, as explained in Chapter 1. During 3D volume manipulation some tools can be applied to improve the quality of the 3D image. Threshold: The function “Threshold” or “Gray Threshold” defines the level of gray scale used in the display of the 3D image calculation (Fig. 3.17). This knob can be used mainly to eliminate weak artifacts and speckles to highlight structures with true

42 

 3 3D Rendering of a Volume

Fig. 3.17: Increasing the level of “threshold” and its impact on the 3D image. The magenta color shadow appears only as long as the knob is used. There is no „ideal“ threshold level but the level is selected according to the result seen on the screen.

Fig. 3.18: In 3D rendering modes different colors can be chosen from gray to different sepia colors as well as the new skin-like HD-live. Most patients associate however the sepia color with the 3D color, and is still the most popular 3D color.

3.7 Magicut, the electronic scalpel 

 43

signals. A very low threshold (50) can be applied to highlight bones in maximum mode or other structures in the inversion mode. Sometimes, the umbilical cord can be faded out with an increase in threshold. Transparency and Gain: The level of transparency can be increased and the image appears transparent in its depth. More gray scale information can also be obtained by increasing the gain, but this results in more artifacts and less detail. Brightness and contrast: This can be subsequently modified only to a modest degree in most 3D systems and are used to enhance the image. Color tints: Different color tints can be selected to color the 3D image, such as the classical sepia, but also gray, blue, ice or different skin tones. This coloring is often used to increase the 3D effect (Fig. 3.18). Most users have only a small number of colors that they use regularly.

3.7 Magicut, the electronic scalpel It is rare that the user manages to acquire a very good 3D image in a single attempt without the need for further corrections. In most cases of static 3D volumes, the image

Fig. 3.19: The electronic scalpel is also known by the name “Magicut”. After a volume data set is frozen, the volume can be rotated in all directions and undesired information can be removed. On the left this structure (placenta or uterine wall) is obstructing the face and has to be removed (arrows). After a vertical rotation (right) the interfering information can be clearly identified and removed with Magicut (see next figure).

44 

 3 3D Rendering of a Volume

Fig. 3.20: Left: With Magicut the structure in front of the face is removed after rotating the volume. The face then appears clearer on the right side, but there still are structures besides and behind the head, which can be removed.

Fig. 3.21: Left: Adjacent to the head the disturbing structures (see Fig. 3.20) can also be removed with Magicut. The image on the right is already very good but can still be improved as shown in next figure.

is improved after some retouching and the use of some of the manipulation tools described above. This is often needed to better visualize some regions, or simply for aesthetic reasons. The electronic scalpel, also called Magicut, can be used after the image is fixed. Different tools can be used here such as the deletion of the surrounding structures. By rotating the volume, the structure to be deleted can be shown floating freely and can simply be deleted without affecting the surrounding structures. Figures 3.19 to 3.22 provide an example of the use of the Magicut tool to obtain the optimal image. One special function of Magicut is the depth or selective deletion, which allows the user to selectively delete slice-by-slice a specific area without deleting the structures behind it. Of special interest is the use of Magicut in 3D volumes acquired with color Doppler and displayed in glass-body mode. It is possible in such cases to sepa-

3.7 Magicut, the electronic scalpel 

 45

Fig. 3.22: The figure on the left is rotated and visualized from the top resulting in the figure in the middle. The aim is now to remove the information in front and behind the face. The result is then seen in the right panel with a face appearing like an artistic “bas-relief”. Figure 3.18 was manipulated in a similar manner.

Fig. 3.23: Instead of using manually Magicut to remove structures in front of the face, the recent software also makes it possible to automatically detect this information and to remove it. This feature called “Sono-Render-Live” adjusts as demonstrated in the lower panel, the green line in a curved shape (arrows) to fit to the region of interest. The sensitivity of this tool can also be adjusted.

46 

 3 3D Rendering of a Volume

rately erase either the structures on the grayscale image or those of color Doppler or both. Please refer to Chapter 12 for more details. A newly introduced function called “Sono-Render Live” (Fig. 3.23) makes automatic modification of the shape of the green line during volume rendering possible. Instead of the complicated deletion of some structures with Magicut, the software identifies, as illustrated in Fig. 3.23, the free fluid between the face and the anterior wall or placenta, and places the projection line (even curved) in this area so that the face appears instantly. This tool is mainly important during a live 4D examination in which the use of Magicut would be too time-consuming.

3.8 Multiple light sources and “HD-live studio” The introduction of a new light source a few years ago (see Fig. 3.5) provided the new possibility to improve the 3D effect in many rendering images, especially in combination with HD-live. In the most recent software release, an amelioration of this artistic approach to 3D is facilitated by the possibility of using up to three light sources at the same time, as in photography studios, and is therefore called “HD-live studio” (Figs. 3.24–3.27). The examiner needs to have some understanding in using these

(a)

(b)

(c)

(d)

Fig. 3.24: The fetus at 12 weeks is displayed in 3D HD-live smooth with one light source. The examples in (b), (c) and (d) illustrate the same volume but applying the new HD-live studio with three light sources and special light effects. The circles indicate the light sources used.

3.8 Multiple light sources and “HD-live studio” 

 47

Fig. 3.25: HD-live displayed with multiple light sources applied on an embryo at 8 weeks (left) and the same fetus two weeks later at 10 weeks (right) of gestation.

Fig. 3.26: HD-live display with multiple light sources applied to a fetus at 11 weeks (left) and the same fetus after removing the different neighboring structures with Magicut.

Fig. 3.27: The fetal face in 3D can be displayed very softly and artistically with the use of multiple light spots as revealed in these examples.

48 

 3 3D Rendering of a Volume

sources, as the position of each light source, its distance to the object and its type, can be changed separately. Preliminary examples of the use of these techniques are illustrated in Figs. 3.24 to 3.27.

3.9 Conclusions The 3D rendering of a volume is far more complex than navigation in the different planes and requires intensive familiarization with the 3D software and its different manipulation tools. The use of the render box, the green line and the orientation are the basics to be learned before further steps of volume manipulation are applied. The ultrasound information included in the render box can be displayed in 3D, either in a surface or in a transparent mode display (Fig. 3.28). The Magicut tool is used to clean the image and highlight the structures of interest, while the light source can be used to increase the spatial impression. The different render modes and other tools are discussed in the next chapters of this section.

Volume rendering

Surface modes

Transparency modes

Volume calculation

– – – – – –

– – – – – –

– VOCAL – Sono AVC

Surface smooth Surface texture Gradient light Light HD-live surface HD-live smooth

Maximum mode Minimum mode Inversion mode X-ray mode HD-live silhouette Glass body mode

Fig. 3.28: Overview of the different volume rendering modes either in surface mode or in the different transparency modes with the different displays as illustrated in Figs. 3.9–3.12.

4 Volume Contrast Imaging (VCI) 4.1 Introduction During certain 4D or static 3D examinations, it may be of great additional value to obtain a thin 3D slice of the studied view instead of a single plane. The advantage of this approach is an increase in the resolution and contrast and a decrease in the artifacts. This is the principle of Volume Contrast Imaging (VCI). VCI can be applied in real-time mode (as in 4D) used as VCI-A and VCI-C or VCI-Omniview, by getting a slice of the A-plane or the C-Plane or an Omniview plane respectively. In recent software the term VCI-C was replaced by the term VCI-Omniview. In static 3D, the static VCI is applied.

4.2 Principle of VCI A single reconstructed 2D image out of a 3D volume includes true information as well as artifacts called “noise” or “speckles”. With the activation of the VCI tool and the choice of a thin slice, the artifacts are reduced and the quality of the image resolution and the contrast are increased (Fig. 4.1 right). The principle is simple and is illustrated in Figs. 4.2. and 4.3. In Fig. 4.2, high amplitude peaks represent the true ultrasound information, while low amplitude peaks represent speckles and artifacts. Comparing two successive planes of an image, true information is found in the same spots with the same intensity on the images, while artifacts differ in intensity and position. Superimposing successive images, the required information from anatomic structures is enhanced whereas randomly gener-

Fig. 4.1: The left image of the embryo was reconstructed from a volume and reveals a low resolution with speckle. In the right image VCI was activated with the result of less artifacts and an increased resolution.

50 

 4 Volume Contrast Imaging (VCI)

Strong signals from anatomical structures

Volume Contrast Imaging VCI

Weak signals from artifacts (speckle, noise)

Fig. 4.2: Principles of Volume Contrast Imaging (VCI). The figure in VCI is reconstructed from several adjacent images (here two are shown). Signals from true tissue information are high and present at the same place in adjacent images, while signals from noise and speckles are weak and present at different places. The sum of two adjacent images (VCI) increases the intensity from true signals and information from noise and speckle is too low and almost eliminated.

Fig. 4.3: Series of schemes illustrating the VCI effect. In this fetal face from a 13 weeks fetus the true information present in adjacent images is from nasal bone, maxilla, chin, and brain tissue. However, artifacts are found in the different parts of the images, as illustrated with stars and circles. Adding three images leads to an increase in the true information and an almost disappearance of the artifacts around (compare with clinical example in the following figures).

4.2 Principle of VCI 

 51

Fig. 4.4: Two images from a static 3D volume of a brain in tomography mode left panel as native image. The right panel shows the image after activating VCI (here 1 mm) (arrow), which appears clearer and with a better contrast.

Fig. 4.5: Omniview display of the midsagittal view of an embryo. In the upper panel, native reconstruction of the plane of interest has a reduced resolution (upper right image), while in the lower panel VCI tool was activated and provides better contrast and resolution of the reconstructed image (lower right image).

52 

 4 Volume Contrast Imaging (VCI)

Fig. 4.6: 3D volume with a coronal visualization of lung, heart, diaphragm and liver. The upper images present a native image with tomography mode. The lower images show the result after activating VCI and increasing contrast and details recognition.

ated noise and speckles in different slices are reduced or sometimes eliminated (Figs. 4.2 and 4.3). In Fig. 4.3 the principle is illustrated with a schematic diagram of a face where the final image from a VCI slice of the face shows more resolution and contrast than each of the single successive images. An example is provided in Fig. 4.4. In this tomography mode two planes of the intracranial structures are visualized. The images on the left are the original volume images whereas the images on the right are the images after activating the VCI with increasing the contrast. In this case the X-Ray contrast mode was activated. Another example in early gestation can be observed in Fig. 4.5 and that of lung and liver imaging is demonstrated in Fig. 4.6.

4.3 Static VCI 

 53

4.3 Static VCI VCI can be applied to any multiplanar, tomographic or selected plane display (as in Omniview) to enhance the image quality and contrast (see Chapters 2, 5 and 6 for examples). The image appears as a plane but in reality it is a thin slice. The slice thickness can be selected from 1 to 20 mm depending on the information to be displayed. The rendering mode of the slice can be selected as in regular 3D rendering, as surface, maximum, minimum or X-Ray modes. X-Ray mode: This mode is ideally used for enhancement of tissue information and is used in the imaging of brain, lungs, kidneys, nuchal translucency and others. In most cases a thin slice 1–5 mm is selected (see examples in Figs. 4.4–4.8). Maximum mode: is ideally used to demonstrate spine, extremities, long bones or skull bones (Fig. 4.9). A good slice is selected between 5–20 mm thickness. Figure 4.10 shows an intrauterine device with pregnancy demonstrated with VCI with maximum mode. Minimum mode is good for use in anechoic structures and can be used in combination with X-Ray mode. Inversion mode until recently was not available with VCI but is now available with the electronic probe in 4D (Fig. 4.11)

Fig. 4.7: Left: fetus with a thickened nuchal translucency (arrow) as seen by a transabdominal examination. The same case is shown on the right following a transvaginal examination with a 3D volume acquisition and reconstruction of the sagittal view. Using VCI increases the image quality. The severity of nuchal thickening (arrows) can be better appreciated and a precise measurement can be performed.

54 

 4 Volume Contrast Imaging (VCI)

Fig. 4.8: In this fetus lying in vertex presentation, the corpus callosum cannot be visualized. After an axial 3D volume acquisition of the head and placing the reconstructed plane along the falx cerebri and cavum septi pellucidi (CSP) in the left image, the corpus callosum (CC) can be reconstructed (right). The image quality is improved by adding VCI with 2 mm thickness.

Fig. 4.9: Lateral acquisition of a 3D static volume of the fetal head with a VCI of 20 mm and maximum mode display revealing the skull bones with the corresponding sutures.

4.3 Static VCI 

 55

Fig. 4.10: Pregnancy with an intrauterine device (IUD). On the left, the IUD can be seen horizontally and its shape cannot be assessed in 2D ultrasound. On the right, the reconstruction of the IUD with 3D volume acquisition and the use of Omniview and VCI illustrate the shape of the IUD in a projected mode.

Fig. 4.11: Acquisition at the level of the abdomen demonstrating the kidneys with the tomography mode. The combination of a VCI-slice of 2 mm with minimum mode display highlights the hypoechoic renal pelvis and reveals the presence of a mild pyelectasis.

56 

 4 Volume Contrast Imaging (VCI)

Fig. 4.12: In this case, Omniview with a VCI slice of 18 mm and surface mode was used. For surface mode display often it is better to use conventional 3D or 4D acquisition modes instead of Omniview with VCI.

Fig. 4.13: A demonstration of the hard palate with a curved Omniview line and VCI. The display was selected as a mixture of maximum and surface modes.

Surface mode is rarely used, since a thin slice is rarely needed to demonstrate a surface. Instead a standard 3D or 4D is usually more useful since the 3D effect of surface mode is enhanced when the volume is larger. Occasionally, surface mode is combined with X-Ray and maximum modes (Fig. 4.13).

4.4 4D with VCI-Omniview During a 4D examination, the examiner can also directly draw a straight or a curved line along the region of interest to obtain a corresponding section or view. The result

4.4 4D with VCI-Omniview 

 57

Fig. 4.14: 4D with Omniview and VCI with direct demonstration of the vermis (short arrow) and the corpus callosum (long arrow). The line is drawn during the 4D examination and the VCI slice (here 2 mm) is activated.

Fig. 4.15: Lateral view of the fetal skull in 4D. A curved Omniview line was drawn lateral to the skull and a slice of 12 mm thickness was selected. The maximum mode display then makes direct visualization of skull bones with the coronary suture possible.

58 

 4 Volume Contrast Imaging (VCI)

Fig. 4.16: Direct demonstration of spine and ribs with 4D ultrasound and curved line Omniview and VCI of 14 mm thickness using a mechanical transducer.

can be improved by combining with VCI with an appropriate thickness. The 3D result is directly displayed side-by-side of the 2D image. The authors have good experience with this technique and use it in screening examinations. In a fetus with cephalic presentation, the online reconstruction of the corpus callosum and vermis (Fig. 4.14) can be directly achieved by selecting a straight line with a thin layer of 1–3 mm and the X-Ray mode. Another possibility is the combination of VCI with maximum mode for the demonstration of skull bones with sutures (Fig. 4.15) or spine with ribs (Fig. 4.16). Figures 4.14 and 4.15 illustrate examples of the use of VCI with maximum mode.

4.5 4D with VCI-A VCI of the A plane is a technique of scanning with a slice instead of only with 2D plane. This technique can be utilized using a mechanical probe (Figs. 4.17, 4.18) but with low frame rate and poor resolution. Resolution was improved with the advent of the electronic matrix transducer (see Chapter 1), which enables the examiner to make a rapid image calculation (Figs. 4.19, 4.20). Slice thickness and rendering display can be adjusted as needed. VCI-A can be used to examine the fetal lung, heart, kidneys, face, brain and other organs. Figures 4.17–4.20 present images acquired with VCI-A. In our experience, combining this technique with X-Ray mode can be used to improve con-

4.5 4D with VCI-A 

 59

Fig. 4.17: Direct transabdominal visualization of the corpus callosum using VCI-A. The left image presents the direct insonation in 2D and the right image is the live 4D display using a 5 mm slice in VCI-A with increased contrast.

Fig. 4.18: Direct demonstration of a four-chamber plane in 2D (left) and after activation of VCI-A (right). The right image shows a better contrast when using an 8 mm VCI-A slice. This image was obtained with a mechanical 3D transducer and a low frame rate of 14 Hz. Better resolution can be achieved with an electronic transducer, as illustrated in next images.

60 

 4 Volume Contrast Imaging (VCI)

Fig. 4.19: Left: a fetus with occipital encephalocele in 2D with low contrast. The right image reveals an improved image after activating the VCI-A with a 5mm slice. This was obtained with an electronic transducer and an image frame rate of 23 Hz.

Fig. 4.20: Two cardiac anomalies demonstrated with VCI-A. The heart appears with a higher contrast when using a slice of 3 mm (left) and 2 mm (right). The left image shows a hypoplastic left heart syndrome with a small left ventricle (LV). The right image shows a normal looking four-chamber-view, but behind the heart, the descending aorta (Ao) can be seen and to its right the dilated azygos vein (AZ) in a fetus with an interruption of the inferior vena cava with azygos continuity. These images were acquired with an electronic transducer and image frame rate of 47 Hz and 35 Hz resp.; right ventricle (RV).

trast discrimination between adjacent regions as heart-thymus, cardiac myocardium-lumen, corpus callosum-cortex or kidneys-bowel. The skeletal system can be well highlighted during the live examination when combined with maximum mode. VCI-A used on a matrix probe can also be combined with inversion mode, and this is further discussed below in Chapter 10.

4.5 4D with VCI-A 

 61

4.6 Conclusions VCI is an interesting additional tool to use in 3D and 4D examinations, which makes rapid use of a 3D slice possible instead of going through the many steps of volume acquisition and rendering. Combined with Omniview, its potential use is increased, especially when curved lines are applied.

5 Multiplanar display I – Orthogonal Mode and Omniview Planes 5.1 Principle An ultrasound examination is still based on the demonstration of standard 2D cross-sectional images of the organ or region examined. Hence, most examiners attempt visualizing such “standard” planes during their examination and some examiners may still feel unfamiliar with the successive images displayed in the tomographic or orthogonal mode. The fetal profile, the four-chamber-view of the heart, the mid-sagittal view of the corpus callosum or the longitudinal view of the spine are all standard planes to be visualized during a routine examination. Some planes, however, cannot be, effortlessly, achieved during fetal ultrasound examinations. The aim of this chapter is to demonstrate methods to obtain those typical cross-sectional planes out of a volume data set and their clinical application. The potential use of such a feature is still not fully explored, but in the future it may constitute an integral part of the regular examination, at least once automation and image pattern recognition is widely applied in ultrasound imaging. Another main advantage of storing a volume data set is the ability to perform a virtual offline examination from the stored volumes. This feature, the so-called “virtual second opinion”, can be carried out remotely with benefits that have been proven in several single and multicenter studies

5.2 Multiplanar reconstruction and the different ways of displaying cross-sectional images The demonstration of single images from a digital volume data set can occur in different ways. In the field of imaging, the general term used for such technique is multiplanar reconstruction, “MPR”. In 3D ultrasound the nomenclature used differ slightly from one manufacturing company to another. In the system used by the authors the term “multiplanar” is often used as a synonym for “orthogonal mode”. In this book we will use “multiplanar reconstruction” or “multiplanar display” as a broader term and discuss the different modalities separately: Currently the following three modalities of multiplanar reconstruction in volume ultrasound are available: – Single or multiple images in multiplanar orthogonal mode – Single or multiple images in tomographic mode – Single image slices acquired by selective cutting within the volume using tools like “Omniview”. With the latter it is possible not only to cut within the volume using a straight line but also to adjust a curved line or to draw any multipoint line and get an “anyplane”.

5.3 Practical approach in orthogonal mode 

 63

In any of the modalities used in order to improve the quality of reconstructed images and reduce the speckles, the authors recommend applying the “Volume Contrast Imaging” (VCI) function as discussed in Chapter 4, or to use the 3D-SRI speckle reduction filter, where available.

5.3 Practical approach in orthogonal mode Before acquiring a volume the preset can be selected to display the result either in orthogonal or in tomographic mode. Once the images are displayed the examiner seeks first the most familiar image to start the manipulation on. In some situations it is helpful to scroll through the volume or to navigate in the different planes by using the intersection dot as explained in Chapter 2. Once the image, which is close to the ideal plane, is reached the examiner then uses rotation (spinning) in the different planes to align the structure of interest along one of the typical fetal axes (falx, spine, aorta, etc.), which would facilitate orientation. Figures 5.1–5.3 provide a step-by-step of how an ideal midsagittal view with the nuchal translucency and the nasal bone is generated out of a transvaginal volume of

Fig. 5.1: Step-by-step reconstruction of a plane out of a volume data set. During a transvaginal examination, it is often difficult to manipulate the transducer to obtain the ideal view of the structure of interest. In this case the examiner tried to obtain a view of the profile and decided to do it by reconstruction. The acquisition of the fetal face is performed from the side, as close as possible to the final plane of interest (see next images).

64 

 5 Multiplanar display I – Orthogonal Mode and Omniview Planes

Fig. 5.2: Taking the volume of Fig. 5.1 the VCI tool is activated to enhance contrast and the intersection dot is placed on a well identifiable structure as the falx cerebri (arrow). In the bottom the falx is oblique in plane C. Therefore, the lower image is turned until the falx aligns with the X-axis and the plane in B is aligned as well. The result is seen in Fig. 5.3.

the fetal face: Due to the limitations of transducer manipulation, the profile could not be initially visualized and a volume was thus acquired. After activating the VCI the falx cerebri was in a first step sought (B plane in Fig. 5.2) and was aligned along the Y-axis. In the C plane, the falx is still oblique and will be aligned along the X-axis in the next step (Fig. 5.3). In this plane, the profile is clearly visible and the visualization of the single plane (Fig. 5.4.) now makes the measurement of the nasal bone and nuchal translucency possible. Figures 5.5 and 5.6 reveal how the maxilla was visualized by manipulating a volume in a first and second trimester fetus.

5.4 Practical approach in obtaining an “anyplane” using Omniview tool A good alternative is using the new Omniview tool. After few adjustments of the image to identify parts of the structure of interest, the examiner can draw directly within the volume a straight or a curved line and get simultaneously the reconstructed image. Since the reconstructed “Omniview-image” appears simultaneously, an adjustment of

5.4 Practical approach in obtaining an “anyplane” using Omniview tool 

 65

Fig. 5.3: Continuing Fig. 5.2. Now the falx lies in the X-axis and the intersection dot is on the falx cerebri and in the plane A the profile is well recognized. Fig. 5.4 reveals the final result.

Fig. 5.4: Result of a fetal profile reconstructed from the volume data set of an oblique view of the fetal face (see Figs. 5.1–5.3). Now nasal bone and nuchal translucency are well seen and the nuchal translucency is measured.

66 

 5 Multiplanar display I – Orthogonal Mode and Omniview Planes

Fig. 5.5: Orientation in a transvaginally acquired 3D volume of a fetal face at 13 weeks’ to demonstrate the maxilla. The intersection dot lies on the maxilla and is observable in all planes. The contrast of the image was increased by choosing a VCI slice of 1 mm in combination with X-Ray contrast and maximum mode.

Fig. 5.6: Demonstration of the maxilla in second trimester in the orthogonal mode demonstrated without VCI. The intersection dot and the planes are adjusted in a way to observe a sagittal view of the maxilla in the plane A (upper left) with a perpendicular plane in the upper right panel. Plane C confirms that the volume has a good orientation.

5.5 Typical applications of Omniview planes 

 67

Fig. 5.7: Use of Omniview on a 3D volume of thorax and abdomen. The user can draw up to three lines and in this case two lines generate axial planes across the heart (upper right panel) and at the level of the urinary bladder (lower right panel). The third horizontal line is a frontal view of thorax, lungs, diaphragm, stomach and bladder (Lower left panel). While planes 1 and 2 resulted from straight lines, the line 3 was selected as curved line.

the placed line can be directly achieved. In the actual software up to three lines can be drawn at the same time and are recognizable by different colors (Figs. 5.7, 5.8). After a line is placed it can be moved in parallel or tilted. An Omniview line can be drawn as a straight, curved or a free drawn line (Fig. 5.7). The resulting image can be used either as projected line or, in a few cases, with a curved line, which can also be displayed as a stretched line. In order to improve image quality, it is recommended to reduce speckles by using either the 3D-SRI filter or to combine it with VCI-mode. Interestingly the use of Omniview-tool is not only limited to a static 3D volume but can be used in a 4D or a STIC volume as well.

5.5 Typical applications of Omniview planes Thorax and abdomen: Figure 5.7 reveals that Omniview can ideally be used in visualizing thoracic and abdominal organs, where typical cross-sectional planes are documented. Figure 5.8 demonstrates a simple way to highlight the kidneys in a volume. Fetal brain: Figures 5.9 to 5.11 provide examples of fetal neurosonography where Omniview allowed a rapid reconstruction of the corpus callosum, vermis and a coronal view of cavum septi pellucidi and other structures.

68 

 5 Multiplanar display I – Orthogonal Mode and Omniview Planes

Fig. 5.8: Use of Omniview in the visualization of kidneys. The 3D volume was acquired in a fetal dorsoanterior position and kidneys are found left and right to the spine. Two Omniview lines are drawn parasagittal (1, yellow line, 2, magenta line) and one frontal (3, cyan line) highlighting the kidneys from different perspectives.

Fig. 5.9: Omniview with VCI for the demonstration of the corpus callosum. The falx cerebri and cavum septi pellucidi are used as landmarks.

5.5 Typical applications of Omniview planes 

 69

Fig. 5.10: Omniview with VCI for the demonstration of vermis and brain stem.

Fig. 5.11: After a lateral static 3D acquisition of a fetal head, three Omniview lines are drawn to demonstrate the corpus callosum (CC) in a sagittal plane, a coronal plane for the cavum septi pellucidi (CSP) and another posterior coronal plane to visualize the cerebellum (CER).

70 

 5 Multiplanar display I – Orthogonal Mode and Omniview Planes

Fetal skeleton: The fetal spine and skull bones can also be demonstrated quite well by combining Omniview with VCI and maximum mode, as shown in Figs. 5.12 and 5.13. Depending on the examined organ and fetal position, a decision can be made as to whether a straight or curved line should be selected (Figs. 5.12, 5.13). The maxilla with the hard and soft palate can often be visualized with the orthogonal mode (Figs. 5.5, 5.6) but in some occasions it is more reliable to use Omniview for a targeted visualization either with a curved or a drawn line (Figs. 5.14, 5.15). Fetal heart: Omniview can be used on the fetal heart either with STIC in gray scale or in color Doppler. Standard views as the four-chamber-view and the three-vessel-trachea view can be well and rapidly demonstrated using this tool (Fig. 5.16). A direct view over the atrioventricular valves can demonstrate the en-face view and the valvular apparatus. Early pregnancy: Early scan performed before 14 weeks’ gestation has a limited possibility of transducer manipulation. In such situations Omniview helps in getting reconstructed planes of some typical regions of interest. Figure 5.17 provides an example of the intracranial translucency. Interesting but not yet of any clinical value is the free hand drawing of an Omniview line, as illustrated in the example of a stretched embryo in Fig. 5.18.

Fig. 5.12: On a static 3D volume a curved Omniview line with a VCI slice of 12 mm and maximum mode display demonstrates spine and ribs in this case.

5.5 Typical applications of Omniview planes 

 71

Fig. 5.13: Skull bones can be clearly observed and identified after a lateral acquisition of a fetal head and the use of Omniview, here as a curved line, with 19mm wide thickness and maximum mode.

Fig. 5.14: After a 3D volume acquisition of a face from below, the maxilla with hard palate can be demonstrated by using a curved Omniview line, a VCI of 4 mm and in this case with maximum mode (compare with Fig. 5.15).

72 

 5 Multiplanar display I – Orthogonal Mode and Omniview Planes

Fig. 5.15: In this example a free-hand Omniview line was drawn along hard and soft palate and a VCI slice of 4 mm was selected. The palate and the uvula region are visualized.

Fig. 5.16: Omniview can also be used on the fetal heart, in this case in combination with color Doppler. In the orientation plane in the upper left panel, three lines were drawn and the result is seen in the three panels, as a four-chamber-view (yellow), a three-vessel-trachea view (magenta) and a frontal view of the cardiac valves (cyan).

5.5 Typical applications of Omniview planes 

 73

Fig. 5.17: On a 3D volume of a fetal brain in early gestation the Omniview line enables the visualization of the intracranial translucency (arrow).

Fig. 5.18: An interesting application of Omniview is the free-hand drawn line. On the example of an embryo at 9 weeks’ gestation, the line can demonstrate a stretched and projected fetus with brain and body.

74 

 5 Multiplanar display I – Orthogonal Mode and Omniview Planes

5.6 Conclusions Initially, navigation within a volume in the different planes needs a learning curve. Scrolling and spinning within the volume enables understanding as to how to obtain the plane of interest easily and to highlight the needed details. In our teaching experience, we observed that once the examiner gets used to the orientation within a volume, he will then easily start using it during routine scans. Particularly, the use of Omniview can be rapidly integrated into a live examination either on a 3D volume or during 4D examination. In this book, many examples are provided on the use of different multiplanar or Omniview tools.

6 Multiplanar Display II: Tomographic Mode 6.1 Principle Three-dimensional ultrasound is often associated with a spatial visualization of the fetus rather than with serial parallel planes as shown in tomographic mode. In the last years, however, increasing experience in 3D ultrasound has shown that one of the main advantages of a digital volume data set, is the post-processing, which makes it possible to obtain any 2D plane (see Chapter 5) or series of planes out of the volume block, especially when the fetus is not easily accessible for a live scan. Storing such a volume block further makes displaying parallel slices of an area similar to those with CT and MR imaging possible. This functions to demonstrate the adjacent structures or to show the extent of a lesion when present. Although an ultrasound examination is still a dynamic online examination generating live planes and instantaneously assessing them, we believe that in the future, tomographic ultrasound imaging will become increasingly important, not only for documenting reports but also in the growing field of ultrasound image automation. This chapter highlights different aspects of the tomographic mode display.

6.2 Practical approach In Chapter 2, orthogonal mode was presented with the display of three planes and the intersection point used for navigation within the volume. One of the other important navigation tools is the “translation” within the volume called scrolling (see Chapter 2). The user, interested in parallel scrolling, can alternatively apply the tomographic mode. Tomographic ultrasound imaging (TUI) is a multiplanar mode display of the volume as parallel planes similar to tomographic images obtained from CT and MR workstations. After choosing the region of interest the examiner controls the number of planes (slices) to display on the screen as well as the interslice distance. The region of interest is set, then the tomographic mode is activated and parallel slices are seen on the screen in addition to the reference image typically found in the upper left corner. The adjustable interslice distance is displayed in the upper corner orientation image. In tomographic mode, all the manipulation tools of the orthogonal mode can be used, such as navigation with the intersection dot, rotation of planes and scrolling within the volume. The user can apply these manipulation tools only in the reference plane, which leads to immediate change in the other planes displayed. In order to improve image quality it is recommended to reduce speckles by adding either the 3D-SRI filter or to activate the VCI-mode (see Chapter 4). Figures 6.1–6.10 demonstrate the different possibilities of tomographic mode. Figure 6.1 reveals the original volume displayed in orthogonal mode. With plane A

76 

 6 Multiplanar Display II: Tomographic Mode

Fig. 6.1: A 3D volume containing the fetal thorax and abdomen, here displayed in orthogonal mode, is used as a basic volume to demonstrate tomography mode in next Figs. 6.2–6.10.

Fig. 6.2: In tomography mode display, the upper left image is the orientation plane. The number of planes can be selected arbitrarily. A green asterisk marks the reference plane and two planes are in front and two behind the reference plane in this case. The interslice distance can be changed accordingly (see red square) and in this case 5.5 mm distance was selected.

6.2 Practical approach 

 77

Fig. 6.3: The same display as in Fig. 6.2, but plane B was activated. The orientation plane is in the upper left panel, with the demonstration of parasagittal planes from left to right.

Fig. 6.4: The same display as in Fig. 6.1 and 6.2, but in this case, plane C is activated with the demonstration of coronal planes from anterior to posterior.

78 

 6 Multiplanar Display II: Tomographic Mode

Fig. 6.5: The same display as in Fig. 6.2, but here the number of displayed slices was changed from 3×2 to 3×3. The interslice distance is now 2.5 mm.

Fig. 6.6: The same display as in Fig. 6.5 but the number of images was changed to 4×4 images.

6.2 Practical approach 

 79

Fig. 6.7: The same display as in Fig. 6.5 but in this figure the interslice distance was augmented to 7.5 mm. The figure in the middle remains unchanged, however the other 6 images change.

Fig. 6.8: In this example with 2×2 images the distance between the slices was increased to 9.5 mm.

80 

 6 Multiplanar Display II: Tomographic Mode

Fig. 6.9: In this example with 2×1 images, the four-chamber view can be observed. The solid line shows the plane of interest on the orientation plane and this approach can be used to scroll through the volume.

Fig. 6.10: Using the same presets as the previous figure, the section in the upper abdomen was selected and now shows the stomach.

selected as reference plane and tomography activated, parallel images to this plane are displayed (Fig. 6.2.). If the examiner chooses the B or C plane as a reference plane, the result is then parallel planes of lateral or coronal views as illustrated in Figs. 6.3 or 6.4. Figure 6.5 shows a tomographic mode image with the typical labeling, the reference plane with an asterisk and the adjacent planes with the – or + sign and a number

6.3 Typical applications in tomographic mode 

 81

which is the multiplication of the interslice distance. If the interslice distance is changed as shown in Fig. 6.6, the reference plane remains but the other images change. The number of slices displayed on the screen can be changed from 2 × 1, 2 × 2, 3 × 2, 3 × 3, 4 × 4 etc. as outlined in Figs. 6.5–6.10. Figures 6.9 and 6.10 show the scrolling within a volume using tomographic mode and the planes can be slightly adjusted by rotating selectively the X-, Y- or Z-axis.

6.3 Typical applications in tomographic mode Tomography of the fetal head, face and brain: Tomographic mode can be ideally used in the assessment of head, face, and brain. For fetal neurosonography either transabdominal (Fig. 6.11, 6.12) or transvaginal (Figs. 6.13, 6.14) volume acquisition can be applied. Tomography provides an overview wherein all intracerebral landmarks can be visualized at one glance (also refer to Chapter 16). The example in Fig. 6.11 illustrates an overview of the normal brain anatomy and Fig. 6.12 reveals a fetus with ventriculomegaly. In the adjacent planes, one can recognize the normal cerebellum and in another plane the dilated third ventricle. Therefore, in this overview, diagnoses such as Chiari II malformation, Dandy-Walker syndrome or holoprosencephaly can be ruled out and the likely diagnosis is aqueduct stenosis. The cavum septi pellucidi is clearly observed and identified in a coronal view in tomographic mode, and Figs. 6.13 and 6.14 illustrate normal and abnormal findings.

Fig. 6.11: 3D volume of a fetal brain demonstrated in tomography mode. Almost all of the information required is visualized at one glance in these axial planes.

82 

 6 Multiplanar Display II: Tomographic Mode

Fig. 6.12: Fetus at 19 weeks’ gestation with ventriculomegaly demonstrated in tomography mode. Following structures can be identified: the bilaterally dilated lateral ventricles, the falx cerebri, the normal appearance of the cerebellum and the dilated 3rd ventricle (3.Ventr.). Most likely, the underlying reason could be attributed to an aqueductal stenosis and the visualized details practically rule out diagnoses such as Chiari II malformation, Dandy Walker syndrome and holoprosencephaly.

Fig. 6.13: Transvaginal neurosonography in tomography mode with coronal planes. Typical structures such as the corpus callosum (CC), the cavum septi pellucidi (CSP) and insula are clearly observed and identified.

6.3 Typical applications in tomographic mode 

 83

Fig. 6.14: Transvaginal neurosonography in two fetuses with abnormal findings. Left: ventriculomegaly with dilatation of the anterior horns with cavum septi pellucidi. Right: Agenesis of septum pellucidum with fusion of the anterior lateral ventricles.

Tomography of thoracic and abdominal organs: Tomography is ideal for an overview of thorax and abdomen especially for the clear delineation of structures as lungs, diaphragm, heart and abdominal organs (Figs. 6.15–6.17). This allows for accurate assessment of the extent of a lesion, such as in hydrothorax (Fig. 6.16) or in hyperechogenic lung (Fig. 6.17). The tomography of the renal system (Figs. 6.18–6.20) is rarely typically used but can have a valuable application when an abnormality is identified. Information from the different abdominal organs can be best displayed in tomography of axial cross-sections showing the typical features such as liver, stomach, bowel, bladder, abdominal wall and kidneys (Fig. 6.6). Tomographic mode is an ideal way of documenting a lesion, particularly in the presence of fetal anomalies. Figures 6.21– 6.23 provide some examples, such as the double bubble sign in duodenal atresia, an ileus in one fetus and the extent of ascites in another fetus. Such image documentation can be of great value for follow up examinations. Tomography of the fetal heart: A complete cardiac examination has to be achieved in different planes therefore tomographic mode can be considered as an ideal tool to provide the complete picture (Figs. 6.24–6.26). Fetal heart tomography can be used either with a grayscale (Fig. 6.24) or with a color-Doppler (Fig. 6.25) STIC volume or rarely in 4D mode. Typical adjacent planes such as a four-chamber-view, a five-cham-

84 

 6 Multiplanar Display II: Tomographic Mode

Fig. 6.15: In this case, the fetal thorax heart, lungs, liver and diaphragm are well recognized in tomography mode.

Fig. 6.16: Mild pleural effusion as demonstrated in 2×1 tomography mode.

6.3 Typical applications in tomographic mode 

 85

Fig. 6.17: This figure illustrates a left sided hyperechogenic lung. Tomography mode displays the localization and the extent of the lesion, as well the difference in the echogenicity of the contralateral lung.

Fig. 6.18: This 3D volume acquisition reveals the lumbar region with both kidneys (arrows), here in tomography mode in transverse planes.

86 

 6 Multiplanar Display II: Tomographic Mode

Fig. 6.19: This 3D volume acquisition depicts the lumbar region with both kidneys (arrows), here in tomography mode in sagittal and parasagittal planes.

Fig. 6.20: Fetus with multicystic renal dysplasia displayed in tomography mode. An overview of the lesion can be better demonstrated with tomography mode.

6.3 Typical applications in tomographic mode 

 87

Fig. 6.21: Transverse 3D volume acquisition of the upper abdomen in tomography mode in a fetus with double bubble sign (*) and suspected duodenal atresia.

Fig. 6.22: Tomography mode of the abdomen in a fetus with ileus and bowel perforation. The stomach (*) can be seen in the lower planes.

88 

 6 Multiplanar Display II: Tomographic Mode

Fig. 6.23: Fetus with ascites and skin edema in cardiac failure. The extent of ascites can be assessed and documented with tomography mode in comparison to single images. These findings are better compared when using tomography mode, especially in follow up examinations.

ber-view and a three-vessel-trachea view can be well and rapidly demonstrated with this tool. More on fetal heart tomography is discussed in the chapter on the fetal heart. Tomography in early gestation: In early gestation, optimal information is best provided by combining transvaginal 3D ultrasound with tomographic mode (Figs. 6.27, 6.28). Due to the limitations of transvaginal probe manipulation, typical planes are more easily reconstructed from a volume than directly visualized in 2D. The acquisition of a 3D volume and its display in multiplanar mode, especially in tomographic mode, provides a good overview, particularly for regions such as the brain and face (Figs. 6.27, 6.28) also for the thorax, abdomen (Fig. 6.29) and others.

6.4 Conclusions Tomographic mode display provides an optimal overview of the region of interest. The all-in-one view of an organ, along with its neighboring structures, makes an accurate examination possible and is helpful when documenting a finding. The possibility of visualizing this region in a range of 2 to 16 successive planes at a time provides a flexibility to display the individual information needed. With more experience, typical

6.3 Typical applications in tomographic mode 

 89

Fig. 6.24: Tomography mode of a STIC volume of the heart. The structures can be visualized from the upper abdomen to the great vessels.

Fig. 6.25: Tomography mode of a STIC volume acquisition in color Doppler in the cardiac phase between diastole and systole. The four-chamber-view can be recognized in diastole (lower middle panel) and the systole is seen in the three-vessel-view; aorta (AO), left ventricle (LV), pulmonary artery (PA), right ventricle (RV).

90 

 6 Multiplanar Display II: Tomographic Mode

Fig. 6.26: Tomography mode of a STIC volume acquisition in color Doppler in a fetus with a rightsided aortic arch. The four-chamber-view is seen in the lower right panel. In the upper middle panel, the trachea can be identified (arrow) between aorta (Ao) and pulmonary artery (PA); left ventricle (LV), right ventricle (RV).

Fig. 6.27: First Trimester screening with the view of the profile in tomography mode. Nasal bone (yellow arrow), maxilla, mandible, both eyes (white arrows) and the posterior fossa with the intracranial translucency (*) are viewable together in one display.

6.3 Typical applications in tomographic mode 

 91

Fig. 6.28: Tomography mode of an axial view to the fetal brain at 12 weeks’ gestation with the brain hemispheres, the large choroid plexus and the posterior fossa.

Fig. 6.29: Tomography mode of the body of a fetus at 13 weeks’ gestation with the demonstration of diaphragm (yellow arrow), lungs, liver, stomach (*), kidneys (arrows) and the left-sided heart position.

92 

 6 Multiplanar Display II: Tomographic Mode

examination standards can be identified for the different body parts where volume depth and interslice distance can be saved in specific presets. Fetal heart and brain are ideal regions to be examined with this tool, and Chapters 16 and 20 present some abnormal findings demonstrated in these planes, and Chapter 19 presents some abnormal findings of fetal thoracic, gastrointestinal and renal organs.

7 Surface Mode Rendering and HD-Live 7.1 Principle Surface mode is, in general, the most popular and most commonly used 3D and 4D display mode. It is used to render an image of the surface of a structure, which is best visualized when positioned in the interface between fluid and that structure. Within the render box, the surface mode displays the most superficial layer nearest to the green rendering line (see Chapter 2). It is used to easily demonstrate the face, anterior or posterior surface of the body, the limbs or the complete fetus in early gestation. In addition, structures within the fetal body can be displayed with surface mode as cardiac chambers, intracerebral ventricular system, lungs and others. Surface mode can be used in different volume acquisitions as 3D static, 4D, or STIC or in combination with live or static Omniview.

7.2 Practical approach In order to acquire an adequate 3D or 4D volume, the examiner essentially aims for an initial 2D image with a high contrast between adjacent structures, as between the anechoic amniotic fluid and the echoic fetal skin. The presets of the 2D image were discussed in Chapter 1. Figures 7.1.–7.3 demonstrate the impact of a prior optimal gray scale “gain level” on the resulting acquisition. A dark amniotic fluid in 2D is a prerequisite for a good surface mode image as shown in Figs. 7.1 and 7.3. The positioning of the region of interest should allow, where possible, the insonation to be perpendicular on, and not parallel to, the surface to be rendered (as explained in Figs. 7.4 and 7.5). In Fig. 7.4, the arm is clearly observed and identified in 2D, however the 3D results are not satisfactory. Only a perpendicular insonation of the arm, as shown in Fig. 7.5, results in an adequate 3D image. Ideally the object of interest should lie horizontally and parallel to the camera (rendering) line (Fig. 7.6). During 3D static volume acquisition we also recommend choosing a wide box to include a larger area than the only selected region of interest (Fig. 7.7). This approach avoids missing fetal parts in the 3D rendered image as shown in Figure 7.7. Particularly in early gestation where the complete fetus can be visualized, a small volume box may result in a fetus with parts of arms and legs missing on the final 3D image. This, more commonly, applies to static 3D, whereas with 4D, the examiner can adjust the resulting image in live mode accordingly. After volume acquisition, the examiner begins “rendering” by changing the size of the box of the region of interest to include the organs to be displayed. The render box is then fixed and one of the different surface mode functions is activated. The quality of the display differs according to the presets of the ultrasound system and the

94 

 7 Surface Mode Rendering and HD-Live

Fig. 7.1: 3D volume of surface mode of a face. The preset of grayscale image is not optimized and demonstrates a low contrast. The grayish appearing amniotic fluid results in an inadequate 3D surface image. Additional changes should be made, as can be seen in Figs. 7.2 and 7.3.

Fig. 7.2: The same volume is shown in Fig. 7.1 but with post-processing increasing the gain and threshold suppresses the gray amniotic fluid and the 3D image of the face can be seen. The image is acceptable, but still appears too bright (see also Fig. 7.3).

7.2 Practical approach 

 95

Fig. 7.3: This is the same fetus as in Fig. 7.1, but in this case, the 2D image has been optimized prior to volume acquisition. The amniotic fluid now exhibits a good contrast during acquisition and the facial result in 3D surface mode is better than the example in Fig. 7.1.

Fig. 7.4: For a good 3D image, not only the contrast but also the insonation angle is important during volume acquisition. In the panel to the left, the hand is clearly visible in 2D but for a 3D volume acquisition the fingers are parallel to the ultrasound waves and are not well displayed on the 3D surface image as seen in the middle panel. The right image is the result after rotating the volume.

96 

 7 Surface Mode Rendering and HD-Live

Fig. 7.5: In comparison to Fig. 7.4 the insonation of hand and fingers (left panel) is now perpendicular and ideal for a 3D acquisition. The result is better than in Fig. 7.4.

Fig. 7.6: 3D volume acquisition of a face in surface mode with a good insonation. The approach is from the side, with both forehead and face almost horizontal.

examiner can switch between the different surface rendering modes and their combinations. The following modes are currently the most commonly used (also refer to Chapter 3): Surface smooth, Surface texture, Gradient light, and the combination of High-definition (HD-) live surface und HD-smooth. There are no “best” presets, since a mixture of different modes can be also a matter of “optical taste” or preference. Figures 7.8–7.10 illustrate some examples of common combinations used by the different individuals. Initially, the reader can try applying the 40/60 mixture of surface smooth and gradient light in this mode. Reducing the gray threshold and augmenting the transparency can improve the image. Magicut can be used (see Chapter 3) to remove structures in front of the region of interest, provided the removed part casts no shadows on the background image.

7.2 Practical approach 

 97

Fig. 7.7: 3D volume acquisition with a small volume box (upper panel). The face is seen, however, a part of the hand is missing due to the small volume box. Choosing a larger box (lower panel) in static 3D acquisition also makes it possible to include structures in the area of the region of interest. The hands can now be seen in the lower panel.

Fig. 7.8: 3D surface mode of fetal faces displayed with different rendering tools.

98 

 7 Surface Mode Rendering and HD-Live

Fig. 7.9: When visualizing a fetal face, additional neighboring structures can also be visualized. In these examples, an arm and ear, a foot, an umbilical cord (short arrow) and even a true knot of the umbilical cord (long arrow) are seen adjacent to the face.

Recently, we often have used the light source and changed its position to create a depth impression and spatial effect. In combination with dynamic rendering, the image can be improved by choosing the blue color, which lends the amniotic fluid a new dimension (see Chapter 3). The image quality is tremendously improved by the use of the skin-like presets called HD-live, especially when the image smoothness is increased. For more information on 3D facial rendering, please refer to Chapter 18, which is dedicated to the fetal face.

7.3 Typical applications of surface mode Head and face: The most common use of surface mode is for visualizing the fetal face, and this is separately discussed in Chapter 18. The face can be examined in 3D or 4D, at different gestational ages, from different perspectives and displayed with different colors (Figs. 7.8–7.10). In 4D, it is possible to appreciate various fetal facial expressions and movements, including swallowing, yawning, opening of the eyes and many others. In addition to the frontal view, a lateral view enables the visualization of the fetal profile and ear, which can by far be better assessed this way than by using con-

7.3 Typical applications of surface mode 

 99

Fig. 7.10: In surface mode, the fetal hands can be clearly seen, and their normal anatomy assessed (compare with Fig. 7.11).

ventional 2D. In the first half of gestation, the fontanels and sutures of the fetal skull are still large and can be easily seen with the surface mode by reducing the gain or increasing the transparency. The approaches regarding the “how-to” when it comes to displaying the face is discussed in Chapters 3 and 18. The post-processing manipulation of a fetal face volume has also been explained above in Chapter 3. Fetal limbs: Arms, legs, hands and feet can be visualized well from different angles and with different resolutions using surface mode. In most situations hands are in proximity of the face and are displayed with it (Figs. 7.9–7.11). An increase in the quality of acquisition provides then a better demonstration of fingers and toes. Further improvement of the image is achieved by adjusting the softness of the image and the position of the light source (Figs. 7.12, 7.13). Anomalies, such as the absence of extremities, polydactyly, clubfeet and so on, can therefore be visualized well using the 3D surface mode (Figs. 7.11–7.13). Demonstration of body surface: The fetal dorsal and ventral surface with the umbilical cord insertion can be easily visualized in early gestation. These can also be demonstrated in more advanced gestational ages, provided there is enough amniotic

100 

 7 Surface Mode Rendering and HD-Live

Fig. 7.11: Anomalies of the hands (arrows) displayed with surface mode: Polydactyly left, mitten-hand in Apert syndrome (middle) and absent hand (right).

Fig. 7.12: The feet are easily visualized in surface mode, often side-by-side or sometimes crossed (compare with Fig. 7.13).

7.3 Typical applications of surface mode 

 101

Fig. 7.13: 3D Surface mode of feet anomalies such as clubfoot (left), absent foot (middle) and foot edema (right) in a fetus with Turner’s syndrome.

fluid volume to facilitate a surface mode view. Fetal anomalies as gastroschisis, omphalocele (Fig. 7.14), spina bifida (Fig. 7.15), sacrococcygeal teratoma (Fig. 7.16) and other anomalies viewed on the surface can be clearly observed in surface mode. In gastroschisis, the bowel loops can be visualized in detail in early and late gestation, as will be demonstrated in Chapter 19. Identifying and displaying the gender can be ideally achieved using surface mode. Related anomalies, such as hypospadia or rare instances of hypertrophy of the clitoris, can be well delineated from normal external genitalia (Chapter 19). Overview of the complete fetus and fetuses of multiple gestations: Instead of magnifying to limit the 3D view to the face, limbs or other fetal parts, the examiner can also attempt to display the complete fetus. Ideally, a complete view of the fetus is

Fig. 7.14: 3D surface mode in two fetuses with omphalocele (arrow), left at 12 and right at 18 weeks’ gestation.

102 

 7 Surface Mode Rendering and HD-Live

Fig. 7.15: 3D surface mode of the back of a normal fetus (left), a fetus with myelomeningocele (middle) and a fetus with myeloschisis (right).

possible between 8 and 18 weeks’ gestation (Figs. 7.17–7.16). At later stages of gestation, the fetus is generally too large to be completely visualized in one image. In multiple gestations, surface mode is ideal for obtaining a complete overview of the fetuses (Fig. 7.18). The amniotic membrane in monochorionic twins is often too thin to demonstrate, but on the other hand can be easily differentiated from a thick separating chorion/amnion layer in dichorionic twins. The position and number of fetuses can be visualized well using the surface mode. Placenta, umbilical cord and amniotic membranes: The overview provided with surface mode used to display the fetus can also demonstrate the surrounding structures such as the placenta, the umbilical cord at its insertion and course, amniotic bands and various uterine anomalies.

Fig. 7.16: 3D surface mode of the complete fetus at 22 weeks’ gestation (left). By comparison, the right panel illustrates a fetus with a sacrococcygeal teratoma.

7.3 Typical applications of surface mode 

 103

Fig. 7.17: 3D surface mode in two fetuses at 13 weeks’ gestation showing the complete body.

Fig. 7.18: 3D surface mode in twin pregnancies.

Visualization within the body, such as the heart, brain and other organs: Surface mode can be applied to internal body organs, especially the heart (Chapter 20), the brain (Chapter 16) thorax and abdomen (Chapter 19). When applied to the heart, the cardiac cavities can be well seen in the four-chamber-view. Cardiac phases, diastole and systole, can be well identified in a STIC acquisition as well (Fig. 7.19). Other planes can be well recognized if needed, such as the five-chamber-view or the three-vesselsview and en-face views of the atrioventricular or semilunar valves. In examining other organs, surface mode is rarely used in normal fetal examinations. However, in some anomalies, surface mode can be utilized, especially when an increased fluid interface is present as found in ascites (Fig. 7.20), duodenal atresia with double bubble sign (Fig. 7.21 left), hydrothorax (Fig. 7.21 right), hydrocephaly (Fig. 7.22), megacystis, cystic kidneys, hydronephrosis and others.

104 

 7 Surface Mode Rendering and HD-Live

Fig. 7.19: STIC volume acquisition of a heart with a view into the ventricles in surface mode, in this case, during systole with closed atrioventricular valves (arrows) and in diastole with opened valves.

Fig. 7.20: Ascites in 3D surface mode with view into the ascites on liver and bowel. Note the position of the “green projection line” placed into the ascites.

(a)

(b)

Fig. 7.21: Surface mode in two fetuses with malformations. Axial view at the level of the abdomen in a fetus (a) with duodenal atresia and double bubble sign (*). Axial view at the level of the thorax in a fetus (b) with pleural effusion (*). The left lung (LL) and the heart (H) are shifted to the right; right lung (RL), left (L).

7.3 Typical applications of surface mode 

 105

Fig. 7.22: Two fetuses with hydrocephaly after a transvaginal 3D acquisition with the demonstration of the dilated lateral ventricles in surface mode. The right figure demonstrates how the ventricles communicate across the midline and the choroid plexus (*) hangs across the midline to the opposite site (arrow) in severe ventriculomegaly.

7.4 Conclusions It is recommended that examiners acquire the surface mode manipulation skills, since this is the most widely used application in 3D fetal imaging. The documentations of normal body surface findings are becoming increasingly important to complete a 2D assessment of a fetus. In abnormal findings, surface mode can rapidly provide an overview of the anomaly encountered, thus making it more understandable for patients and peers.

8 Maximum Mode Rendering 8.1 Principle The maximum mode is mainly used for the spatial visualization of hyperechogenic structures as the fetal bones. In this transparency mode all hyperechogenic structures found within the render box are highlighted and displayed as a projection. In the upper panel of Figure 8.1, we see the face of a fetus rendered with surface mode and after activation of maximum mode (lower panel) the skin is not seen anymore and only hyperechogenic signals from the facial bones are displayed. Another example is provided in Fig. 8.2. In general, cranial bones, ribs and other curvilinear bones cannot be properly observed in a single 2D plane, and one of the advantages of maximum mode is the ability to demonstrate a projection of the bones.

(a)

(b)

Fig. 8.1: 3D volume of a face in surface mode (a) and after switching to maximum mode (b) with a choice of a narrow volume box. In the lower image, one can recognize the individual facial bones with the metopic suture (arrow), orbits, nasal bones, maxilla and mandible.

8.2 Practical approach  

 107

Fig. 8.2: Demonstration of an arm in surface mode (left) and in maximum mode (right). For maximum mode effect, the size of the box was reduced to only include the arm, with the result that structures behind the arm are not seen.

8.2 Practical approach During volume acquisition, care should be taken to record a volume large enough to include the complete region of interest. A better result is achieved if the 2D image gain is reduced and the contrast increased during volume acquisition to allow bones to appear “bright” and the surrounding tissue as “dark”. In early gestation, it often appears difficult to display the bones in 3D due to their reduced ossification and in third trimester the skin of the fetus has an increased echogenicity and often overlaps the information from the bony structures. Therefore, in our experience, maximum mode is best performed between 15 and 25 weeks’ gestation enabling clear bones visualization. An acquisition box large enough to include the region of interest is selected (Fig. 8.3) once the preset of the 2D image is adjusted. In general, it is better to use a flat box depth when only including the superficial bones with very little information from the neighboring tissue or skin (Fig. 8.4 and 8.5). The resolution of the 3D volume (“low”, “mid1” to “maximum”) depends on the duration of volume acquisition as shown in Fig. 8.6. Maximum mode is not only used in 3D static and 4D volume acquisition (Fig. 8.6), but also with VCI-Omniview (Figs. 8.7, 8.8) and a slice thickness of 15 to 20 mm is recommended in all these cases. Generally speaking, a “maximum mode” of 100 % is selected, but occasionally a mixture of maximum with surface mode (80/20 %) with an increased threshold can provide a better image. An interesting tool is also the examination with VCI-A (see chapter 4) in combination with maximum mode: The 4D examination, ideally performed using a matrix probe, enables the visualization of the bones of interest using a slice of 15–20 mm thickness (Figs. 8.9–8.11).

108 

 8 Maximum Mode Rendering

Fig. 8.3: Volume acquisition of a spine with ribs prior to display with maximum mode, but here first seen in the orthogonal display. Note that the image is rather dark with increased contrast in order to better highlight the bones.

Fig. 8.4: In this example, the size of the volume box is still large (double arrow). In such a case, all signals within the box are calculated while only the information from the bony structures is needed. A better result can be achieved with a narrow box, as illustrated in the next figure.

8.2 Practical approach  

 109

Fig. 8.5: Conversely to Fig. 8.4 the volume box is now narrow (double arrow) to mainly include the bony structures. The 3D image now has a better contrast and reveals more details.

Fig. 8.6: 3D acquisition of a spine in two different resolutions and maximum mode rendering. The acquisition was achieved in the upper panel using a “mid1” quality, whereas and “max” was used in the lower panel. From the results it is possible to recognize the different 3D image resolutions.

110 

 8 Maximum Mode Rendering

Fig. 8.7: Use of Volume Contrast Imaging (VCI) and Omniview tool as VCI-Omniview during a 4D examination. During the 2D examination of the spine, the Omniview line is placed along the spine with a slice thickness of 17 mm and the maximum mode selected (see next figure).

Fig. 8.8: VCI-Omniview, as explained in previous figure, here provided in different resolutions at acquisition (arrows) during 4D examination. The image on the left had a resolution of “low”, the middle image “mid” and right image had a “high” resolution.

8.2 Practical approach  

 111

Fig. 8.9: Volume Contrast Imaging (VCI) of the A-plane, as VCI-A-acquisition (arrow). Instead of obtaining 2D images, the examination is performed in 4D with the live acquisition of a slice (here, 8 mm thickness). In this case, maximum mode is activated. A hand and fingers are displayed in different 2D planes (left), but in a VCI-A slice, both can be displayed together in one slice (right).

Fig. 8.10: VCI-A (see explanation in Fig. 8.9) of the spine and scapula in a slice thickness of 12 mm in maximum mode.

112 

 8 Maximum Mode Rendering

Fig. 8.11: VCI-A of the spine in a normal fetus in the upper and in a fetus with a hemivertebra (arrow) in the lower image.

8.3 Typical applications of maximum mode In the following, some clinical aspects are briefly presented and abnormal cases are demonstrated in Chapter 17 and 18 on the fetal skeleton and fetal face. Visualization of spine and ribs: A dorsal view with a narrow 3D/4D box over the spine is ideal, with VCI-Omniview as a straight or curved line or with VCI-A (Figs. 8.4–8.8 and Figs. 8.10–8.13). Figure 8.12 displays ribs with 13, 16 and 21 weeks’ gestation and Figure 8.13 shows a dorsal and lateral view of the spine. In this view, spine shape and symmetry of vertebral bodies are well seen, a view which is ideal in the demonstration of spina bifida, hemivertebra, kyphoscoliosis, ribs number and others (Fig. 8.14). Also refer to Chapter 17. Fontal view of the face: An acquisition of a volume of the face from the front enables the visualization of the bony face (Fig. 8.15) with frontal bones with metopic suture, orbits with nasal bones, maxilla and mandible. Absent nasal bones (Fig. 8.16), abnormal metopic suture, facial clefts, abnormal orbit size are the main fields of interest (see Chapter 18). Cranial bones and sutures: Maximum mode is ideal for visualizing the curved shape of cranial bones with sutures and fontanelles (Fig. 8.17). This approach is also excellent for the demonstration of wide sutures, abnormal ossifications as the prematurely closed sutures in craniosynostosis.

8.3 Typical applications of maximum mode 

 113

Fig. 8.12: Fetal spine and ribs in a fetus at 13 (left), at 16 (middle) and at 21 weeks’ gestation (right). Note the increased ossification of spine and ribs with advancing gestation.

Fig. 8.13: Spine with a view from dorsal (left) and from lateral (right) in maximum mode.

114 

 8 Maximum Mode Rendering

Fig. 8.14: Rib numbers and vertebral bodies: In the left image, one can count typically 12 rib pairs, in the fetus in the middle there are 11 rib pairs and the right fetus displays evidence of a hemivertebra (arrow) with a kinking of the spine.

Fig. 8.15: During a 4D examination, placing the VCI-Omniview line with a maximum mode rendering directly on the face is possible. In this example, the slice thickness is 12 mm. The fetal face exhibits details, as was illustrated earlier in Fig. 8.1.

8.3 Typical applications of maximum mode 

 115

Fig. 8.16: Fetus with absent ossification of nasal bone in 2D (left) and in 3D maximum mode from lateral (middle) and from anterior (right).

Fig. 8.17: The cranial bones (left) can be visualized well with a lateral insonation and in maximum mode display. Following bones are recognized: Frontal (F), parietal (P), sphenoid (S), Temporal (T) and occipital bones (O) as well as the mandible (M). Right: in a 3D acquisition from the top with maximum mode the big fontanelle (*) is well seen.

116 

 8 Maximum Mode Rendering

Fig. 8.18: This fetal arm lies horizontally on the 2D image (left). This is the ideal position to be examined in 3D/4D (right), in this case with VCI-Omniview with a slice thickness of 12 mm and maximum mode.

Visualization of long bones and limbs: The long bones of arm and leg together with hands and feet can be observed clearly using maximum mode (Fig. 8.18). The 3D rendering is ideal when long bones lie horizontally with an almost perpendicular insonation (Fig. 8.18). The proportion of bones, skeletal anomalies, clubfeet and abnormalities in hands and feet are important questions of interest.

8.4 Conclusions Maximum mode is the ideal 3D tool for demonstrating the different parts of the fetal skeleton. The easiest way to learn is to start with a static 3D of the fetal spine and long bones. Best results are achieved in a perpendicular insonation of horizontal lying bones. A thin slice either in 3D static or in VCI-Omniview enables the selection of the region of interest. Chapter 17 discusses some 3D skeletal anomalies in greater detail.

9 The Minimum Mode 9.1 Principle In general, fluid-filled structures are easily recognizable on ultrasound due to their echolucency and sharp borders to adjacent neighboring structures. The advantage of the transparency minimum mode is the ability of rendering information within the volume box by highlighting hypo- or anechoic structures. Other tools used for the demonstration of echolucent organs are inversion mode (see Chapter 10) and silhouette mode (Chapter 11).

9.2 Practical approach Before volume acquisition, care should be taken in preparing the 2D image by optimizing the contrast in a way that fluid is seen as “black” in color without artifacts and speckles (Fig. 9.1). Ideally, the acquisition is achieved from a perspective with the lowest shadowing from bones as possible, since shadows will act on the rendered image in the same manner as fluid. For a volume to be rendered with a minimum mode, the examiner should select a flat volume box to primarily include solely the organs of interest, with only very little information from additional neighboring tissue (Figs. 9.1–9.3). Within the box, the presence of amniotic fluid should be avoided, as it casts a large black shadow (Fig. 9.1). In other words the anterior and posterior line of the volume box should be placed in the tissue and not in the amniotic fluid (Fig. 9.2).

Fig. 9.1: The render box has been placed over the fetal abdomen and the minimum mode activated. The box is deep and includes amniotic fluid, and therefore the image appears almost black in minimum mode, while no structures are identifiable (see next Fig. 9.2).

118 

 9 The Minimum Mode

Fig. 9.2: The Fig. 9.1 render box is now less deep and less amniotic fluid is present in the volume box. Thorax and abdomen contours can be better recognized (see further Fig. 9.3).

Fig. 9.3: The render box is now flat and minimum mode can reveal the hypoechoic organs such as the heart (H), stomach (*), gallbladder (GB) and bladder (BL).

A good result is often achieved with a “minimum mode” combined with “X-Ray mode” (80/20 % mix). However, the “threshold” should be increased and in some occasions post-processing change of contrast and gain may improve the image result. A rotation along the vertical Y-axis often provides a better 3D effect in the region of interest (Figs. 9.4 and 9.5).

9.3 Typical applications of minimum mode 

 119

Fig. 9.4: Thorax and abdomen in minimum mode in anterio-posterior (left) and lateral projection (right). In both views, typical structures are demonstrated, such as the stomach, gallbladder, heart position and umbilical vein.

Fig. 9.5: Heart, lungs and diaphragm are projected in minimum mode in anterio-posterior and in lateral projection

9.3 Typical applications of minimum mode Typical structures that are displayed with minimum mode in the normal fetus are the echolucent organs as bladder, stomach, gallbladder, umbilical vein and portal venous system in the abdomen (Figs. 9.1–9.3), in thorax the heart with the great vessels (Figs. 9.4, 9.5) and in the head the intracerebral ventricular system. Since some fetal anomalies are often associated with increased fluid accumulation, these can be clearly demonstrated, not only in minimum mode (Figs. 9.6–9.15), but also with inversion or silhouette modes (see Chapter 10 and 11).

120 

 9 The Minimum Mode

Fig. 9.6: In comparison to the fetus in Fig. 9.4 (left), this fetus has a partial situs inversus and the 3D minimum mode display illustrates the stomach (*) on the right side (R) and the heart on the left (L).

Intraabdominal organs with vasculature: One of the typical approaches easily used in combination with minimum mode is the frontal acquisition of abdomen and thorax (Figs. 9.4, 9.5). The 3D view is then either from frontal or lateral with the projection of situs with stomach, heart, diaphragm but also umbilical vein with gallbladder, inferior vena cava and descending aorta. In this view, situs inversus or ambiguus can be well recognized (Fig. 9.6). In a lateral view, it is easy to differentiate an abnormal course of the umbilical vein in agenesis or atypical course of the ductus venosus from a normal finding (Fig. 9.7). The absence of the stomach filling or better a dilated stomach as observed in double bubble sign (Fig. 9.8) can be well-documented using

Fig. 9.7: Abnormal course of Ductus venosus (arrow) with color Doppler (left) and in projection in minimum mode (right); umbilical vein (UV), aorta (AO), inferior vena cava (IVC).

9.3 Typical applications of minimum mode 

 121

Fig. 9.8: Stomach in duodenal atresia with double bubble sign, as illustrated in 2D (left), in 3D minimum mode in coronal (middle) and lateral projection (right).

Fig. 9.9: Bilateral pyelectasis in axial view in 2D (left) and in minimum mode in coronal projection (right).

Fig. 9.10: Severe hydronephrosis with megaureter and vesico-ureteral reflux displayed in minimum mode.

minimum mode. Other abnormal conditions in the abdomen with increased fluid are the presence of megacystis, hydronephrosis with or without a dilated ureter (Figs. 9.9, 9.10), multicystic renal dysplasia (Fig. 9.11) and others. Ascites is better displayed using surface mode, as illustrated in Chapter 19. Thorax with heart and great vessels: A frontal acquisition of the thoracic cavity using minimum mode reveals the heart shape with the crossing of the vessels, as well as the both slightly echogenic lungs and the dark border of the diaphragm (Fig. 9.5). A lateral view makes the demonstration of the crossing of the great vessels with the

122 

 9 The Minimum Mode

Fig. 9.11: Multicystic renal dysplasia with many cysts of different size in 2D (left) and in minimum mode (right).

Fig. 9.12: A fetus with unilateral pleural effusion in minimum mode. Left) the anterio-posterior projection with the heart (H) shifted to the right (R), and stomach (*) on the left (L). The image on the right shows the projection from a lateral view.

aortic arch possible (Figs. 9.4, 9.5). Abnormal findings such as lungs cysts, hydrothorax (Fig. 9.12), and stomach position in diaphragmatic hernia (Fig. 9.13) and others can be clearly observed and identified with this rendering mode. Heart defects, however, are more difficult to demonstrate, unless the size or course of the great vessels is affected (Fig. 9.14). For this purpose, we generally prefer to use inversion mode.

9.3 Typical applications of minimum mode 

 123

Fig. 9.13: Left: A fetus with a left-sided congenital diaphragmatic hernia with the stomach (*) left (L) in thorax and the heart (H) shifted to the right (R) in 2D. Right: anterio-posterior projection in minimum mode revealing the position of the stomach in the thoracic cavity at the same level as the heart and to its left. Compare with a normal finding in Fig. 9.4 to the left.

Fig. 9.14: 3D lateral projection in minimum mode of a fetal heart in complete transposition of the great arteries with the parallel course of aorta (AO) and pulmonary artery (PA) arising from the right (RV) and left ventricle (LV) respectively. Stomach (*) is seen in the abdomen.

Intracerebral ventricular system: The fluid-filled ventricular system can also be well demonstrated with minimum mode (Fig. 9.15). However, bone shadowing is a main limitation of this application in the 2nd and 3rd trimester, and we therefore recommend the volume acquisition ideally be performed through the fontanelle. Minimum mode can ideally be applied in abnormalities with increased fluid accumulation as found in ventriculomegaly, hydrocephaly, holoprosencephaly, absent septum pellucidum and others. An interesting application is the demonstration of the cerebral

124 

 9 The Minimum Mode

Fig. 9.15: Projection with Omniview of the intracerebral ventricular system at 9 weeks’ gestation. In minimum mode both lateral ventricles (*) can be well identified as well as the developing third (3v) and fourth (4v) ventricles.

ventricles in early gestation (Fig 9.15), even before 10 weeks’ gestation. At this stage, skull bones are only slightly ossified and the ventricles adequately filled with fluid. A combination of minimum and X-Ray mode is good for obtaining an appropriate contrast image as shown in Figure 9.15. This approach has been replaced in recent years by silhouette mode (see Chapter 11).

9.4 Conclusions Minimum mode can be used as a projection highlighting the anechoic structures across a volume, similar to an X-Ray examination in radiology. Transparent structures and their neighboring organs can be clearly observed and identified and abnormally increased fluid in the fetal body can be well displayed. Diagnoses as hydronephrosis, hydrothorax, double bubble, cystic lesions and hydrocephaly can be visualized using minimum mode. Interestingly, hyperechoic lesions as seen in hyperechogenic lungs or kidneys can be well highlighted in comparison to neighboring tissues as well. The two prerequisites for a good result are, on the one hand, avoiding bone shadowing during acquisition and on the other, using a narrow display box mainly avoiding the presence of amniotic fluid. In recent years, however, minimum mode has been less used as a single render mode and other transparent modes have gained in preference.

10 The Inversion Mode 10.1 Introduction In Chapter 9 we explained the principle and clinical use of minimum mode display. Inversion mode rendering, on the other hand, starts from the minimum mode rendering and merely inverts the color of the information (similar to negative/positive film), thus presenting the hypoechoic structures as echogenic solid structures. It blackens most of the surrounding tissue information. The image is similar to a 3D digital cast of the structures of interest and the spatial depth is better appreciated in comparison to minimum mode. As opposed to minimum mode, Magicut (see Chapter 3) can be applied on an inversion mode volume to remove artifacts around the region of interest.

10.2 Practical approach Similar to minimum mode acquisition, the volume should ideally be acquired with as less shadow as possible, since shadow will be displayed on inversion mode as echogenic information. Before volume acquisition, the image contrast should be increased to have a clear black-gray discrimination and a better border recognition. The volume depth for inversion mode should include the complete region to be demonstrated. After a volume is acquired and the inversion mode selected, the image turns to black with some information displayed in inversion mode (Fig. 10.1). The size of the box has to be adapted to include the region of interest and then the „threshold“ level has to be increased (level 70 or more) until the result of the inversion appears on the screen (Fig. 10.2). In some systems the preset of inversion is the color „light“, but the authors prefer to use “gradient light” or HD-live, which can be well combined with a surface mode. Magicut can be used to remove additional neighboring artifacts (see Chapter 3, Fig. 10.3) and the “gain” and “threshold” buttons can be used to improve the image. Figures 10.1 to 10.3 illustrate an example of step-by-step image display. Inversion mode can be applied to 3D and 4D volumes. Recently, with the advent of the electronic matrix probe, inversion mode has also become available for use in combination with the VCI-A live scanning mode (see Chapter 4) (Fig. 10.4). In this mode, a slice thickness between 1 and 20 mm can be selected and displayed in inversion mode. All anechoic spaces such as the heart, great vessels, stomach and others can be displayed in a live scanning mode. A good example can be seen in Fig. 10.4; the method is discussed briefly in Chapter 20.

126 

 10 The Inversion Mode

Step 1: Activate inversion mode and select gradient light

Fig. 10.1: Main steps for a 3D rendering with inversion mode as demonstrated on a STIC volume of a heart. The render box is placed over the heart, the inversion mode activated and gradient light is selected (continued in Fig. 10.2) Step 2: Increase the gray threshold

Fig. 10.2: In a second step, the threshold level is increased, for example from 30 to 60 (arrow), and gain level is adapted until the targeted anatomic details are visualized (continued in Fig. 10.3). Step 3: Remove artifacts with Magicut

Fig. 10.3: In a third step, artifacts from ribs shadowing and other interfering structures are erased with the Magicut electronic scalpel and the image is finalized by adjusting threshold and gain.

10.3 Typical applications of inversion mode 

 127

Fig. 10.4: 4D examination of thorax and abdomen in a longitudinal view with VCI-A in combination with inversion mode with the electronic transducer. Choosing an 8 mm thickness layer the examiner can see at the same time the aorta (Ao), the inferior vena cava (IVC), the umbilical vein (UV) and the heart in a projection highlighted with the inversion mode rendering.

10.3 Typical applications of inversion mode There are many similarities between the use of minimum and inversion mode and we recommend referring to Chapter 9 for the organs or regions of interest. Thorax und Abdomen: In inversion mode, the anechoic structures can be visualized in the thorax and abdomen under normal and abnormal conditions. Typical structures are the stomach (Fig. 10.5), bladder, gallbladder (Fig. 10.6) and different vessels in thorax and abdomen (Figs. 10.4, 10.7). Intracerebral ventricular system: Fluid filled ventricular system can be demonstrated with inversion mode especially in the early embryonic period (Fig. 10.8). One of the main limitations of inversion mode is the lack of the demonstration of the surrounding structures. However, inversion mode has been used clinically to study the embryology of brain development, especially the ventricular system between 8 and 13 weeks of gestation (Fig. 10.8) and differentiated from conditions as holoprosencephaly. Later in gestation inversion mode can be used in conditions with increased fluid accumulation, as in ventriculomegaly, which is best demonstrated after transvaginal volume acquisition (Figs. 10.9, 10.10). Urogenital system: Abnormal findings of kidneys when associated with fluid accumulation can be clearly demonstrated and identified with inversion mode. Typically con-

128 

 10 The Inversion Mode

Fig. 10.5: Stomach of a normal fetus in minimum mode (Left) and in inversion mode (middle image). By comparison, the image to the right reveals a stomach with the double bubble sign that typical of duodenal atresia. The gallbladder (arrow) is also displayed.

Fig. 10.6: Axial view of the abdomen with gallbladder in minimum mode (left) and inversion mode (right).

Fig. 10.7: Left: 3D volume acquisition of an axial cross-section of the fetal abdomen at the level of stomach (*) and liver at 33 weeks’ gestation with the umbilical vein (UV) and hepatic vessels. Middle: The display in inversion mode enables the demonstration of stomach (*), hepatic veins (HV) and umbilical vein (UV) with the portal system. Right: In this case, the stomach and hepatic veins were digitally removed and the umbilical vein can thus be recognized with its connection to the portal sinus (PS).

10.3 Typical applications of inversion mode 

 129

Fig. 10.8: The intracerebral ventricular system of a 9 week-old embryo as displayed in minimum mode (left) and in inversion mode (right); lateral ventricle (LV), rhombencephalon (Rb), third ventricle (3V).

Fig. 10.9: Ventriculomegaly in a fetus at 30 weeks’ gestation after transvaginal 3D volume acquisition and inversion mode rendering. The ventricular system is presented along with other neighboring information that is mainly the result of shadowing (upper left). After the removal of artifacts with Magicut, only both lateral ventricles (LV) are displayed with the cavum septi pellucidi (CSP) between, as seen from a lateral view (upper right) and from a cranial view (lower panel).

130 

 10 The Inversion Mode

Fig. 10.10: Ventricular system of a fetus at 20 weeks with agenesis of the septum pellucidum after a transvaginal 3D volume acquisition. Left: In 2D, both anterior horns (*) of lateral ventricles are communicating due to the absence of the laminae of the septum pellucidum. Right: After inversion mode rendering and manipulation with Magicut, the ventricles communicating along the midline are clearly observable from a cranial view. Compare with the lower image in Fig. 10.9.

ditions examined are multicystic renal dysplasia (Fig. 10.11), hydronephrosis (Fig. 10.12, 10.13), megacystis and others. Some examples are presented in Figs. 10.10 to 10.13. Heart and great vessels: One of the main fields of inversion mode is the heart and its neighboring vessels, where the spatial orientation can be clearly demonstrated (see Figs. 10.1–10.3). Inversion mode can be used in both static 3D and in STIC (Figs. 10.14) or in combination with VCI-A in normal and abnormal cases (Figs. 10.3, 10.4, 10.14–10.16). A good contrast volume can be displayed from the front to show the atria, ventricles and the crossing of the great vessels. Inversion mode can be displayed in surface smooth, gradient light or HD-live mode.

Fig. 10.11: Multicystic renal dysplasia in 2D (left) and in inversion mode (right). The individual cysts are clearly observable in 2D but are spatially better demonstrated in 3D inversion mode with HD-live color.

10.3 Typical applications of inversion mode 

 131

Fig. 10.12: Hydronephrosis in a fetus with vesico-ureteral reflux in 2D (left) and in inversion mode (right). Dilated pelvis (*), the calyces and ureter (U) are well recognized.

Fig. 10.13: Hydronephrosis in a fetus with vesico-ureteral reflux displayed in minimum mode (left) and in inversion mode (right) with the spatial visualization of the dilated ureter (U), pelvis (*) and calyces.

Fig. 10.14: STIC acquisition of two hearts as revealed by inversion mode rendering. Left: Normal heart with the right (RV) and left (LV) ventricle and the normal crossing of aorta (AO) and pulmonary artery (PA). In the fetus on the right, there is a transposition of the great arteries with parallel course of AO and PA.

132 

 10 The Inversion Mode

Fig. 10.15: 4D acquisition in a cross section of the mediastinum at the level of great vessels with VCI-A in inversion mode (electronic matrix transducer). In this projection, the crossing of the great arteries is well recognized; aorta (AO), pulmonary artery (PA).

Fig. 10.16: 4D acquisition in an oblique cross-section of the mediastinum at the level of great vessels with VCI-A in combination with inversion mode (electronic matrix transducer). With this projection the parallel course of the great vessels in this fetus with transposition of the great arteries is well recognized; aorta (Ao), left ventricle (LV), pulmonary artery (PA), right ventricle (RV).

10.4 Conclusions Fluid in the fetal body with a good discrimination to its neighboring tissue and not in the shadow of bones is the ideal region to be displayed with the inversion mode. The image is similar to a digital cast and can be improved by changing the direction of light. A prerequisite for a good image is an optimized contrast 2D image before acquisition and a good balance when using the threshold and the gain buttons. Often, a Magicut is needed to remove additional artifact information.

11 The Silhouette Tool 11.1 Principle When rendering structures within a 3D render box, the examiner can generally choose between surface and transparent modes or a mixture of both. New software introduced in 2014 enables (see Chapter 7) the demonstration of contours of the structures present in the volume (Fig. 11.1). This tool is called silhouette, and in the actual software can only be used in combination with the HD-live rendering mode. The intensity of silhouette contour can be increased gradually (currently from 0–100). For this purpose, transparency and gain functions are used in the optimization of the image results. This chapter shares the authors’ first experiences when using this application. We believe that the potential of this new method has not been yet fully exploited.

11.2 Practical application A prerequisite for using the silhouette tool is the activation of HD-live mode. The result essentially depends on the size and the amount of the information within the render box. The examiner can select the intensity of the silhouette level, depending on the structures to be demonstrated. Silhouette images range from the mild smoothing of a

Fig. 11.1: Transvaginal 3D volume acquisition of a fetus at 12 weeks’ gestation displayed with HD-live and silhouette effect. The placenta anterior to the fetus has only been partly removed with Magicut, but the silhouette effect enables creating a partly transparent effect with the placenta. The silhouette is level 50 of transparency.

134 

 11 The Silhouette Tool

Fig. 11.2: Profile of a fetus with HD-live and different silhouette levels. The image to the left displays the profile without the silhouette effect (0) and then increasing the level of silhouette transparency from 25, to 50 and 70. Note the increasing transparency and smoothing of the image in addition to the wax-like and glossy effect.

Fig. 11.3: Embryo at 8 weeks’ gestation without silhouette effect (left), then with smooth silhouette effect (level 40), where the contours are well seen (middle) and then an almost transparent embryo when silhouette is high (level 80) (right). In the right embryo, the intracerebral ventricles have started to become visible.

surface mode image providing contours that appear wax-like (Level 0–10) (Fig. 11.2) to the sole display of contours with almost complete transparency of the surrounding structures (Level 60–100) (Fig. 11.3). The silhouette tool is actually the most powerful tool to effectively highlight the contours of anechoic structures within a render box. In early pregnancy, the fetus or embryo can be visualized completely with the silhouette tool (Fig. 11.1, 11.3, 11.4). Fluid accumulations in the body such as a thickened nuchal translucency (Fig. 11.5) or cystic structures or other anomalies can be highlighted very well using this function. A silhouette image can be easily manipulated with the Magicut electronic scalpel. This can be applied in two ways. In one approach, the image is first optimized in the

11.2 Practical application 

 135

Fig. 11.4: Fetus at 12 weeks’ gestation with silhouette effect and different light source positions. Left image with light source from cranial and right image with light source from posterior.

Fig. 11.5: Fetus with thickened nuchal translucency at 11 weeks’ gestation. In the left image in surface rendering mode, a thickened region of the neck is slightly evident (arrow). In the right image and using silhouette, increased nuchal fluid can be visualized well (arrow).

HD-live function (e.g., a fetus in first trimester or a face in the second trimester), and then unneeded information is removed using Magicut. In the next step the silhouette tool is used to emphasize the contours making the structures anterior to the region of interest more transparent (Figs. 11.3, 11.5). In another approach, the silhouette function is activated in a first step on the raw volume data set (e.g. in early pregnancy) (Fig. 11.6) and the transparency level is increased until only contours are visible (Fig. 11.6b). At this point, Magicut can be used to easily remove the unneeded structures that can be well differentiated from their neighboring tissue (Fig. 11.6c). In a next step, silhouette is reduced to optimize the image (Fig. 11.6d). The example in Figure 11.6 illustrates this step-by-step approach.

136 

(a)

(b)

(c)

(d)

 11 The Silhouette Tool

11.3 Typical applications of the silhouette tool 

 137

One of the important steps that significantly improve the silhouette image is the change of the light source position (Fig. 11.6d) (see Chapter 3). In the right lower corner of the image, the light source position is displayed. The user can lighten the volume from the side, from above or for thin volumes even from behind resulting in different light effects.

11.3 Typical applications of the silhouette tool In our first experiences, we did achieve good results when applying it to some conditions, which are presented below. We therefore encourage to the user explore new applications using the silhouette tool. Early pregnancy: Ranging from the visualization of the 5 mm embryo up to the fetus at 14 weeks’ gestation, the silhouette can be applied during the complete first trimester to provide surprisingly impressive images (Figs. 11.1, 11.3–11.8). A prerequisite, however, is an excellent 3D volume quality, which usually is achieved with a transvaginal transducer. Ideally, the volume size is selected as large as possible, which in turn allows for a better visualization of the embryo/fetus and its surrounding area. The amniotic cavity can be easily visualized with this tool, thus contributing to a good differentiation of a multiple pregnancy. The intracranial structures can also be visualized well in this time interval. Body contours: Body contours are softened with the silhouette tool. In the first, second or third trimester, the silhouette provides a soft “veil” on the surface of the face (Fig. 11.2, 11.9) For this purpose, the silhouette tool can be highly useful in abnormal conditions where the skin contour is involved, such as in myelomeningocele (Fig. 11.10), omphalocele, gastroschisis, cleft lip and palate (Fig. 11.11) or in thickened nuchal translucency (Fig. 11.5). Bony structures such as the spine and ribs can also be imaged as contours with the silhouette after increasing the threshold (see Fig. 11.12).

◂ Fig. 11.6: Step-by-step use of silhouette tool with Magicut in early gestation in a 12 week-old fetus. (a) After a transvaginal acquisition of a large volume box HD-live is activated. In figure (b) the silhouette level is increased to the maximum (level 100) until the fetal and surrounding contours are seen. (c) The volume is rotated and by using Magicut, the structures to be removed are better to identify and can be erased subsequently. (d) The volume is rotated as in the first step and silhouette is reduced (to level 20) and the light source adjusted.

138 

 11 The Silhouette Tool

Fig. 11.7: A fetus at 13 weeks’ gestation with conventional surface mode (left) and after HD-live render mode activation with a low silhouette level (Level 15). The fetus appears slightly transparent (ribs), but neither the intracranial ventricles nor information within or behind the fetus are displayed.

Fig. 11.8: A fetus with triploidy with narrow chest and head-abdomen discrepancy. Body contours and some intracerebral details are seen with silhouette effect.

Fig. 11.9: A fetal face in HD-live surface rendering on the left and in the right panel after adding mild silhouette effect leading to a wax-like skin.

11.3 Typical applications of the silhouette tool 

(a)

 139

(b)

Fig. 11.10: Two fetuses in early gestation with anomalies. (a) Fetus at 11 weeks’ gestation with an intrahepatic cyst (arrow), which is visible with the transparency effect of silhouette; (b) Fetus with myelomeningocele (arrow), which can be well distinguished from the neighboring structures, such as the umbilical cord.

Fig. 11.11: Two fetuses at 22 weeks’ gestation with facial clefts displayed with HD-live and silhouette. Fetus with cleft lip (left) and with cleft lip and palate (right). Adapting the position of the light source the finding can be better highlighted.

Fetal heart: The silhouette can also be applied to a STIC heart volume. This enables a good demonstration of the contours of the myocardium, valves and papillary muscles (Fig. 11.13a). Anomalies of the cardiac chambers and great vessels can be highlighted. Figure 11.13b displays a fetus with intracardiac rhabdomyoma, where the tumors are well differentiated from the adjacent structures. Silhouette can also be combined with HD-live color Doppler flow (see Glass-body mode in Chapter 12), where the grayscale information is well smoothed in the image and recently also the color Doppler information. Intracerebral ventricular system and other hypoechoic structures in the body: The silhouette is an ideal tool for displaying the hypoechoic structures and can be

140 

 11 The Silhouette Tool

used in the visualization of the intracerebral ventricular system (Fig. 11.14). Silhouette is ideal for spatial visualization of the ventricular system of the embryo, at a time where ossification of the cranial bones has not yet occurred. Figures 11.15 and 11.16 illustrate such conditions. Later in pregnancy, the interventricular system can be clearly demonstrated and identified by scanning through the fontanels and conditions as ventriculomegaly (Fig. 11.17), or holoprosencephaly can be visualized well. Other anechoic spaces in the body such as the normal or abnormal stomach, the mul-

Fig. 11.12: Spine and ribs of a fetus at 13 and at 22 weeks’ gestation with HD-live, high threshold and silhouette.

(a)

(b)

Fig. 11.13: (a) Four-chamber view of a normal heart at 22 weeks’ gestation with the use of silhouette. In comparison the right fetus (b) has heart tumors as rhabdomyoma (arrows).

11.3 Typical applications of the silhouette tool 

 141

ticystic (Fig. 11.10a) or severely dilated kidneys and other accumulations of fluid can be regions of interest to be visualized with this tool. The regions of interest are similar to those discussed in Chapters 9 and 10 covering the use of minimum and inversion mode.

Fig. 11.14: Fetus with agenesis of septum pellucidum (left) with the lateral display of the corpus callosum and on the right in a coronal view with the typical image of the fused anterior ventricles in the midline with the absence of a separating septum pellucidum.

Fig. 11.15: Embryo at 8 weeks’ gestation with HD-live and silhouette effect. The transparency enables to recognize the intracranial ventricles (see next figure).

142 

 11 The Silhouette Tool

Fet. 11.16: Embryo at 8 weeks’ gestation with high level of silhouette effect, with a view from lateral (left) and then a view from ventral (right). Note the clear display of the ventricular walls.

Fig. 11.17: Two fetuses with ventriculomegaly at 14 (left) and at 17 weeks’ gestation (right). The view is across the fontanelle and the silhouette effect is activated. Note that the dilated ventricles with large plexus choroidei are recognized with the silhouette effect.

11.4 Conclusions The recently introduced silhouette tool displays images with an almost artistic effect, yet with increasing experience, the clinical benefit becomes readily apparent. The application of silhouette in early gestation provides a rapid overview of the position and shape of the embryo and fetus. Surface regions can certainly be displayed smoothly with this tool, but its real power mainly lies in the visualization of anechoic structures within the render box. As opposed to inversion mode, the surrounding structures are visible when using silhouette. One of the promising applications is the ability to provide a visualization of the embryo’s ventricular system in early pregnancy. As the tool becomes more commonly used, further applications of this new will become apparent.

12 The Glass-Body Mode and HD-Live Flow 12.1 Principle It is well known that color Doppler sonography in the fetus is not only used to examine the heart, but also in the assessment of different organs in normal fetuses and in fetuses with malformations. The examined vessels generally have a spatial course and the 3D reconstruction can demonstrate the course and branching of the vessels. There are different methods of 3D rendering of the vessels, such as inversion or minimum mode, where only the lumen is visualized with 3D. Smaller vessels can be made visible only by the demonstration of blood flow, by using color Doppler, power Doppler or high-definition flow. In this chapter, the term color Doppler is used for all three Doppler tools. 3D visualization as static 3D, 4D or STIC in combination with color Doppler can be displayed with the tool named 3D glass-body mode. This mode can visualize blood flow either separately in 3D or together with the surrounding structures as glass-body mode (Figs. 12.1, 12.2).

Fig. 12.1: Volume acquisition of the thoraco-abdominal vessels with STIC or static 3D in combination with color Doppler. In 3D rendering mode, the user can choose between different displays, either only grayscale (upper left), only color Doppler information (upper right) or a mixture of both as glassbody mode (lower panel); hepatic vein (HV), umbilical vein (UV), inferior vena cava (IVC), aorta (AO).

144 

 12 The Glass-Body Mode and HD-Live Flow

Fig. 12.2: 3D glass-body mode with different levels of transparency: in the image optimization the user can choose the level of mix between grayscale and color Doppler information separately. In the left panel an example of a placenta and in the right panel an example of thoracoabdominal vessels. In the upper images the mix grayscale to color is 100/0 %. The images in the middle are the result of a 50/50 % mix. The best effect is achieved with a 10/90 % mix with the color Doppler information selected as surface mode.

12.2 Practical approach Prior to volume acquisition, the user should optimize the color presets to improve the visualization of the blood flow in the heart or in the vessels of interest. For a volume acquisition in static 3D, both frame rate and persistence should be kept at high level. The more images displayed per second in 2D, the more images with color information can be then acquired in a 3D volume. If the persistence is low and high pulsations are present, then many images are stored in the volume without color information. The 3D reconstruction of the vessel then reveals interruptions in its course. An exception is STIC volumes, wherein pulsations are needed. Prior to volume acquisition, it is recommended that a sweep be performed with the transducer to check whether all vessels can be easily visualized and are potentially present in the volume to be acquired. The volume is then acquired at a middle resolution, using either static 3D or STIC. An examination with live 4D in combination with

12.2 Practical approach 

 145

Fig. 12.3: 3D glass-body mode with a manipulation with Magicut. Different Magicut functions can be selected either that both grayscale and color-Doppler information are erased (middle) or only gray scale or only color Doppler.

color Doppler is also possible, but actually exhibits limitations due to the lower resolution. After the volume has been acquired, the user can select the render mode display either in B-mode alone, in color Doppler alone, or in a combination of the two as glassbody mode (Fig. 12.1). For a better result in glass-body mode, the degree of transparency should be adapted as shown in following the steps as outlined in Figure 12.2. Magicut can also be used to selectively remove grayscale structures anterior to or around the region of interest in order to highlight the color Doppler information (Figs. 12.3 to 12.8). It is important to emphasize that Magicut offers additional functions in glass-body mode, including the possibility of deleting either the grayscale or color Doppler information separately, or both, together (Fig. 12.3). The best way to proceed is to acquire an umbilical cord in the 3D glass-body mode and to try the different tools. In Figs. 12.3 to 12.10, examples of umbilical cords are illustrated in which the Magicut was used to edit and selectively delete information. Artifacts due to small signals from the vessels can also be selectively removed.

Fig. 12.4: Placenta with the umbilical cord insertion in 3D glass-body mode (left) and after Magicut manipulation (right).

146 

 12 The Glass-Body Mode and HD-Live Flow

Fig. 12.5: 3D glass-body mode of the umbilical cord insertion on the placenta in an anterior placenta (upper images), a posterior placenta (lower left) and as velamentous insertion in placenta bipartita (lower right).

Fig. 12.6: Upper left: An umbilical cord coil as seen in grayscale. Upper right: the perfusion is demonstrated with high-definition flow. Lower left: a static 3D volume is acquired and the lower right image illustrates the result after volume manipulation with Magicut.

12.2 Practical approach 

 147

Fig. 12.7: 3D glass-body mode in three different umbilical cords with different courses.

Fig. 12.8: In color Doppler, a true or a false knot in the umbilical cord is suspected (left). In 3D glassbody mode (right), the true knot is recognized due to the spatial display.

Fig. 12.9: Left: Color Doppler in the lower uterine segment indicates (arrows) free vessels along the internal cervical os, as vasa praevia. In the right figure, in 3D glass-body mode one can recognize the umbilical cord insertion as a velamentous insertion with the course of the vessels along the cervix.

148 

 12 The Glass-Body Mode and HD-Live Flow

Fig. 12.10: Fetus with a single umbilical artery and fivefold umbilical cord around the neck in HD-flow (left) and in 3D glass-body mode and HD-live-flow display (right).

12.3 Glass-body mode with HD-live flow function New software has been recently released that uses the light source, which was already discussed in Chapter 2, now in combination with glass-body mode. This software, called HD-live flow, enhances the spatial and depth effect of the vessel course. Figure 12.11 presents two fetuses with the conventional 3D glass-body mode and in comparison using HD-live flow. Many of the figures in this chapter were displayed with this new tool.

12.4 Typical applications in the glass-body mode Visualization of the umbilical and placental vessels: The visualization of the placental and umbilical vessels is generally easy to achieve (Figs. 12.1–12.10) due to absence of fetal movements. They are the ideal vessels to be examined when learning the technique. From a clinical point of view, the origin and course of the umbilical cord can be assessed to visualize typical conditions, which include velamentous insertion (Fig. 12.5), vasa previa (Fig. 12.9), umbilical cord knot (Fig. 12.8), nuchal cord (Fig. 12.10) and others. Visualization of the liver and abdominal vessels: Either in a longitudinal (Figs. 12.11–12.13) or in an axial cross-section (Fig. 12.14) of the abdomen, the intrahepatic veins, inferior vena cava and descending aorta can be well demonstrated. From a clinical point of view, this approach can be used in suspected agenesis or abnormal

12.4 Typical applications in the glass-body mode 

 149

course of the ductus venosus (Fig. 12.12) as well as in interrupted inferior vena cava with azygos continuation (Fig. 12.13) and in other rare atypical courses of vessels. In anomalies affecting the ductus venosus, the examiner should focus on the visualization of the portal system, which can be visualized well using 3D color Doppler in a cranial-caudal acquisition (Fig. 12.14). Visualization of heart and great vessels: The largest experience with glass-body mode is available from 3D fetal echocardiography (Fig. 12.15, 12.16) (see also Chapter 20). The visualization of the cardiac chambers with a septal defect or a hypo-

Fig. 12.11: Longitudinal view of the abdominal vessels with the drainage of the ductus venosus (DV) together with inferior vena cava (IVC) and hepatic vein (HV) at the subdiaphragmatic vestibulum level, left in the conventional glass-body mode and on the right the same view in another fetus displayed with HD-live flow and light source (in the bottom of the image); aorta (AO), umbilical vein (UV).

Fig. 12.12: 3D glass-body mode with HD-live flow display in a fetus (left) without ductus venosus with the connection of the umbilical vein (UV) directly into the inferior vena cava (IVC). In the fetus on the right, the connection of the ductus venosus is in an atypical ectatic vein with a course to the left side of the IVC; aorta (AO), hepatic veins (HV). Compare with the normal finding in Fig. 12.11.

150 

 12 The Glass-Body Mode and HD-Live Flow

Fig. 12.13: 2D color Doppler (left), 3D glass-body mode (middle) and HD-live flow display (right) in a fetus with interruption of the inferior vena cava and azygos vein continuation. Two vessels aorta (AO) and azygos vein have a side-by-side course with different directions of blood flow.

Fig. 12.14: Abdomen axial cross-section view in color Doppler from a cranial view on the intrahepatic vessels. In (a) demonstrated in the conventional 3D glass-body mode and in (b) with the new HD-live flow display. In (a) the spatial course of many vessels is visualized but the image in (b) reveals a better effect of depth with a good discrimination of the vessels of interest. Scrolling plane by plane (b), (c) and (d) reveals the hepatic veins (HV) and the different parts of the portal system; Ductus (DV) venosus, umbilical vein (UV), portal vein (PV), inferior vena cava (IVC), aorta (AO).

12.4 Typical applications in the glass-body mode 

 151

plastic ventricle is rarely the anomaly of interest in glass-body mode, but rather anomalies involving the great vessels (Fig. 12.15, 12.16). Difference in size, in blood flow direction, spatial arrangement or course of the vessels are some of the information that can be demonstrated by using the 3D glass-body mode. Typical anomalies providing a good 3D images include transposition of the great arteries (Fig. 12.16b), right or double aortic arch, hypoplastic left heart syndrome, aortic coarctation and can be well differentiated from a normal finding. The best view is generally achieved with a cranial to caudal view from the perspective of the mediastinum or from the upper left side. Visualization of the intracerebral vessels: Intracerebral arteries and veins can be visualized using glass-body mode, ideally from a sagittal section where in addition to the pericallosal artery the internal cerebral veins with the straight sinus and the superior sagittal sinus can be demonstrated (Fig. 12.17). Clinical conditions such as the abnormal course of the anterior cerebral artery in complete or partial agenesis of the corpus callosum, abnormal vasculature in vein of Galen aneurysmal malformation or other disorders can be visualized well with this technique (see Chapter 16). The intra-

Fig. 12.15: STIC with color Doppler and glass-body mode rendering of a heart. In the background the ventricles are identified while the crossing of the great vessels appears in the foreground. Compare the difference with figure 12.16 with the HD-live flow display; aorta (AO), left ventricle (LV), pulmonary artery (PA), right ventricle (RV).

152 

(a)

 12 The Glass-Body Mode and HD-Live Flow

(b)

Fig. 12.16: STIC volume of a heart in glass-body mode and HD-live flow display in a normal fetus (a) and (b) in a fetus with a d-transposition of the great arteries (curved arrows), aorta (AO), left ventricle (LV), pulmonary artery (PA), right ventricle (RV).

cranial venous anatomy in 3D is a new field of research, either examining the relationship between venous development and cortical maturation or focusing on the course of the veins in different brain anomalies, but in these cases best images are demonstrated by using the transvaginal approach. Figure 12.18 demonstrates the 3D rendering of the circulus of Willis in 3D glass-body mode.

Fig. 12.17: Intracranial arteries and veins in 3D glass-body mode (left) and HD-live flow display (right). The view is a sagittal view on the anterior cerebral artery, the pericallosal artery, the internal cerebral vein (ICV) and superior sagittal sinus (SSS).

12.5 HD-live flow using the color silhouette tool 

 153

Fig. 12.18: Circulus of Willis in 3D glass-body mode.

12.5 HD-live flow using the color silhouette tool In the most recent software release 2016/2017, a new application of silhouette was introduced and can be applied in the color Doppler 3D when displayed in HD-live flow. This can be used for color Doppler, High-definition flow and power Doppler (Fig. 12.19– 12.23). With the actual HD-live flow a glossy surface of the flow is displayed and gives

(a)

(b)

(c)

Fig. 12.19: Umbilical cord insertion on the placenta displayed with the three different 3D glass-body mode tools. In (a) the usual glass-body mode, in (b) in combination with HD-live flow and in (c) displayed in the newest combination of silhouette tool with glass-body-mode. Note in (c) that the vessels become almost transparent and the borders of the vessels are clearly seen as a silhouette.

154 

 12 The Glass-Body Mode and HD-Live Flow

(a)

(b)

(c)

(d)

Fig. 12.20: Two examples of an umbilical cord displayed in HD-live flow, left panels (a), (c) and after the use of the new silhouette tool for color Doppler, right panels (b), (d). Note the transparency of the vessels with this new mode

(a)

(b)

Fig. 12.21: A sagittal view on heart and abdominal vessels (as in Fig. 12.11) in a fetus displayed with 3D-HD-Flow (a) and power Doppler glass-body mode (b) both with in combination with silhouette effect; aorta (Ao), ductus venosus (DV), inferior vena cava (IVC), umbilical artery (UA), umbilical vein (UV).

a spatial effect of the visualized vessels (Fig. 12.20a,c). In the new software of color Doppler silhouette, blood flow becomes more transparent and the border of the vessels and blood flow are displayed (Fig. 12.20b,d). This color silhouette of flow enables to see the shape of flow even behind the vessels, as illustrated in the few examples in Figs. 12.19–19.23. The clinical use needs to be studied further.

12.5 HD-live flow using the color silhouette tool 

(a)

 155

(b)

Fig. 12.22: STIC volumes of two normal hearts at 13 (a) and 22 (b) weeks’ gestation in glass-body mode and HD-live flow display in combination with the new silhouette tool for color Doppler. Note the transparency of the vessels in this new display.

(a)

(b)

Fig. 12.23: (a) Color Doppler and HD Flow with silhouette display in a fetus with interrupted inferior vena cava showing aorta (Ao) and azygos vein (Azyg.) side by side (compare with Fig. 12.13). (b) The same display in a fetus with an aberrant right subclavian artery (ARSA). PA, pulmonary artery.

12.6 Conclusions Glass-body mode is generally used to visualize blood flow in the heart and in the vessels by combining color Doppler and 3D. The vessels can be viewed alone or together with the neighboring structures displayed in grayscale. Not only the heart, but also other regions with a well-developed vasculature as liver, brain, lung or placenta are good areas for application of glass-body mode. The fetal cardiac examination with a view from the mediastinum provides a spatial demonstration of the heart with the crossing of the great vessels. The combination with HD-live flow enables a significantly better spatial visualization of the blood vessels and has currently become an important adjunct to glass-body mode display.

13 The B-Flow Mode 13.1 Principle B-flow technology is a special technique that enables visualization of blood flow from the grayscale image without the use of Doppler signals. This technique makes the direct visualization of blood cell reflectors in grayscale mode possible and the information is displayed together with the 2D information but mainly flow events are visualized, while any other specific information from neighboring structures is not clearly visible. On the screen, the blood flow then appears in a contrast-rich color against the surrounding information (Fig. 13.1). Both the grayscale as well as the B-flow image have the same principle of visualization, so the image frame rate remains unchanged. Blood flow as event can be visualized, but unlike Doppler no information on flow direction, speed, or turbulences is available with the B-flow. Since the information is angle-independent, it can be used for vessels with a horizontal course (Fig. 13.1). B-flow is very sensitive and tiny vessels can be visualized side-by-side to the large ones. Therefore the B-flow image compared with color Doppler not only has a significantly higher frame rate, but also displays a better spatial resolution. B-flow can be used in both static 3D (Fig. 13.2) as well as STIC volume acquisition (Figs. 13.3, 13.4).

Fig. 13.1: Longitudinal view of thorax and abdomen in B-flow demonstrating the heart (H), aorta (AO) and abdominal vessels with umbilical vein (UV) and ductus venosus (DV). Neighboring structures cannot be visualized in B-flow mode.

13.1 Principle 

 157

Fig. 13.2: Static 3D rendering with B-flow of heart, aorta (AO) and abdominal vessels as inferior vena cava (IVC), ductus venosus (DV) and umbilical vein (UV).

Fig. 13.3: STIC volume with B-Flow after a STIC acquisition in gray mode (left) and in gradient light display (right).

Fig. 13.4: STIC Volume with B-flow displayed with HD-live; aorta (AO), ductus venosus (DV), inferior vena cava (IVC), umbilical vein (UV), umbilical artery (UA).

158 

 13 The B-Flow Mode

13.2 Practical approach Prior to volume acquisition, the examiner should adjust the preset of the grayscale image and the B-flow. In our experience, the key features to be optimized are the sensitivity and persistence of B-flow. A high sensitivity and moderate persistence is suitable for visualizing pulsations on the heart. By contrast, a high persistence and lower sensitivity is needed for small vessels and veins to prevent image overlapping. Prior to volume acquisition, a sweep can be performed to check whether the region of interest includes all vessels of interest. The volume is then acquired using either static 3D or STIC mode. After volume acquisition, the examiner can check whether all vessels of interest are included in the volume box and do not exhibit motion artifacts. Often images with B-flow are not informative enough and the user is encouraged to switch to the rendering mode and visualize the result in the 3D render mode. In our experience, increasing the gain and choosing the 3D surface mode for the display in gradient light provides a good image of the vessels of interest. Artifacts due to signals from tiny vessels or movements can be selectively removed with the Magicut electronic scalpel.

13.3 Typical applications of the B-Flow mode Only few examiners have focused on the use of B-flow in combination with 3D or STIC. These have mainly focused on the heart and the great vessels as well as tiny vessels such as the pulmonary veins or the fetal vessels in early gestation. Visualization of intrahepatic and abdominal vessels: The longitudinal acquisition of a volume provides a good approach to demonstrate the aorta, inferior vena cava umbilical vein with Ductus venosus and the intrahepatic vessels (Figs. 13.2–13.6). In our opinion, this is the best plane to acquire experience with this technique. Visualization of heart and great vessels: The heart and the great vessels are best acquired using STIC. Lateral, ventral or cranial acquisition provides the good results, for example, in demonstrating course and crossing of the great vessels. Figures 13.7 and 13.8 present examples of the application on the heart. Other areas: Other areas with good perfusion are also suitable, including the placenta, umbilical cord, intracranial vessels, as illustrated in Figs. 13.9 and 13.10.

13.3 Typical applications of the B-Flow mode 

 159

Fig. 13.5: Abnormal dilated umbilical vein (UV) in grayscale (left) and in static 3D-B-Flow rendering (right).

Fig. 13.6: Static 3D rendering after B-Flow demonstration of the aortic arch (AO) and inferior vena cava (IVC).

Fig. 13.7: Right aortic arch (RAO) with left ductus arteriosus (DA) and U-Sign visualized with STIC B-Flow and 3D rendering with gradient light (left) and HD-live (right); descending aorta (AOD), pulmonary artery (PA).

160 

 13 The B-Flow Mode

Fig. 13.8: Double aortic arch in B-Flow and 3D rendering. One recognizes the right (RAO) and left aortic arch (LAO), the ductus arteriosus (DA) and the pulmonary artery (PA) with the left branch (LPA). All three vessels merge into the descending aorta (AO), pulmonary artery (PA). The asterisk indicates the position of the trachea, which cannot be seen in B-Flow mode.

Fig. 13.9: STIC volume display with B-flow, revealing a true knot of the umbilical cord (left) and in surface mode B-flow rendering display

13.3 Typical applications of the B-Flow mode 

 161

Fig. 13.10: Static 3D B-flow demonstration of the intracranial vessels in a fetus with an aneurysm of the vein of Galen (arrows).

13.4 Conclusions The initial optimism surrounding the application of B-flow in combination with 3D or STIC has leveled off somewhat in recent years. The application is ideal for use in areas with small vessels and blood flows, where the examiner must consider whether the spatial demonstration of the vessels without the surroundings provides enough information. The authors prefer to use the bidirectional color Doppler in 3D and STIC for the examination of tiny and large vessels, as already explained in Chapter 12.

14 Biplane Display Using the Electronic Matrix Transducer 14.1 Principle One of the special features of an electronic matrix probe is the construction of the transducer with multiple rows of crystals instead of a single one, found in conventional mechanical transducers. With these multiple rows of crystals (64 rows in some), the transducer footprint has more than 8,000 elements, hence the name “matrix array transducers”. In conventional mechanical 3D probes, a single row of crystals is used to generate the 2D image and once the 3D acquisition is selected, a mechanical motor sweeps the ultrasound beam in order to generate multiple 2D planes compiled to form the 3D volume. With the use of rapid processors in computers, matrix transducers are able to electronically steer the ultrasound beam through a selected volume box and to acquire volumes 2 to 4 times faster than a 3D mechanical transducer. This rapid acquisition of ultrasound planes makes an enhanced resolution within the 3D volume possible, as well as the simultaneous display of two planes in real-time in a promising display, which is referred to as “biplane display”. Additionally, the transducer can display a thin slice of the region of interest called VCI-A in 4D (see Chapter 4) much more rapidly than is possible with a mechanical probe. In this chapter, we present our preliminary experiences with biplane display, providing typical examples.

14.2 Practical approach The examination is first performed in 2D with the matrix probe and the image and the region of interest adjusted and optimized (see Chapter 1). The aperture angle of the image should be kept as narrow as possible, and the biplane display is then activated. The image is instantly split into two images A and B as dual image (Fig. 14.1). The left image is plane A and is the scanning plane, where a vertical line appears. This line can be freely controlled, moved and placed along the region of interest. The right image is plane B and is the orthogonal plane along the line placed in plane A. (Fig. 14.1). While the examination is performed as usual as can be seen in the images in plane A on the left, the panel on the right side simultaneously reveals the orthogonal corresponding images along the biplane-line placed in A. The biplane examination can be performed in grayscale or in combination with color-Doppler. The use of zoom enables the magnification of one region to better focus on it. From a practical point of view, biplane can be used in two ways, as illustrated in Figs. 14.1 to 14.3: One approach is to keep the position of the line unchanged and move the transducer so that the struc-

14.2 Practical approach 

 163

Fig. 14.1: Examination of the spine with biplane mode with the dual image. The examination plane is on the left and the image on the right is a perpendicular cross-section plane along the line placed in the left image, here at the level of a thoracic vertebral body. Ribs can be seen in the right image. See also Figs. 14.2 and 14.3.

Fig. 14.2: Biplane examination of the spine. The previous figure was first visualized but in order to visualize the lumbosacral region, the position of the biplane line was kept unchanged and the transducer moved toward the sacral region. Another possibility is provided in next figure.

164 

 14 Biplane Display Using the Electronic Matrix Transducer

Fig. 14.3: Biplane examination of the spine as in the two previous figures. Another possibility of visualizing the lumbosacral region is to control and move the biplane line and place it along the sacral region as showed in this example.

tures of interest are arranged successively along the biplane-line (Fig. 14.2) and the orthogonal images are generated on the right panel. Another approach is to keep the scanning image still and to move the biplane-line to produce successive corresponding orthogonal images in the right panel (Fig. 14.3). In the latter approach, the ultrasound system must permanently re-assess the line position and calculate all images, which is accompanied by a slight delay in scanning. The authors recommend that the user simply tries to work with this interesting type of display; in this chapter, we share some applications acquired through our preliminary experiences.

14.3 Typical applications of biplane mode Once the examiner uses the biplane-display in many examinations, he realizes that this new modality of scanning is not limited to screening examinations, but can also be used in suspected fetal anomalies. Examination of head and face: The head and face are routinely examined in several planes in 2D and 3D multiplanar mode. The biplane mode therefore offers an ideal tool for the demonstration of many anatomical structures. While the head, for example, is examined in a transverse plane, in the biplane image the cavum septi pellucidi (Fig. 14.4), the lateral ventricles, the Sylvian fissure or the posterior fossa can be visu-

14.3 Typical applications of biplane mode 

 165

Fig. 14.4: Examination of the brain using the biplane mode at the level of the cavum septi pellucidi (*) in both orthogonal planes. The original plane on the left is the standard axial plane of the head. In the image to the right, both anterior horns (short arrows) and the corpus callosum (long arrow) are seen, which are not seen in the left plane.

Fig. 14.5: Agenesis of corpus callosum visualized in biplane mode. The head is examined in the standard axial plane as can be seen in figure 14.4, but in this case the cavum septi pellucidi is absent (?) in both planes. In the biplane image on the right, the anterior horns are seen shifted laterally.

166 

 14 Biplane Display Using the Electronic Matrix Transducer

Fig. 14.6: Examination of the head in a coronal view and in biplane mode a midsagittal view of the corpus callosum is visualized (arrows).

Fig. 14.7: A fetus with an occipital encephalocele in biplane mode. Brain tissue can be recognized in the cele.

14.3 Typical applications of biplane mode 

 167

Fig. 14.8: Axial view of the head in a fetus with choroid plexus cysts. In the left image, only one cyst can be seen, but the biplane image reveals both.

alized simultaneously. Anomalies such as agenesis of the corpus callosum can be suspected with this mode as can be seen in Fig. 14.5. If the fetus is in breech presentation, the brain can be examined through the fontanelle in a coronal view and the corpus callosum can be visualized simultaneously in biplane mode (Fig. 14.6). Other brain anomalies can be visualized in one plane and verified with the biplane mode in the other (Fig. 14.7–14.8). Additionally, biplane mode is particularly helpful in the assessment of the fetal face (Fig. 14.9, 14.10), where the examination can be started either from the profile in a sagittal insonation or from an axial or a coronal view (Fig. 14.9–14.14). The simplest approach is probably to obtain a profile and simultaneously tilt the biplane-line from the plane of the eyes (Fig. 14.9), down to the nose and then to the upper and lower jaw (Fig. 14.10). Facial anomalies such as cleft lip, cleft palate and other malformations can be clearly demonstrated and identified with the biplane mode and the abnormal finding is better assessed when displayed in the two planes at the same time (Fig. 14.11– 14.14). A similar approach can also be performed already in the first trimester screening as illustrated in Fig. 14.13. Examination of the heart: The biplane mode is an interesting new tool for the examination of the heart, chest and mediastinum. At the level of the heart, the four-cham-

168 

 14 Biplane Display Using the Electronic Matrix Transducer

Fig. 14.9: Biplane mode in the examination of the fetal face. The profile of the fetus is visualized and the biplane is placed at the level of the eyes. On the left the eyes are not seen but in the orthogonal biplane image both eyes and orbits are displayed (see also Fig. 14.10).

Fig. 14.10: Biplane mode of the face. The profile is seen as in the previous figure but the biplane line is placed now at the level of the mouth with the visualization of the intact maxilla (compare with next figure).

14.3 Typical applications of biplane mode 

 169

Fig. 14.11: Biplane mode of the face in a bilateral cleft lip and palate (arrows). The biplane examination is performed by visualizing the profile and placing the line along the maxilla.

Fig. 14.12: Biplane mode in a bilateral cleft lip and palate (arrows). The biplane mode is the result of a coronal view of the face.

170 

 14 Biplane Display Using the Electronic Matrix Transducer

Fig. 14.13: Biplane mode in a bilateral cleft lip and palate (arrows) in a fetus at 13 weeks’ gestation. In the left image the “maxillary gap” is seen, where the line is placed and the suspicion is confirmed in the resulting biplane image.

Fig. 14.14: Biplane mode in a fetus with lymphangioma of the neck. The extent of the finding can be better appreciated by adding the orthogonal plane in the assessment in 2D. Also compare with Fig. 18.21.

14.3 Typical applications of biplane mode 

 171

ber view can be examined while simultaneously, a sagittal section of the aortic arch, ductal arch (Fig. 14.15) and venous system can also be visualized. Anomalies affecting the great arteries or the venous system in the mediastinum can be demonstrated in these planes simultaneously (Fig. 14.16). An interesting view is the visualization of the interventricular septum in two planes, mainly the direct view of the septal surface (Figs. 14.17–14.19). This novel view makes it possible to check the integrity of the interventricular septum in grayscale or in combination with color Doppler. Figures 14.15– 14.20 illustrate examples of fetal hearts under normal and abnormal conditions. Examination of chest, abdomen, skeletal system and other areas: The biplane display also has significant potential when it comes to the examination of different fetal organs. The visualization of the spine has already been demonstrated in Fig. 14.1 and this approach can aid in the accurate assessment of the height of the lesion in spina bifida (Fig. 14.21) or in hemivertebra. The lungs and abdominal organs can also be examined quite well using biplane mode and this display facilitates obtaining a better overview of normal and abnormal conditions. Figures 14.22 to 14.26 illustrate examples of some anomalies demonstrated in biplane mode.

Fig 14.15: Biplane mode in a normal heart. The examination is performed in the five-chamber view plane (left). The biplane view simultaneously displays the aortic arch.

172 

 14 Biplane Display Using the Electronic Matrix Transducer

Fig. 14.16: Biplane mode of a heart with a left persistent superior vena cava (arrows). The examination is performed in the four-chamber-view (left) and the biplane mode reveals the left superior caval vein with its course from the neck toward the heart in an orthogonal plane (arrows).

Fig. 14.17: Biplane mode display of the interventricular septum of the heart with a muscular ventricular septal defect (arrow) in grayscale. The defect is suspected in the left and confirmed in the right image.

14.3 Typical applications of biplane mode 

 173

Fig. 14.18: Biplane mode display of the interventricular septum in color Doppler in a heart with a muscular ventricular septal defect (arrow). The defect is suspected in the left and confirmed in the right image.

Fig. 14.19: Biplane mode of the interventricular septum in a fetus with heart tumors, diagnosed as rhabdomyomas. A large rhabdomyoma (*) is found in the region of the septum and left ventricle. In biplane mode, it is possible to observe that the aortic valve is not obstructed (arrow) by the tumor.

174 

 14 Biplane Display Using the Electronic Matrix Transducer

Fig. 14.20: Biplane mode in color Doppler in a fetus with transposition of the great arteries. The parallel course of the great vessels (arrows) is recognized in the orthogonal biplane.

Fig. 14.21: Myelomeningocele in biplane mode in a fetus at 21 weeks’ gestation. Moving and controlling the biplane line facilitates a good assessment of the level of the lesion.

14.3 Typical applications of biplane mode 

 175

Fig. 14.22: Both kidneys can be ideally assessed in two orthogonal planes, as shown in this biplane mode example.

Fig. 14.23: A fetus with multicystic renal dysplasia in biplane mode.

176 

 14 Biplane Display Using the Electronic Matrix Transducer

Fig. 14.24: Biplane mode in color Doppler in a fetus with an omphalocele with the demonstration of aorta (AO) and umbilical vein (UV).

Fig. 14.25: Biplane visualization of a hyperechogenic lung in lung sequestration (*). In the right part of the figure, which is orthogonal to the left panel, the normal upper lung lobe (arrow) is recognizable and has a normal size and echogenicity.

14.3 Typical applications of biplane mode 

 177

Fig. 14.26: Biplane visualization of a fetus with ascites (*).The extent of the finding can be better appreciated in this mode.

14.4 Conclusions Given the fact that it is a newly developed modality, the examination with biplane mode requires a learning curve in order to familiarize oneself with it and rapidly integrate it into routine screening. Our preliminary experience has shown that the main benefit of biplanar display mode is ability to obtain information quickly and simultaneously on two planes, which proves to be superior to that of a single 2D image.

15 Calculation of 3D Volumes 15.1 Principle Fetal biometry is an intrinsic part of the prenatal ultrasound examination. During a routine examination, measurements of diameters, circumferences or areas are conducted and compared with reference ranges. During a fetal examination, volume calculations are rarely performed and when needed, the examiner simply makes his or her assessment based on an ideal form of the region of interest when calculating distances and areas. The acquisition of a 3D volume is a good prerequisite for a reliable measurement of a volume. Generally speaking, few techniques are available for volume calculation and depending on the region of interest, the calculation can be accomplished quickly and easily, or in a complicated and time-consuming way. In this chapter, we discuss the two main tools that are typically used for volume calculation.

15.2 Practical approach 3D volume measurements can be performed in different ways. The most well-known and common technique is with VOCAL-software (see below). Additional tools have been introduced in recent years that facilitate the automatic rapid measurement of echolucent regions. There is general agreement that the need to conduct measurements in prenatal diagnosis will increase in the future and that there will thus be a call for more simplified volume measurement tools. As it is, the few techniques available for conducting volume measurements are quite time-consuming in their implementation, and this certainly goes a long way to explaining why most volume calculations are reported in research studies yet not used in clinical practice. In the following, we discuss two techniques, namely, VOCAL and Sono-AVC software.

15.2.1 Virtual Organ Computer-aided Analysis (VOCAL)-Software VOCAL software is still the most commonly used technique for a calculation of a volume. Following a static 3D volume acquisition, the structure to be measured is displayed in the orthogonal mode and magnified in order to be placed in the center of the image. Once the VOCAL software is activated, a vertical line appears with two triangles present on the two poles of this line. The user manually moves each triangle, placing each on the poles of the area to be measured (Fig. 15.1). In the next step, the outline drawing is selected either manually, semiautomatically, or automatically. The automatic drawing of the contours is reliable when a single echolucent structure with well-defined borders is selected such as the stomach, bladder, or a cyst, which

15.2 Practical approach 

 179

Fig. 15.1: Step-by-step 3D volume calculation using VOCAL: Once the region of interest is displayed in orthogonal mode, the VOCAL function is then selected. A vertical line appears with two triangles. These are placed manually at the two poles of the selected area, in this case the lung.

Fig. 15.2: The next step in VOCAL volume calculation (see Fig. 15.1): After the region of interest has been magnified and the triangles placed on the poles, the type of outline drawing is selected, either as manual or semiautomatic. Once the outline is drawn well, the measurement is then confirmed and an automatic rotation of the volume to the next image occurs.

180 

 15 Calculation of 3D Volumes

Fig. 15.3: The next step in VOCAL volume calculation (see Figs. 15.1, 15.2): The user now proceeds in a similar fashion to Fig. 15.2, moving from image to image and drawing the outline and confirming the result until all steps have been completed. The number of rotations can be selected by the operator prior to volume calculation.

Fig. 15.4: The next step in VOCAL volume calculation (see Figs. 15.1–15.3): At the conclusion of the previous steps of drawing the lines and rotating the volume the result is displayed on the screen highlighting the measured region of interest to the lower right, in this instance the lung, after calculation. At this stage, some corrections can be made by reviewing one or other of the planes to adjust contour drawings.

15.2 Practical approach 

 181

Fig. 15.5: The 3D VOCAL result for the previously assessed lung. The result can be displayed in different colors as a solid area (left) or as a mesh (right).

however is rarely the case. In most cases, however, for the evaluation of kidneys, lungs, placenta and other structures, automatic recognition of contours is difficult and selecting the manual or semiautomatic function is recommended. This then allows the user to draw or modify the outline according to the ultrasound information on the screen (Fig. 15.2). Once the user has finished drawing the outline, this step is manually confirmed and the image automatically changes to the next image by a few degrees rotation along the long axis. The same steps are confirmed and adjusted and manually corrected in each plane (Figs. 15.3, 15.4) until a complete 180° rotation is achieved. The more rotation steps that are selected, the more precise the volume calculation will be. Figures 15.1 to 15.5 demonstrate the step-by-step approach for lung volume measurement using VOCAL. The visualization of the calculated volume is displayed at the end on the screen either with a solid or a mesh envelope (Fig. 15.5).

15.2.2 Sono Automatic Volume Calculation (Sono AVC™) This recently released software volume calculation tool is used in gynecological ultrasound for the automated measurement of cysts and follicles. The software automatically recognizes single or multiple echolucent areas as cysts and calculates the corresponding volumes accordingly (Figs. 15.6–15.10). The user selects the area with the organs of interest by placing them within the render box. The structures to be measured can be added or removed selectively by means of a simple mouse click. It should be kept in mind that the software automatically recognizes echolucent areas and therefore shadows can lead to artifacts. On the other hand, this technique is the quickest technique for volume calculation of cysts, especially when multiple cysts need to be measured (Fig. 15.10). Therefore, the measurement of a filled stomach

182 

 15 Calculation of 3D Volumes

Fig. 15.6: 3D volume calculation with Sono Automatic Volume Calculation (Sono-AVC). After selecting the region of interest where the liquid is to be measured (here the stomach), the region can be selectively clicked with the mouse while activating Sono-AVC (see Fig. 15.7).

Fig. 15.7: 3D volume calculation with Sono-AVC: Following the mouse click, the liquid is identified and the volume displayed. The 3D shape of the stomach is displayed and the volume calculated.

15.2 Practical approach 

 183

Fig. 15.8: 3D volume calculation with Sono-AVC, here in a fetus with double bubble sign in duodenal atresia.

Fig. 15.9: 3D volume calculation with Sono-AVC in a fetus with hydronephrosis in pelvic-ureteral junction obstruction.

184 

 15 Calculation of 3D Volumes

Fig. 15.10: 3D volume calculation with Sono-AVC in a fetus with multicystic renal dysplasia. The volumes for different individual cysts can be separately calculated and displayed. Measurements are illustrated for the different cysts in different colors, and the numbers refer to the region measured.

(Figs. 15.7, 15.8), or the fluid volume in the dilated renal pelvis (Fig. 15.9) or the cyst volumes in multicystic kidneys (Fig. 15.10) can easily be calculated. Also refer to the example in Chapter 19.

15.3 Clinical application of volume calculation Volume calculations and corresponding reference ranges were reported for early gestation including volume of placenta, gestational sac and embryo. The lung is another common organ for measurement in normal fetuses and fetuses at high-risk for pulmonary hypoplasia. Volume measurements were reported for different structures such as the liver, brain, placenta, kidneys, lateral ventricles, cardiac cavities and others. One of the main applications of volume measurements is the fetal weight estimation by calculating the volume of an extremity or in combination with other volume measurements. However, the routine use of volume calculation is still uncommon and is mainly performed in research studies.

15.4 Conclusions 3D volume measurements are important in selected cases in prenatal ultrasound, but performing such calculations is still quite time-consuming. VOCAL and Sono-AVC are the most commonly used tools and require a degree of experience before being used effectively and easily, which limits their use in routine ultrasounds.

Part III: Clinical Applications of Prenatal Diagnosis

16 3D Fetal Neurosonography 16.1 Introduction The ultrasound examination of the fetal brain mainly focuses on the interval between 15 and 40 weeks of gestation. The first section of this chapter deals with this interval and demonstrates the potential of 3D ultrasound under normal and abnormal conditions. Nevertheless, recently, with the increased use of high-resolution transvaginal sonography in combination with 3D techniques, effective sonoembryology of the brain between 7 and 14 weeks of gestation has become possible. We discuss some aspects of this development in the second part of the chapter.

16.2 Fetal neurosonography with 3D ultrasound Ultrasound screening of the fetal brain is performed from 15 weeks of gestation onwards, mainly by demonstrating axial planes that are typically used to measure the biparietal and transcerebellar diameters. In fetuses at high-risk for CNS anomalies or when an abnormality is suspected, during routine screening additional coronal and sagittal sectional planes are recommended as part of a comprehensive fetal neurosonogram. Additional planes are often difficult to obtain, especially in unfavorable fetal positions, such as in a vertex presentation. Usually at this point, additional time-consuming transvaginal sonography has to be performed. With 3D neurosonography, the examiner can acquire a volume and as explained in Chapters 2–5, subsequently reconstruct any sectional plane needed. Alternatively, he can examine directly with 4D and generate these planes online during the live examination. One of the main advantages of 3D multiplanar mode is the possibility of virtually reconstructing the midline structures from a volume data set that has been acquired from an axial or oblique insonation of the fetal head. Moreover, tomographic mode enables the visualization of the region of interest together with the adjacent structures in one single image. The 3D volume acquisition can be achieved from different insonation angles as an axial, coronal or sagittal approach. Axial 3D-acquisition: The easiest acquisition is achieved from an axial view of the head, often when the fetus is in a vertex presentation. In tomographic display mode (Fig. 16.1), the parallel cross-sectional planes provide a good overview of the brain anatomy. As illustrated in Figure 16.1, the landmarks such as the cerebellum with cisterna magna, cortex tissue, posterior and anterior horns of the lateral ventricle and falx cerebri with cavum septi pellucidi can be visualized in a single image. Figures 16.1 to 16.4 illustrate 3D volume images with multiplanar mode of normal and abnormal fetal brains.

188 

 16 3D Fetal Neurosonography

Fig. 16.1: A 3D volume data set of an axial acquisition of a fetal brain displayed in tomographic mode. The different planes displayed provide an overview of the main structures of a normal brain as the falx cerebri (Falx), the lateral ventricles (Lat.V), the choroid plexus (Plexus), the thalami (Th), the cavum septi pellucidi (Csp), the Sylvian fissure (arrow), the cortex and the cerebellum with cisterna magna (circle).

Sagittal or coronal 3D-acquisitions: A better resolution of the midline structures is provided by a 3D acquisition through the anterior fontanelle, transabdominally when the fetus is in breech presentation (Figs. 16.5, 16.6) or transvaginally in vertex presentation (Fig. 16.7). A volume acquired through the fontanelle can be used to display a series of coronal or parasagittal sectional planes. The best resolution is provided by transvaginal volume acquisition (Figs. 16.7–16.9). Figures 16.5 to 16.9 illustrate examples of transabdominal & transvaginal 3D volumes acquired through the fontanelle in normal and abnormal fetuses. The intracranial structures are best assessed with a multiplanar reconstruction display such as the orthogonal, tomographic or Omniview mode. Quite often, the additional combination with volume contrast imaging (VCI) (see Chapter 4) improves the resolution of the reconstructed image, as illustrated in Figures 16.6–16.10.

16.2 Fetal neurosonography with 3D ultrasound 

 189

Fig. 16.2: A fetus with holoprosencephaly displayed in tomographic mode.

Fig. 16.3: A fetus with agenesis of the corpus callosum demonstrated in tomographic mode. The cavum septi pellucidi is absent (?) and in the midline there is a dilated interhemispheric fissure (arrow). The shape of the lateral ventricles (Lat.V) demonstrates the typical colpocephaly; falx cerebri (Falx).

190 

 16 3D Fetal Neurosonography

Fig. 16.4: Schizencephaly (circle) in tomography mode. In the planes, cranial and caudal to the lesion the cortex appears intact.

Fig. 16.5: Coronal sectional planes in tomographic mode after a transabdominal volume acquisition through the fontanelle. Following structures can be recognized: The interhemispheric fissure (IHF), the corpus callosum (CC), the cavum septi pellucidi (Csp), the thalami (Th), the insula (Ins) and the anterior horns with the lateral ventricles (Lat.Vent.).

16.2 Fetal neurosonography with 3D ultrasound 

 191

Fig. 16.6: Sagittal sectional planes after a transabdominal 3D acquisition through the fontanelle with a tomographic mode rendering The midline structures as the corpus callosum (CC) and the vermis with the posterior fossa are well seen.

Fig. 16.7: Sagittal and parasagittal sectional planes after a transvaginal 3D acquisition through the fontanelle with a rendering in tomographic mode. The focus is on the midline structures, which are well recognized as the corpus callosum (CC), vermis and lateral ventricles (Lat.Vent.). With the choice of a larger interslice distance the insula (Ins) could have been displayed as well.

192 

 16 3D Fetal Neurosonography

Fig. 16.8: Coronal sections after a transabdominal 3D volume acquisition through the fontanelle in a fetus with agenesis of the corpus callosum. In this view no corpus callosum can be seen but the typical “steer horn” shape (circle). The frontal anterior horns (*) are in this anomaly compressed and lateralized.

16.3 3D visualization of specific brain structures In fetal neurosonography, some structures need to be visualized and have to be reconstructed with 3D multiplanar rendering. The steps for the corpus callosum and vermis are explained below: The corpus callosum: For the experienced examiner, the visualization of the corpus callosum is considered as a part of a comprehensive ultrasound examination. This structure is either demonstrated directly or with a rapid reconstruction of a sagittal plane after of a 3D volume acquisition from an axial plane. An important landmark is the cavum septi pellucidi as orientation point, both during volume acquisition and 3D rendering (Figs. 16.10, 16.11). Figures 16.10 and 16.11 explain the 3D reconstruction of the corpus callosum step-by-step. In addition to this visualization using 3D static mode, the corpus callosum can also be directly reconstructed in 4D during a live examination, for example by using

16.3 3D visualization of specific brain structures 

 193

Fig. 16.9: Coronal sections after a transvaginal 3D volume acquisition through the fontanelle in a fetus with agenesis of the corpus callosum (circle) similar to the case in Fig. 16.8, but this fetus additionally has a schizencephaly (arrows), which thanks to the display in tomographic mode can be recognized in the adjacent planes.

Fig. 16.10: Despite the vertex position in this fetus the corpus callosum, which cannot be seen, can be reconstructed from an axial volume acquisition. The orientation is best achieved by locating the cavum septi pellucidi (CSP) and placing the intersection dot in the Csp. The axes of the head (dashed arrows) are still oblique but should be aligned with the horizontal line (see next figure).

194 

 16 3D Fetal Neurosonography

Fig. 16.11: After the intersection dot was placed on the cavum septi pellucidi, the planes A and B are rotated in a way that the axis of the falx cerebri is aligned along the horizontal axis (dashed line). Now the corpus callosum (CC) appears in the C-plane.

VCI-Omniview. In the presence of a complete agenesis of the corpus callosum, the axial view of the brain reveals the typical teardrop shape of the lateral ventricles, called colpocephaly, as well as the absence of the cavum septi pellucidi (Fig. 16.3). In the coronal view, the absence of the cavum septi pellucidi is confirmed and the anterior horns are displaced laterally providing a typical image described as the “steer horn” shape (Figs. 16.8–16.9). The cerebellar vermis: Cerebellar anatomy is generally assessed in the axial view. This includes the demonstration of the normal shape of both hemispheres with the vermis present in between, and the cisterna cerebello medullaris additionally has a normal size, while the inferior part of the vermis is visualized and separates the 4th ventricle from the cisterna. Ideally, not only the cerebellum and cisterna magna should be included in a 3D volume but, when possible, the brain stem should also be as well (Fig. 16.12). After volume acquisition, the images are rotated in such a way that the middle axis and vermis are aligned. In the C-plane, the vermis shape and size are then recognized especially in its relationship to the cisterna magna and brain stem (Fig. 16.12). Similarly to the corpus callosum, the cerebellar vermis can also be directly visualized during a 4D examination using VCI-Omniview, as previously explained (Fig. 4.14).

16.3 3D visualization of specific brain structures 

 195

Fig. 16.12: A vertex presentation of the cerebellar vermis (arrow) can be reconstructed after a 3D volume acquisition. In this reconstruction, the corpus callosum (CC) can also be seen in this midline section. The asterisk is placed in the cisterna magna.

Fig. 16.13: Orthogonal sectional planes after a transvaginal volume acquisition through the fontanelle in a normal fetus. The intersection dot is placed in the upper plane on the chiasma opticum (long arrows) and the image in the upper right plane was rotated in a way to have the base of the skull horizontal and visualize the chiasma opticum (short arrows) in plane C (lower plane). Note the X-shape of the chiasma (lower plane).

196 

 16 3D Fetal Neurosonography

On encountering suspicious findings while examining the axial planes, the demonstration of a midsagittal view of the vermis with its neighboring structures is of great importance when attempting to diagnose anomalies. In this view, the shape, size and position of the vermis can be objectively assessed. Conditions such as a mega cisterna magna can also be well differentiated from a Blake’s pouch cyst, a partial or complete agenesis of the vermis or a true Dandy-Walker malformation. Other brain structures in 3D multiplanar render mode: Some structures in the brain can be demonstrated directly in a 2D scan, but many others are difficult to visualize depending on the fetal position. In such conditions, it is worth learning the use of the 3D volume acquisition with the reconstruction of the plane of interest. The free movement of the transducer is limited and the structures can be more easily visualized in 3D, especially during a transvaginal examination. Parasagittal views can demonstrate the horns of the lateral ventricles, whereas coronal views provide insight into the basal ganglia, the symmetry of the cortex and other structures of the brain. At the base of the skull, even the optic chiasm can occasionally be visualized using 3D, as illustrated in Fig. 16.13.

16.4 Reconstruction of fetal brain structures in 3D rendering The use of 3D is mainly used for the multiplanar visualization of specific fetal brain structures, but in some situations there is still a space for 3D volume rendering of some regions using the different modes presented in this book. Our experience has shown that some rendering modes like surface, minimum, inversion or glass-body mode as well as Silhouette and Sono-AVC tools are able to be implemented well in fetal neurosonography. Figures 16.14–16.18 demonstrate some examples of different modes and Chapter 21 provides additional images in early brain development in 3D.

16.5 The intracranial vascular system in color Doppler Major intracerebral arteries and veins can be visualized well either from an axial or a sagittal approach. The left and right internal carotid arteries and the basilar artery enter the skull at its base to soon form the Circulus of Willis, which can be easily visualized using color Doppler and 3D glass-body mode (Fig. 12.18). One of the main arteries, visualized in a midsagittal view, is the anterior cerebral artery that continues along the corpus callosum to form the pericallosal and callosomargnial artery. In fetuses with partial or complete agenesis of the corpus callosum, these arteries demonstrate an abnormal course, as can be seen in Fig. 16.19. Recently, the intracranial venous system has been intensively explored. Interest not only focused on the

16.5 The intracranial vascular system in color Doppler 

 197

Fig. 16.14: Surface mode and cranial view into the brain at the level of the transventricular plane, in a normal fetus (a) and in fetuses with abnormal findings. Fetus in (b) with open spina bifida and abnormal head shape (“lemon shaped”)(arrows), in (c) in ventriculomegaly (arrow) and in (d) in choroid plexus cysts (arrows).

Fig. 16.15: Surface mode with HD-live silhouette with cranial view into the brain, in a normal fetus (a) and in fetuses with anomalies, in (b) with holoprosencephaly and monoventricle (curved arrow), in (c) in ventriculomegaly (double arrow) and in (d) in Dandy-Walker syndrome with a dilated posterior fossa with an absence of the cerebellum (arrow).

198 

 16 3D Fetal Neurosonography

Fig. 16.16: Upper panel: Fetus with bilateral ventriculomegaly with the cavum septi pellucidi in between displayed in minimum mode (upper left) and in inversion mode (upper right). Lower panel: fetus with holoprosencephaly and monoventricle in minimum mode (lower left) and the inversion mode of the ventricular shape (lower right).

Fig. 16.17: Fetus with occipital encephalocele in tomographic mode with brain tissue in the cele (*) (see also Fig. 16.18 left).

16.5 The intracranial vascular system in color Doppler 

 199

Fig. 16.18: Surface mode rendering in two fetuses with an occipital encephalocele (left) and suboccipital meningocele (right). In the fetus on the left, brain tissue is recognized in the cele (*) (same fetus as in Fig. 16.17).

Fig. 16.19: Glass-body mode of the anterior cerebral artery with an atypical course in two fetuses with a complete (left) and a partial (right) agenesis of the corpus callosum.

sinuses as the superior and inferior sagittal sinus, straight and transversal sinus, but also on other veins such as the vein of Galen, the internal cerebral vein, and cortical veins (Fig. 12.17). The typical anomalies affecting the veins include the vein of Galen aneurysmal malformation (Fig. 16.20), the pial arteriovenous malformations or abnormal courses of veins, such as falcine sinus.

200 

 16 3D Fetal Neurosonography

Fig. 16.20: Two fetuses with an aneurysm of the Galen vein in color Doppler in glass-body mode and HD-live flow.

16.6 Fetal neurosonography before 14 weeks of gestation Interest in normal fetal anatomy and anomalies in the first 14 weeks’ gestation has increased following the introduction and routine use of nuchal translucency screening. For many years, the evaluation of the brain at this gestational age was reduced to the demonstration of the skull, excluding anencephaly and the visualization of the falx cerebri, excluding alobar holoprosencephaly. With the advent of the intracranial translucency and its potential for early detection of an open spina bifida, there was an increased interest in understanding brain development and anatomy in the first trimester. In many conditions, a 3D volume with tomographic mode display (Fig. 16.21) provides a good overview of the intracranial anatomy, providing the ability to differentiate between normal and abnormal findings (Fig. 16.22). Figure 16.23, in Omniview mode, reveals the intracerebral changes in the brain of a 12 week-old fetus with open spina bifida. Another example is provided in Figure 21.23. Few scientists have further examined the embryonic development of the human brain before 10 weeks’ gestation with 3D ultrasound (Figs. 16.24, 16.25). These days, the study of the embryonic brain can be performed in vivo with 3D ultrasound. In so doing, different multiplanar mode displays are applied to demonstrate the regions of interest, plane by plane. Interestingly, few volume-rendering modes are also able to demonstrate the developing ventricular system. Different 3D render modes are shown in Figs. 16.26–16.29. In the future, with the aid of these techniques, it is expected that more knowledge will become available during this early stage of brain development and that high-risk patients will then be able to be examined earlier in gestation.

16.6 Fetal neurosonography before 14 weeks of gestation 

 201

Fig. 16.21: Transvaginal 3D volume acquisition of head and brain displayed in tomographic mode. In this overview many structures can be seen in one glance as the choroid plexus of the lateral ventricle (Plexus), the falx cerebri (Falx), the aqueduct of Sylvius between both cerebral peduncles (*) and the fourth ventricle as intracranial translucency in an axial view (arrow).

Fig. 16.22: Fetus at 12 weeks’ gestation with holoprosencephaly clearly demonstrated and identified in tomographic mode. No midline is recognized in comparison to Figure 16.21.

202 

 16 3D Fetal Neurosonography

Fig. 16.23: Fetus with open spina bifida and intracranial changes, displayed in Omniview planes. Upper right: axial view of the posterior fossa at the level of the cerebral peduncles with the aqueduct of Sylvius (*), shifted toward the occipital bone. Lower left, compressed posterior fossa without a transparency (arrow). Lower right) thickened brain stem (double arrow).

Fig. 16.24: 3D volume of head and brain at 9 weeks in the orthogonal mode. Both hemispheres are separated; the choroid plexuses and the rhombencephalon can be also good identified.

16.6 Fetal neurosonography before 14 weeks of gestation 

 203

Fig. 16.25: 3D volume of a head at 9 weeks’ gestation. Using Omniview, a midline has been placed and a midsagittal view of the embryonic brain is demonstrated.

Fig. 16.26: Left: Head of a fetus at 12 weeks’ gestation in 3D in surface mode and HD-live display. In the right figure, the head has been opened with Magicut and both hemispheres are recognized with both choroid plexuses (*) separated by the falx cerebri (Falx).

204 

 16 3D Fetal Neurosonography

Fig. 16.27: Head of two fetuses at 12 weeks’ of gestation with a view from cranial displayed in surface mode and HD-live with silhouette effect. Left: Normal anatomy with falx cerebri (Falx) and both halves of the brain with the large plexus choroidei (*). By comparison, in the right image the fetus exhibits the typical features of holoprosencephaly with monoventricle (double arrow) with lack of separation of the thalami.

Fig. 16.28: The lateral ventricles can be visualized before 11 weeks’ gestation with inversion mode (left) and Sono-AVC (right) as digital cast.

16.6 Fetal neurosonography before 14 weeks of gestation 

 205

Fig. 16.29: Head of a normal fetus at 12 weeks’ gestation from a front and lateral view displayed with a silhouette tool and revealing both lateral ventricles.

16.7 Conclusions Fetal neurosonography is an important component of a comprehensive ultrasound to rule out fetal brain malformations, especially in the second half of gestation. The combination of 3D with 2D ultrasound facilitates the assessment of the fetal brain providing the possibility of reconstructing planes that are inaccessible with routine scanning. The reconstruction, as well as the offline analysis and detailed visualization of structures using the multiplanar mode, represents the important benefits of 3D examinations. The tomographic demonstration of sectional planes also offers a reliable tool for comparison with other diagnostic modalities as the MR-examination of fetal brain. The study of early embryonic brain development under normal and abnormal conditions using 3D ultrasound provides significant future potential.

17 3D of the Fetal Skeleton 17.1 Limitations in the assessment of the fetal skeleton using 2D ultrasound The examination of the fetal bones using 2D ultrasound is often limited to the bones that are easily accessible. During a routine screening long bones are measured, the spine is visualized in different planes and the hands and feet are demonstrated. Cranial bones are not evaluated in 2D ultrasound unless indirectly by visualizing the profile or during the measurement of the biparietal diameter. A better approach to visualize the cranial bones and other parts of the fetal skeleton is the examination with 3D or 4D ultrasound in combination with maximum mode rendering, as explained in Chapter 8. This mode enables the demonstration of the skeletal system under normal and abnormal conditions. The combination of the use of multiplanar modes as the orthogonal, tomography or Omniview modes with VCI (see Chapters 3, 4 and 5) can help in the extraction of typical bones displayed in maximum mode. The ideal technique, however, remains the 3D volume acquisition with a maximum mode rendering, as outlined earlier in Chapter 8. The recent introduction of a high resolution VCI-A with maximum mode during a live 4D examination with the electronic matrix transducer (as detailed in Chapter 4) is promising. This chapter discusses the examination of the skeletal system under normal and abnormal conditions.

17.2 3D of fetal spine and ribs The fetal spine can be imaged using various 3D methods, as illustrated in Figs. 17.1–17.6 and explained earlier in Chapter 8. These tools facilitate a good visualization of the spine with the vertebral bodies and arches at the different stages of ossification. In Fig. 8.12, the varying degrees of ossification between the 1st and 2nd trimester can be well recognized. Navigation through the volume enables also the visualization of the vertebral bodies with the corresponding intervertebral disks (Figs. 17.3–17.5). Vertebral bodies can be also visualized separately using Magicut or multiplanar mode, as shown in Fig. 17.5. In a coronal projection, the spine is seen with the ribs, which makes it possible to assess the symmetry and number of ribs (Fig. 17.1). Typical anomalies that are amenable to be assessed using 3D include the different forms of open and closed spina bifida, such as myelocele (Figs. 17.7, 17.8), myelomeningocele (Fig. 17.9), meningocele, lipomeningocele and others. Figures 17.1 to 17.9 demonstrate typical images of fetuses with normal and abnormal spines. Other vertebral abnormalities as hemivertebra or more severe findings, such as kyphoscoliosis, are often already identified using 2D, but the complete picture in its

17.2 3D of fetal spine and ribs 

 207

Fig. 17.1: 3D volume data set of a spine (fetus at 22 weeks) displayed in multiplanar orthogonal rendering mode (left) and in maximum mode (right).

Fig. 17.2: After a static 3D volume acquisition, the user can apply the Omniview mode to demonstrate the planes of interest. Three lines have been placed on the reference plane (upper left): the yellow line reveals a sagittal plane (upper right), while the two oblique planes (magenta and cyan) are placed at the level of the lateral spinal arches and vertebral bodies.

208 

 17 3D of the Fetal Skeleton

Fig. 17.3: Omniview can also be used during the live 4D examination to demonstrate spine and ribs. This is achieved by combining with Volume Contrast Imaging (VCI) (here 14mm thickness) in combination with maximum mode.

Fig. 17.4: In a 3D volume with maximum mode rendering the image can be rotated or the perspective can be changed. Left: View from dorsal on spine and ribs. Middle: Lateral view on the spine with the intact skin covering the spine and right: view in a deeper layer from dorsal with a direct view on the vertebral bodies.

17.2 3D of fetal spine and ribs 

 209

Fig. 17.5: The user can also selectively cut out anatomic structures out of a volume. In this example, one vertebra was cut out (left) and magnified (right). In a cross-section, one can recognize the three ossification centers as the vertebral body and the laminae of the vertebral arches (*).

Fig. 17.6: Biplane display of the spine. In the left image, the spine is visualized in a sagittal plane and an axial cross-section of the spine (right) is displayed along the biplane line.

210 

 17 3D of the Fetal Skeleton

Fig. 17.7: Omniview planes in a fetus with myeloschisis (myelocele) (arrow) revealing an axial cross-section at the level of the defect (upper right), at a level few vertebrae higher than the defect (lower right) and in a coronal direct view on the defect (lower left).

Fig. 17.8: Omniview in a fetus with myeloschisis (myelocele) with a 17 mm slice and surface mode with a direct view of the defect (arrow).

17.2 3D of fetal spine and ribs 

 211

Fig. 17.9: Lateral view of the back of a fetus with lumbosacral myelomeningocele (arrow) in surface mode (left) and in maximum mode (right).

Fig. 17.10: Three fetuses with hemivertebra (arrow) in maximum mode with several deviations of the spine. The extent of deviation is better appreciated and demonstrated with 3D maximum mode rendering.

212 

 17 3D of the Fetal Skeleton

Fig. 17.11: Fetus with a closed spina bifida and severe spine and ribs anomalies in a case of spondylocostal dysostosis. The finding can be better appreciated by switching from surface mode (left) to maximum mode (right).

Fig. 17.12: Spine in maximum mode in a normal fetus (left) and in a fetus with a rare skeletal disease (right). Note the thin ribs and the abnormal arrangements of the low thoracic ribs

17.3 3D of the fetal limbs 

 213

Fig. 17.13: Lateral view of the spine in maximum mode in a normal fetus (left). In the two other fetuses in the middle and right images, one can recognize the interruption of the lumbosacral spine in a case of segmental spinal dysgenesis. In the fetus to the right, there is a severe caudal regression syndrome in the presence of maternal diabetes mellitus.

full extent is better demonstrated using 3D with the maximum mode rendering, as can be seen in Figures 17.10 and 17.11. Anomalies that affect the ribs, either as an isolated finding or part of syndromic conditions, are uncommon. Figures 17.11–17.13 reveal fetuses with anomalies affecting ribs and spine.

17.3 3D of the fetal limbs The limbs can be visualized in 3D not only using the surface mode but also in their bony parts by switching to maximum mode rendering (Figs. 17.14 to 17.18). Ideally, the acquisition plane should be a perpendicular view of the arm or leg (see Figs. 7.4, 7.5 and 8.18) in order to obtain a good perspective of the examined limb. Ideally, the arm or leg should lie horizontally during volume acquisition and the hand or foot should be included in the volume, which makes a good volume often a challenge. However, once this approach is successful, the visualization of the limb reliably confirms the

214 

(a)

 17 3D of the Fetal Skeleton

(b)

(c)

(d)

Fig. 17.14: Maximum mode with the demonstration of the forearm with radius (R), ulna (U) and hand; (a) normal fetus, (b) Fetus with a “mitten-hand” with syndactyly in Apert syndrome, (c) Fetus with absent hand, (d) fetus with radius aplasia, short ulna (arrow) and typical hand position.

Fig. 17.15: A fetus with absent forearm in surface mode (left) and in maximum mode (right).

17.3 3D of the fetal limbs 

 215

Fig. 17.16: The arms of three fetuses with skeletal dysplasias displayed in maximum mode rendering. The fetus on the left had a short-rib polydactyly syndrome as Ellis-van-Creveld syndrome, the fetus in the middle had an osteogenesis imperfecta and in the fetus on the right had a thanatophoric dysplasia.

Fig. 17.17: Clubfoot in a fetus displayed in surface mode (left), in VCI-A with maximum mode (middle) and in static 3D with maximum mode (right).

216 

 17 3D of the Fetal Skeleton

Fig. 17.18: Bilateral clubfeet in surface mode (left) and in maximum mode (right).

normal anatomy and in abnormal conditions the extent of the lesion can be well documented to its full extent (Figs. 17.14 to 17.18). Typical findings affecting the upper and lower limbs can be complex and their spectrum wide (see Figs. 17.14–17.18), as is the case in partial or complete absence of an extremity, in radius aplasia, in various skeletal dysplasias with shortened, bowed or fractured bones or in the frequent finding of a clubfoot. Isolated anomalies of the hands including polydactyly, oligodactyly, cleft hand or syndactyly can be well documented by selective rendering of the hand.

17.4 3D of the facial and cranial bones A 3D visualization of the bones of the head includes both the facial bones (Fig. 17.19) as well as the other cranial bones with their corresponding sutures and fontanelles (Figs. 17.20, 17.21). The information provided by 2D ultrasound in this field is limited and 3D has an incomparable advantage in the rendering of bones. The patterns of the metopic suture have already been intensively examined under normal and abnormal conditions (Fig. 17.19) in a few studies. Some of the typical abnormal findings are the fused metopic suture in fetuses with alobar holoprosencephaly or the wide metopic suture typically found in Apert syndrome in association with craniosynostosis of the coronary suture (Fig. 17.19, Fig. 18.19). The hypoplastic or non-ossified nasal bone can be well detected in the frontal and lateral view (see Fig. 8.15). In the presence of encephaloceles or other “tumors” on the head or face, 3D sonography can help in the assessment as to whether a bony defect is present and can help demonstrate its size. The presence of additional bones, so-called Wormian bones, can occasionally be found in the metopic suture or in the fontanelles, but their clinical impact is still not understood.

17.4 3D of the facial and cranial bones 

 217

Fig. 17.19: 3D maximum mode with a view from anterior on the bony face with the frontal bones and metopic suture. The fetus on the left is a normal fetus; the fetus in the middle has a wide metopic suture in Apert syndrome due to coronal suture synostosis, and the fetus on the right has a holoprosencephaly with synostosis of the metopic suture with cleft lip and palate (*) and trisomy 13.

Fig. 17.20: Maximum mode of a lateral view of the cranial bones in a normal fetus (left) and in a fetus with Apert syndrome and coronal suture synostosis (right). The coronal suture in the fetus on the left can be recognized (arrow) but appears fused in the right fetus (?).

218 

 17 3D of the Fetal Skeleton

Fig. 17.21: A fetus with an unknown skeletal disease of the family of cleidocranial dysplasia. Note the abnormal ossification of the parietal bone (circle). Compare with a normal ossification in Fig. 17.20 left.

17.5 Conclusions Maximum mode is a good prerequisite for accurately examining the fetal skeleton using 3D volume display. Most bones of the body are better examined in a 3D volume than in 2D ultrasound. The normal anatomy can be well differentiated from abnormal conditions affecting the areas of interest such as the spine and ribs, the upper and lower limbs and the bony face and skull. A good prerequisite is a good insonation angle and a high-contrast image. Anomalies of limbs and spine either isolated or as part of skeletal dysplasias can be clearly demonstrated and identified using maximum mode. The assessment of the bony face and skull can be of significant help when assessing syndromic conditions, but there is a learning curve required when obtaining reliable images.

18 3D of the Fetal Face 18.1 The sonographic examination of the face in 2D and 3D ultrasound The 2D ultrasound examination of the fetal face typically focuses on a midsagittal view in order to visualize the profile and a frontal view to demonstrate both orbits and the mouth-nose triangle. A systematic examination generally comprises panning the face longitudinally from left to right across the profile, including an axial sweep in transverse sections from the orbits across nose and lips, and to the upper and lower jaw. The ears are rarely assessed during a 2D examination. The profile is also one of the most important views a pregnant woman expects to see, as it is one of the few ultrasound images that a layperson can are easily identify. However, a fetal face today is best displayed by 3D surface rendering as illustrated in Chapter 7 and in this chapter. One of the main advantages of the 3D surface rendering mode is the ability of demonstrating the entire face in a single realistic view so that that the fetus is personified, intensifying the bonding between mother and child. It is also noteworthy to observe how the facial features change with advancing gestation (Fig. 18.1). Early in the third

Fig. 18.1: Fetal faces displayed in 3D surface mode. Shape and proportions of the face change significantly between 12 weeks’ (left panel), around 22 weeks’ (middle panel) and around 30 weeks’ gestation (panel right).

220 

 18 3D of the Fetal Face

trimester, the images of the face strikingly start to resemble the features of the neonate (see later). 3D ultrasound of the face is, however, not synonymous with surface mode and includes other rendering modes, depending on the clinical question, as will be discussed in this chapter.

18.2 The face in multiplanar display For the examination of the face in multiplanar display, the acquisition is performed best in an axial section with the face revealed in an anterior position and both orbits visible in the initial image, or in a midsagittal view starting from the profile plane. By navigating through the volume either in multiplanar orthogonal (Fig. 18.2) or in tomography mode (Fig. 18.3), the face can then be visualized with all details needed, such as the forehead, the eyes, the nose, the mouth and jaw (Figs. 18.2, 18.3, 18.4). In some situations, the examiner can use the Omniview mode to selectively display a few structures such as the hard and soft palate in normal fetuses (Fig. 18.5) and for applying this approach in fetuses with a cleft lip and palate (Fig. 18.6). Figure 18.7 presents the face of a fetus with microphthalmia in tomography mode.

Fig. 18.2: Face in multiplanar display in orthogonal mode. The intersection point (navigation point) has been placed on the nose and the images rotated and adapted correspondingly.

18.2 The face in multiplanar display 

 221

Fig. 18.3: Face in multiplanar display in tomography mode. The reference plane at the upper left reveals the profile, while the tomographic images present axial parallel slices of the face from the eyes (lower right) to the mandible (upper right).

Fig. 18.4: Demonstration of the hard palate (thick arrow) in orthogonal mode in a normal fetus (left) and in a fetus with bilateral cleft-lip and palate (two arrows).

222 

 18 3D of the Fetal Face

Fig. 18.5: Selective planes in multiplanar mode, displayed as Omniview planes. Three cross-section planes displayed in yellow, magenta and cyan have been placed in the reference plane in the upper left image to illustrate the typical features. The orbits are visualized in the upper right plane, in the lower left plane the nose-mouth triangle view and in the lower right plane an axial view of the maxilla (compare with Fig. 18.6.).

Fig. 18.6: The fetus with a mediolateral cleft lip and palate (arrow), displayed in Omniview mode. The selective placement of the three lines (yellow, magenta and cyan) in the reference plane is enhanced with the additional use of VCI and makes the selective demonstration of the regions of interest possible. The two orbits appear normal and the defect is displayed in a coronal view (lower left) and in an axial view (lower right).

18.3 The normal face in 3D/4D surface mode 

 223

Fig. 18.7: The face in multiplanar display in tomographic mode in a fetus with unilateral microphthalmia (arrow). The different planes reveal the difference between the normal (short arrow) and abnormal eye (long arrow) (lower right panel).

18.3 The normal face in 3D/4D surface mode 3D and 4D visualization of the fetal face is often the first and most desired application to learn in 3D volume sonography. Before starting, care should be taken to ensure that enough amniotic fluid is present in front of the face and no objects (such as hands or umbilical cord) hide the face during volume acquisition. For a good image acquisition, the examiner should proceed like a photographer by approaching the face from a slightly antero-lateral position. In our experience, the most important areas to focus on are the nose and mouth, which should be the central points during volume acquisition. The volume box should be large enough to include adjacent structures to the face. It is important to emphasize a difference in viewing the profile in 2D and 3D ultrasound: Figures 18.8 and 18.9 present two fetal profiles in which the 2D images are acceptable. However, the 3D image of the face in Fig. 18.8 is not as good as the in Fig. 18.9. In acquiring a good 3D image of the face, we recommend keeping the chin, mouth, nose and forehead at a same horizontal level; otherwise, if the mouth area is lower, the mouth-chin region will not be identifiable in 3D, as Figs. 18.8 and 18.9 demonstrate. The 3D effect is even more realistic if, in addition to the face, hands or other structures can be observed together (Fig. 18.10). After a volume is acquired, a few steps undertaken in volume manipulation can significantly improve the final result (also refer to Chapter 3). The 3D volume is first

224 

 18 3D of the Fetal Face

Fig. 18.8: 3D static acquisition of a fetal profile with mouth and chin low in the image, meaning that the distance from the surface to the chin is long, whereas the distance to the forehead is short (left). The result in 3D demonstrates that the mouth-chin area cannot be optimally observed (see the difference compared to Fig. 18.9).

Fig. 18.9: 3D static acquisition of a fetal profile with the face almost horizontal to the mouth, chin and forehead, and almost at the same level in 2D (left). The distance from the transducer to the chin and to the forehead is similar, resulting in a good 3D picture of the face, especially the mouth and chin area, as compared to the previous figure.

18.3 The normal face in 3D/4D surface mode 

 225

Fig. 18.10: Most facial images are manipulated with the Magicut tool to improve the image. Here are two examples. In the upper images, the structures in front of the hand and behind the head were erased. In the lower images, the primary result (left) is acceptable but the nuchal cord might bother the mother and can easily be removed with Magicut (right).

fixed and “unneeded” structures are removed with Magicut (Fig. 18.10). In some cases modifying the level of gain and threshold can help in reducing obstructing or disturbing artifacts. An almost natural effect is achieved by selecting HD-live smooth, which lends the surface a soft skin tone. Moreover, the rendered image can be processed with a softening filter by increasing both the shadow level and the transparency. The position of the light source can be adjusted to lighten the face from the top rather than the front (see Fig. 3.15). Figures 18.11 and 18.12 present a collection of images that were acquired and manipulated in the manner explained earlier. The intensive use of Magicut and other functions is only possible to a limited extent in a 4D examination due to the fact that the images are permanently changing while the fetus is moving. In a live 4D exam, the examiner is mainly concentrating on other details of the face, such as the opening of mouth or eyes, the facial expressions or the hand movements in front of the face. Figures 18.13 and 18.14 illustrate examples of changes in fetal facial expressions as observed in a sequence of subsequent images from 4D volumes. The 4D examination is of particular interest in the third trimester, when features and grimaces become more explicit (Fig. 18.13). In the third trimester, the fetal physiognomy appears very realistic and close to neonatal features. Figure 18.15 illustrates five 3D fetal profiles after 28 weeks’ gestation that emphasize the differences in fetal faces. Figure 18.16 demonstrates the similarities between two fetal faces and their respective neonatal

226 

 18 3D of the Fetal Face

Fig. 18.11: Collection of 3D face images in surface mode around 20–25 weeks’ gestation. Different facial expressions are evident, and at this gestational age, the orbit region frequently appears to exhibit mild exophthalmia, which is normal. The eyes are always closed.

18.3 The normal face in 3D/4D surface mode 

 227

Fig. 18.12: A typical behavior in fetuses is holding the hand in front of the head and face, which can be visualized well in both 3D and 4D.

Fig. 18.13: A small series of images from a cine loop of 4D volumes illustrating two fetal facial expressions. The upper panel shows the fetus smiling, while the lower panel shows the fetus grimacing.

228 

 18 3D of the Fetal Face

Fig. 18.14: During a 4D examination, it is often possible to observe facial expressions such as yawning, swallowing, showing the tongue, smiling, sucking the thumb, thinking and opening the eyes.

Fig. 18.15: In the 3rd trimester after ca. 28 weeks’ gestation, fetuses start to acquire their own personal facial features. The shape of nose and mouth, the facial proportions and the thickness of the cheeks lends the face its typical features, which resemble the postnatal appearance. Here we can observe the profile of five fetuses with different features. The parents often compare the images with a previous child or themselves.

18.4 The abnormal face in 3D/4D 

 229

Fig. 18.16: Comparison of a profile of two 3rd trimester fetuses in 3D with the postnatal profile. Forehead, nose and mouth are often identical pre- and postnatally.

appearance. The assessment of the lateral part of the face with cheek and ear can also be a part of the 3D assessment of the face, and are discussed in the next section.

18.4 The abnormal face in 3D/4D From the very outset of the use of 3D sonography, there has always been significant interest in the demonstration of facial dysmorphism. In addition to specific facial features in 2D, 3D surface mode rendering is still the main 3D display used for this assessment. This mode makes a good demonstration of the proportions of the face and its different regions possible, such as the forehead, the eyes, nose, mouth, chin and ears. Using this approach, abnormal conditions such as microcephaly, macrocephaly (Fig. 18.17), facial anomalies (Figs. 18.18, 18.19), different types of cleft lip and palate (Fig. 18.20), skin tags (Fig. 18.21), trisomy 21 (Fig. 18.22) and other dysmorphic features such as Pierre-Robin syndrome or the flat profile in Binder syndrome (Fig. 18.23) can

Fig. 18.17: Fetuses with abnormal head shape and size. The fetus on the left has a microcephaly and the fetus on the right has a macrocephaly in Apert syndrome (also refer to Fig. 18.19). Note the disproportion between the forehead and the middle face.

230 

 18 3D of the Fetal Face

Fig. 18.18: A fetus with severe facial anomalies involving eyes, nose and mouth in association with holoprosencephaly; the fetus on the left exhibits a proboscis, the fetus in the middle cyclopia and otocephaly and the fetus on the right fetus exhibits arrhinia, median cleft and hypotelorism.

(a)

(c)

(b)

(d)

(e)

Fig. 18.19: A fetus with Apert syndrome at 23 weeks’ gestation with coronal suture synostosis and wide metopic sutures. The typical characteristics of this abnormality are recognizable in these images: (a) profile in 2D with turricephaly, (b) frontal bossing, (c) macrocephaly with hypertelorism and exophthalmia, (d) increasing transparency demonstrating wide metopic suture. In (e), the fetus has its hand in front of its face, and the typical “mitten-hand” is identifiable.

18.4 The abnormal face in 3D/4D 

(a)

(b)

(c)

(d)

(e)

(f)

 231

Fig. 18.20: Fetuses with different facial clefts (arrows): Unilateral cleft lip (a) and (d), mediolateral cleft lip and palate in (b) and (e) and medial cleft lip and palate in (c) and (f).

Fig 18.21: Left: Fetus with a neck lymphangioma, clearly observable with the surface mode. Right: Fetus with tags on the left cheek.

232 

 18 3D of the Fetal Face

Fig. 18.22: Frontal view of the face in 3D surface mode in two fetuses with trisomy 21. A few fetuses are notable due to their opened mouth (images left). One of the interesting features is the proportion of nose and mouth with a small nose and microstomia. Nose and mouth have the same width as compared to normal fetuses, where mouth is larger than the nose width.

Fig. 18.23: Fetuses with facial dysmorphism in 2D and 3D. Upper images: Fetus with a Pierre-Robin syndrome. The finding is very recognizable in 2D (upper left). A line from the chin to the upper lip has a course far from the forehead and can be used in the 3D image of the face as well. Lower images: Fetus with middle face hypoplasia suspecting Binder syndrome or Binder face. The underlying etiology can be different but most commonly it is a chondrodyplasia punctata. In this case, there was a chromosomal anomaly.

18.4 The abnormal face in 3D/4D 

 233

be clearly demonstrated and identified. In this chapter, the reader can compare normal faces in 3D in Figures 18.8 to 18.16 with abnormal faces in Figures. 18.17 to 18.24. The demonstration of the cheek and the ear, which are not clearly observable in 2D ultrasound, has become also part of the 3D evaluation of the face, under both normal and abnormal conditions (Figs. 18.21, 18.24, 18.25). Figure 18.25 illustrates examples of normal and abnormal findings of the ears.

Fig. 18.24: Fetus with a tumor of the eye and orbit in orthogonal mode (left) and in 3D surface mode (right).

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

Fig. 18.25: The ears also can provide variants that can be visualized well in 3D. Normal ears in the upper images in (a) to (d). In the lower panel of images, abnormal ears are illustrated in (e) small ear in trisomy 21, in (f) a dysplastic ear and in (g) and (h) a microtia in syndromic conditions.

234 

 18 3D of the Fetal Face

18.5 The facial bones in 3D/4D The facial bones can be best demonstrated in maximum mode (see Chapter 7). The volume acquisition can be performed either in static 3D, in 4D or in VCI-Omniview mode. Prerequisites for optimal bone visualization in 3D are the reduction of the gain and the increase of contrast in 2D before volume acquisition. The volume is acquired either from a sagittal view or a lateral insonation of the face depending on the region to be displayed. To better highlight the bones in maximum mode, VCI-Omniview slice thickness should be selected around 15–20 mm. In a frontal view of the face in maximum mode, both frontal bones are identifiable with the metopic suture, the two orbits, nasal bone, maxilla and the mandible (Fig. 18.26a). Figures 8.15, 17.19 and 18.26 reveal normal and different abnormal findings in this view. It is also possible to visualize the profile by rotating this view. A lateral insonation and acquisition can visualize the cranial bones with maxilla and mandible (Figs. 4.9, 4.15, 8.17,17.20,17.21). The reverse face view, a view of the face from behind (Fig. 18.27) was introduced as a novel view for the assessment of facial clefts (Fig. 18.28).

(a)

(b)

(c)

(d)

(e)

(f)

Fig. 18.26: Bony face displayed in maximum mode; (a) a normal fetus with the typical landmarks as the metopic suture (1), nasal bone (2), orbitae (3), maxilla (4), and mandible (5). By comparison, fetuses from (b) to (f) exhibit abnormalities of the bony face as b) fetus with craniosynostosis (here in Apert syndrome) with wide metopic suture, (c) with synostosis of the metopic suture, (d) fetus with absent nasal bone, (e) fetus with mediolateral cleft lip and palate and (f) fetus with left-sided microphthalmia with different size of orbits.

18.5 The facial bones in 3D/4D 

 235

Fig. 18.27: Reverse face view in a normal fetus. This approach demonstrates the face from the inside. The projection line is placed in the face with a view from inside to outside.

Fig. 18.28: Reverse face view in a fetus with mediolateral cleft lip and palate.

18.6 Conclusions Despite the wide range of different displays on offer in volume ultrasound, 3D and 4D visualization of the fetal face is still the most commonly performed examination and is still the first 3D image that an examiner learns to perform. Facial anomalies can be displayed quite well in multiplanar mode, but 3D surface rendering provides a spatial view of the face that is often very similar to the postnatal image. A prerequisite for a good 3D image is using a good preset in grayscale 2D prior to acquisition, a large box that includes adjacent structures such as limbs and a good laterally conducted facial insonation rather than from the front. A step-by-step manipulation of the 3D image with Magicut, with different surface modes and smooth skin then enables the demonstration of a very realistic image. Facial features and grimaces become more apparent in the third trimester and are best visualized with 4D ultrasound. Facial anomalies, such as facial clefts or anomalies of the eyes, nose, lips and ears or some syndromic conditions, can be visualized in 3D quite well, but in general this is an augmentation of the information demonstrated in 2D.

19 3D Intrathoracic and Intraabdominal Organs 19.1 Introduction In the previous chapters, the applications of 3D and 4D in fetal echocardiography, neurosonography and in the examination of the fetal face and fetal skeleton have been discussed. The examination of the intraabdominal and intrathoracic organs under normal and abnormal conditions can also be achieved using different 3D modes as have been presented in the Chapters 1–15. It is generally agreed that a finding demonstrated in the tomography mode provides better documentation than a single image or a collection of single images. In abnormal findings, a 3D volume rendering often provides a better picture on the extent of the finding than is possible in 2D. In this chapter, we discuss the potential applications of using different 3D tools for the intrathoracic and intraabdominal organs, summarized in tables and illustrated with examples.

19.2 Intrathoracic organs The typical anomalies affecting the intrathoracic organs (without the heart) include a congenital diaphragmatic hernia mainly focusing on the shifting of the intrathoracic organs (Figs. 19.1–19.3) and the demonstration of the different lung sizes with the hypoplastic lung on the ipsilateral site. Lung anomalies as the congenital cystic adenomatoid malformation (CCAM) (Fig. 19.4), the bronchopulmonary sequestration (Figs. 19.5–19.9) and other cystic lesions also can be visualized using 3D. In hydrothorax, the extent of the lesion can be better assessed and documented with 3D ultrasound (Fig. 19.10) and the volume can be calculated with VOCAL or Sono-AVC. Tomography mode is the best 3D tool for documenting a lesion with its adjacent organs, but the new biplane mode visualized with the electronic probe (see Chapter 14) provides a reliable overview on the extent of the lung lesion with its neighboring organs during a live examination (see Fig. 14.25). Table 19.1 summarizes common diagnoses affecting the intrathoracic organs with suggestions for possible 3D tools that can be applied. Figures 19.1 to 19.10 illustrate examples of 3D visualization of intrathoracic lesions.

19.2 Intrathoracic organs 

 237

Fig. 19.1: Left: In this fetus with a left-sided diaphragmatic hernia, the stomach (*) is left (L) in the thorax and the heart (H) shifted to the right (R) in 2D. Right: In a coronal 3D projection, here displayed in minimum mode, one can recognize stomach and heart side-by-side. Compare with a normal finding in Fig. 9.4.

Fig. 19.2: A left-sided diaphragmatic hernia demonstrated in 3D tomography mode with the stomach (*) adjacent to the heart (H). This can also be identified in the reference plane in the upper left image.

238 

 19 3D Intrathoracic and Intraabdominal Organs

Fig. 19.3: Axial view of the thorax in surface mode in two fetuses with a left-sided diaphragmatic hernia. In the thorax, the heart is shifted to the right (R) and the right lung (RL) is able to be identified, but the stomach position (*) can be alternately be found either posterior or anterior to the left side of the heart.

Fig. 19.4: An isolated cyst in the right thoracic cavity (arrow) demonstrated in tomography mode. This is likely a bronchogenic cyst and surrounded with hyperechogenic lung tissue, but no additional cysts.

19.2 Intrathoracic organs 

 239

Fig. 19.5: Tomography mode for a fetus with a congenital cystic adenomatoid malformation of the lung (CCAM). Arrows are pointing to the multiple middle-size cysts in one right lung lobe.

Fig. 19.6: Tomographic mode in a left hyperechogenic lung with suspected lung sequestration.

240 

 19 3D Intrathoracic and Intraabdominal Organs

Fig. 19.7: Quantification of the volume of the hyperechogenic lung segment in the previous case using the VOCAL tool (see Chapter 15).

Fig. 19.8: The lung lesion in the fetus is displayed with the orthogonal mode and reveals the extent of the finding.

19.2 Intrathoracic organs 

 241

Fig. 19.9: The presence of a feeding artery (arrow) from the descending aorta (AO) is a typical finding in lung sequestration. This finding is ideally demonstrated with 3D glass-body mode and HD-live flow; Heart (H).

Fig. 19.10: Left-sided hydrothorax (*) with the heart shifted to the right and compression of the left lung (arrow). Left: Cranial view in minimum mode; in the images to the middle and right, a view from the left side into the thorax with surface mode displayed in gradient light (middle) and in HD-live (right).

Table 19.1: Typical intrathoracic anomalies with the potential use of different 3D render modes. Anomalies

3D Techniques

Congenital diaphragmatic hernia

Tomography mode Minimum mode Surface mode VOCAL for lung volume calculation

CCAM: Congenital Cystic Adenomatoid Malformation of the lung

Tomography mode Minimum mode Sono-AVC (cysts volume calculation)

242 

 19 3D Intrathoracic and Intraabdominal Organs

Table 19.1: (Continued) Anomalies

3D Techniques

Bronchopulmonary Sequestration

Tomography mode Minimum mode Glass-body mode for visualization of feeding vessel

Hydrothorax

Tomography mode Minimum mode Surface mode Sono-AVC (Fluid volume calculation)

19.3 Intraabdominal organs 19.3.1 The gastrointestinal tract Anomalies of the gastrointestinal tract (GIT) include abnormal position of the stomach (e.g., situs inversus), obstruction of the GIT (e.g., duodenal atresia, ileus) (Fig. 19.11, 19.12) and abdominal wall defects (Fig. 19.13–19.16). Intrahepatic anomalies mainly

Fig. 19.11: Tomographic mode with an antero-posterior view on thorax, diaphragm (arrow) and abdomen with a dilated stomach and duodenum (*) in double bubble sign in a fetus with trisomy 21 at 27 weeks’ gestation. The double bubble sign can be better displayed with the volume rendering, as illustrated in next figure; heart (H).

19.3 Intraabdominal organs 

 243

Fig. 19.12: Double bubble sign in duodenal atresia in minimum mode (upper left) and in inversion mode with HD-live display (upper right). In the lower left panel, the stomach and duodenum are displayed with Sono-AVC after volume calculation while in lower right image; the visualization with the new silhouette tool is presented. The gallbladder (arrow) is also well-visualized in some of these images.

Fig. 19.13: Omphalocele in a fetus at 12 and 16 weeks’ gestation in surface mode.

244 

 19 3D Intrathoracic and Intraabdominal Organs

Fig. 19.14: Omphalocele in a fetus at 24 and 32 weeks’ gestation in surface mode.

Fig. 19.15: Gastroschisis in a fetus at 21 and 26 weeks’ gestation in surface mode rendering. The later the gestational age the more dilated are the bowels (see next figure).

affect the intrahepatic vessels as the agenesis of the Ductus venosus or the interruption of the intrahepatic inferior vena cava with azygos continuation (see Chapter 12). The presence of ascites, either isolated or as part of generalized fetal hydrops, can be well documented with 3D either with tomography mode (Fig. 6.23) or even in surface mode. Surface mode in ascites resembles the image of “virtual laparoscopy” as presented in Figures 19.17 and 19.18. Table 19.2 summarizes common diagnoses affecting the GIT with suggestions for possible 3D tools that can be used. Figures 19.11 to 19.18 illustrate examples of 3D visualization of anomalies of the GIT.

19.3 Intraabdominal organs 

 245

Fig. 19.16: Gastroschisis in a fetus at 28 and then at 32 weeks’ gestation displayed in surface mode and HD-live rendering. Bowels are often dilated in late gestation. In the third trimester (right), the difference between small intestine (short arrow) and colon (long arrow) can be well recognized, especially when highlighted with the silhouette tool; Knee (K).

Fig. 19.17: A fetus with ascites in 2D image (left) and in surface mode (right) with deep dynamic rendering reminding a “virtual laparoscopy”. One can recognize the liver (L), bowel (short arrow) and the bursa omentalis (*) quite well. The bowel and bursa cannot be easily differentiated (see next figure). The long arrow points to the umbilical vein, which postnatally has its course on the liver surface as ligamentum falciforme.

246 

 19 3D Intrathoracic and Intraabdominal Organs

Fig. 19.18: The same fetus with ascites as in previous figure displayed in “virtual laparscopy” here in HD-Live mode with low (left) and high (right) silhouette level. The bursa omentalis (*) appears more transparent than the bowel (arrow); liver (L).

Table 19.2: Typical anomalies of the gastrointestinal system with the potential use of different 3D render modes. Anomalies

3D Techniques

Situs inversus

Tomography mode Minimum mode

Duodenal atresia

Tomography mode Minimum mode und Inversion mode Sono-AVC (Stomach-Duodenum Volume)

Omphalocele / Gastroschisis

Tomography mode Surface mode Glass-body mode

Ileus

Tomography mode Minimum mode

Intrahepatic vessels

Glass-body mode Minimum mode

Ascites

Tomography mode Minimum mode Surface mode

19.3 Intraabdominal organs 

 247

19.3.2 The urogenital System Anomalies of the urogenital system include the obstruction of the upper and lower urinary tract (Figs. 19.19–19.22), cystic dysplastic kidneys (Figs. 19.23–19.28), anomalies of the renal anlage as pelvic kidney, horseshoe kidney and renal agenesis. Ovarian cysts are additionally included in this group (Figs. 19.29–19.30) as well as the assessment of the external genitalia. The latter can be well evaluated in 3D ultrasound,

Fig. 19.19: Bilateral pyelectasia with multiplanar mode using Omniview planes. The three lines were placed to visualize the right and the left kidney in an anterio-posterior view, as well as in coronal view (lower left panel).

Fig. 19.20: Fetus with a bilateral pyelectasia with the demonstration of the dilated renal pelvis with minimum mode.

248 

 19 3D Intrathoracic and Intraabdominal Organs

Fig. 19.21: Tomographic mode of an axial view of the abdomen in a fetus with a vesico-ureteral reflux with hydronephrosis (arrow) and kinking of the ureter (U). The finding can be better appreciated in a volume rendering display; Bladder (BL).

Fig. 19.22: A fetus with a vesico-ureteral reflux with hydronephrosis revealed in 2D image (left), in inversion mode (middle and right images). The bladder (BL) and the dilated kinked ureter (U) can be well recognized with the hydronephrosis (arrow) in this coronal view.

19.3 Intraabdominal organs 

 249

Fig. 19.23: A fetus with bilateral polycystic kidneys (arrows) in an autosomal recessive polycystic kidney disease displayed in tomographic mode. Tomography rendering provides a good overview of the extent of the finding.

Fig. 19.24: Volume Contrast Imaging of the A-plane (VCI-A) with contrast enhancement of the kidneys (arrow) in the left fetus with a normal kidney and in the right fetus with an enlarged polycystic kidney.

250 

 19 3D Intrathoracic and Intraabdominal Organs

Fig. 19.25: A fetus with a multicystic kidney in multiplanar mode.

Fig. 19.26: A fetus from the previous figure with a multicystic kidney visualized in 3D rendering modes as minimum and inversion mode. In the right panel the individual cysts were displayed and calculated separately with Sono-AVC (see Chapter 15, continued in next figure).

19.3 Intraabdominal organs 

 251

Fig. 19.27: A fetus from the previous images with a multicystic kidney. The rendering is in inversion mode and the display is in HD-live, but in the left panel, the light source has been placed behind the kidney while in the right panel, the silhouette function has been activated.

Fig. 19.28: Multicystic renal dysplasia displayed in surface, in minimum, inversion and Sono-AVC mode.

252 

 19 3D Intrathoracic and Intraabdominal Organs

Fig. 19.29: A fetus at 30 weeks’ gestation with an isolated cyst (arrows) localized in the low left abdomen, beneath the stomach (*). The likely diagnosis in the female fetus is an ovarian cyst, displayed here in tomographic mode. The cyst is typical echolucent without echodensity signals inside it. In the image to the right, the volume of the cyst was calculated with Sono-AVC. Compare with the follow-up in the next figure.

Fig. 19.30: A fetus shown in the previous figure with an ovarian cyst (arrows); here four weeks later with cyst hemorrhage.

19.3 Intraabdominal organs 

 253

Fig. 19.31: 3D surface mode in a male (a) and female fetus (b) and in two fetuses with abnormal genitalia in (c) and (d).

which often enables a good differentiation between normal and abnormal findings (Fig. 19.31). Table 19.3 summarizes common anomalies of the urogenital system with suggestions for possible 3D tools that can be applied. Figures 19.19–19.31 illustrate examples of 3D visualization of lesions of the urogenital system. Table 19.3: Typical anomalies of the urogenital system with the potential use of different 3D render modes. Anomalies

3D Techniques

Pyelectasis, hydronephrosis, pelviureteric junction obstruction, vesico-ureteral reflux, duplex kidney with ureterocele

Tomography mode Minimum mode Inversion mode Sono-AVC

Megacystis

Tomography mode Minimum- und Inversion mode Surface mode

254 

 19 3D Intrathoracic and Intraabdominal Organs

Table 19.3: (Continued). Anomalies

3D Techniques

Multicystic und polycystic renal dysplasia

Tomography mode Minimum mode Inversion mode Sono-AVC

Horseshoe kidney, pelvic kidney,

Tomography mode Omniview mode Volume Contrast Imaging

Genital anomalies

Surface mode Tomography mode

Renal agenesis

Tomography mode Glass-body mode Volume Contrast Imaging

19.4 Conclusions The examination of the intrathoracic and intraabominal organs including the gastrointestinal and renal system can be achieved with both the multiplanar and volume display. From a clinical point of view, the most important tool in abnormalities in these regions is tomography mode, with the demonstration of the examined lesion in its extent and with its surrounding anatomy. Moreover, in some specific conditions in fluid-filled organs such as hydrothorax, ascites, duodenal atresia, hydronephrosis or cystic kidneys, or, anomalies in body contours such as in omphalocele, gastroschisis or abnormal genitalia others, volume displays can then provide a more complete spatial view of the lesion.

20 STIC and 3D/4D Fetal Echocardiography 20.1 The sonographic assessment of the heart in two-dimensional ultrasound A fetal echocardiographic examination includes the demonstration and documentation of a series of planes, which include the axial section of the upper abdomen, the four chamber view, the five chamber view, the short axis view, the three-vessel-trachea view and if needed, longitudinal planes of the aortic arch, the ductal arch and the veins. Improved diagnostic accuracy can be achieved by combining grayscale evaluation with the color Doppler to demonstrate diastolic and systolic hemodynamics in the cardiac chambers and vessels. While atria, ventricles and atrioventricular valves are simultaneously visualized in the single four-chamber-view plane, the great vessels can only be assessed by tilting the transducer to demonstrate their origin and their spatial course and relationship to each other. The documentation of a cardiac examination for a later evaluation or for a second opinion still occurs in many places by storing single images or video clips, the main limitation of which, however, is that they only include what has been seen and recorded by the examiner. 3D/4D fetal echocardiography offers important benefits for all the points outlined above, which are explained below in this chapter.

20.2 Acquiring cardiac volumes Acquiring a cardiac volume, as was already described in Chapter 1, can be performed using static 3D, STIC or 4D with a mechanical or electronic transducer. Acquisition type depends on the question to be answered and the methodology was already explained in Chapter 1. Static 3D acquisition This technique of 3D-acquisition is rapid and has a high-resolution. However, wall and valve movements are the main limitation for fetal heart examinations, as they produce movement artifacts. Despite this limitation, a volume with good resolution often provides acceptable information on the anatomy of the chambers and the great vessels, provided the information needed does not depend on wall movements, as is the case in valve atresia or hypokinesia. These can be reliably assessed by viewing the static 3D size of cardiac structures and their relationships to each other. Static 3D cannot be reliably combined with color Doppler, since direction of blood flow is dependent on the phase of the cardiac cycle. The authors prefer the use of power Doppler in static 3D acquisition, given its uniform color display, especially to demonstrate the course of the vessels.

256 

 20 STIC and 3D/4D Fetal Echocardiography

STIC volume acquisition The best cardiac volumes are acquired with STIC technique and these can be ideally used for off-line evaluation of fetal cardiac structures and movements. STIC volume acquisition can occur in combination with a grayscale (Fig. 20.1), color Doppler (Fig. 20.2), power Doppler, and B-Flow modes. Prior to volume acquisition it is recommended that the examiner optimize the color to clearly

Fig. 20.1: STIC volume of a heart displayed in the three orthogonal planes A, B and C,

Fig. 20.2: STIC volume in color Doppler displayed in the three orthogonal planes A, B and C.

20.3 Fetal echocardiography in 3D/4D multiplanar reconstruction 

 257

visualize flow events in heart and vessels (see Chapter 1). The starting plane for the acquisition mainly depends on both the question of interest and the result expected. Volumes acquired for the demonstration of cardiac cavities can be best acquired starting from the four- or five-chamber-plane, whereas volumes for the assessment of the positions of the great vessels and their course are accquired from an axial plane of the upper mediastinum. A longitudinal or oblique acquisition is recommended if the aortic or ductal arch or the abdominal vessels are to be visualized. 4D acquisition with an electronic matrix transducer: With this probe, the 4D examination can be performed in almost “real-time” with the display of 20–30 volumes per seconds. The 4D examination can be performed in different rendering modes as the orthogonal or tomographic modes or in 3D volume rendering displays. The combination with color Doppler is possible, but the frame rate is often too low. An interesting combination is the combination of 4D with VCI-A (see chapter 4), which displays an image with high-contrast as the result of a thin slice scanning instead of a large 4D volume or a single 2D plane. 4D with a matrix probe can be used in arrhythmia. The combination of 4D with different 3D render modes such as color Doppler or inversion mode should make this technique interesting for new applications in the future.

20.3 Fetal echocardiography in 3D/4D multiplanar reconstruction The ultrasound examination of a fetal heart consists on the visualization of different cross-sectional planes that are close to each other and demonstrating the typical structures of interest. These planes can be generated from an acquired volume and displayed in either orthogonal (Figs. 20.1, 20.2) or tomography mode (Figs. 20.3–20.5) or by using few selective Omniview planes (Figs. 20.6, 20.7). The combination with color Doppler enables the assessment of systolic and diastolic events in the cardiac chambers and great vessels. Any cross-sectional plane can be reconstructed from a good digital static 3D or STIC volume. Also, in STIC volume, a single hypothetical cardiac cycle is stored as an infinite loop. A STIC volume can be visualized in slow motion and stopped at any phase of the cardiac cycle, enabling a detailed analysis of different phases of the cardiac cycle (Fig. 20.8). The intracardiac hemodynamic changes can be particularly well analyzed in STIC volumes acquired in combination with color Doppler. Since the complete heart is included in a digital volume any sectional plane (“plane of interest”) can be reconstructed off-line. This allows the previously described typical planes to be extracted from the volume and the needed cardiac examination can be virtually performed. Extracted images can be demonstrated in one of the multiplanar render mode displays as single, orthogonal, tomographic or Omniview planes. The quality of the reconstructed images in grayscale can be slightly increased by adding VCI or SRI tools (see Chapter 4). In several clinical studies, it was demonstrated that such volumes allow a reliable off-line diagnosis and therefore can

258 

 20 STIC and 3D/4D Fetal Echocardiography

Fig. 20.3: STIC volume in tomographic mode presenting different planes, such as the abdomen with stomach (*), heart in the four-chamber view and aorta (Ao) with pulmonary artery (PA) in the upper mediastinum; left ventricle (LV), right ventricle (RV).

be sent to obtain a second opinion or be used in teaching fetal echocardiography. This approach is ideal, especially for teaching purposes, because a cardiac examination can be simulated on a STIC volume and the examiner can learn how to place the typical sectional planes to obtain the information revealing the diagnosis by working on volume samples of cardiac anomalies. In the Figs. 20.1 to 20.11, some examples of normal and abnormal hearts are presented in different multiplanar mode displays. The examination of the heart with biplane mode offers important additional information and was already discussed in Chapter 14.

20.4 Fetal heart in 3D/4D volume rendering Similar to the imaging of a fetal face in surface mode, heart volumes can also be reconstructed in different 3D rendering modes. The rendering can focus on the demonstration of the surface of the walls and lumen in the ventricles or the great vessels or highlight the visualization of blood flow in the heart and the corresponding vessels. Following rendering modes are generally used: In surface mode, the examiner can emphasize the demonstration of the interface between the cardiac cavities and walls. For methodological aspects, please refer to

20.4 Fetal heart in 3D/4D volume rendering 

 259

Fig. 20.4: STIC volume in tomographic mode with the most important planes as four-chamber-view, Five-chamber-view and three-vessel-trachea view.

Fig. 20.5: STIC volume in tomographic mode in color Doppler in diastole with the filling of the right (RV) and left (LV) ventricle and systole with the visualization of aorta (Ao) in five-chamber-view and aorta and pulmonary artery (PA) in the three-vessel-trachea-view.

260 

 20 STIC and 3D/4D Fetal Echocardiography

Fig. 20.6: STIC volume in Omniview display: In the reference plane (upper panel left) where the heart is seen in a sagittal view the Omniview lines are placed at typical levels demonstrating the fourchamber-view (Plane 1, upper panel right), the five-chamber-view (Plane 2, lower panel right) and the three-vessel-trachea view (Plane 3, lower panel left).

Fig. 20.7: STIC volume in Omniview mode in color Doppler. A curved line was drawn and placed directly in front of the atrioventricular (AV) valves and great vessels. The effect in image b) reveals the flow across both AV valves in the right (RV) and left (LV) ventricle. The great vessels lie typically in a position that aorta (Ao) is embedded between both AV-valves and the pulmonary artery (PA) slightly to its right.

20.4 Fetal heart in 3D/4D volume rendering 

(a)

 261

(b)

Fig. 20.8: STIC volume with the visualization of the four-chamber-view. With a STIC volume each phase of the cardiac cycle can be selected and here reveals the systole (a) and diastole with open valves (b).

Fig. 20.9: STIC volume in tomographic mode, illustrating a fetus with dextrocardia with heart on the right (arrow) and stomach (*) left-sided and the cardiac axis pointing the right (R); left (L).

262 

 20 STIC and 3D/4D Fetal Echocardiography

Fig. 20.10: STIC volume in tomographic mode in color Doppler shows in this fetus in comparison to a normal finding (Fig. 20.5) a transposition of the great arteries with a parallel course (arrows) of aorta (Ao) and pulmonary artery (PA); right ventricle (RV), left ventricle (LV).

Fig. 20.11: STIC volume in tomographic mode in color Doppler in systole in a fetus with pulmonary stenosis. aorta (Ao) arising from the five-chamber view appears normal, but a turbulent flow (circle) can be identified across the pulmonary artery (PA).

20.4 Fetal heart in 3D/4D volume rendering 

 263

Chapter 7. Figures 20.12–20.14 illustrate normal and abnormal findings in the fourchamber view. An interesting reconstruction is possible using minimum mode that resembles a projection of an X-ray image (Fig. 20.15). This particular method was extensively explained in Chapter 9. The use of this rendering mode has become less common in recent years, mainly due to the advent of other more sensitive 3D rendering modes. A much more plastic image is displayed when using inversion mode rendering. The heart can be displayed like a 3D digital casting with the visualization of the chambers and great vessels, as was already explained in Chapter 10 and illustrated in Figs. 10.1–10.3. This mode allows viewing the spatial course of the great vessels, as indicated in Fig. 20.16. Glass-body mode alone (Fig. 20.17) or in combination with the HD-live flow function (Fig. 20.18) (also refer to Chapter 12) enables the examiner to make the best spatial visualizations of blood flow in the ventricles and the great vessels. Anomalies in the ventricular plane (Fig. 20.19) as well as the spatial course of the great vessels (Figs. 20.20–20.22) can be demonstrated using this approach. This mode can be well used in cases with abnormal courses of the great vessels, such as a right or double aortic arch or in transposition of the great vessels, but also in hypoplastic vessels or vessels with an atypical course. An interesting application is provided by the combination of B-flow (Chapter 13) with static 3D or STIC. The sensitive signals of blood flow that are demonstrated with

Fig. 20.12: STIC volume with the four-chamber-view demonstrated in surface mode rendering. The projection line (“green line”) is placed over the chambers and under the origin of the aorta.

264 

(a)

 20 STIC and 3D/4D Fetal Echocardiography

(b)

(c)

Fig. 20.13: STIC volume of the four-chamber view in surface mode rendering: In the normal heart (a) both right (RV) and left (LV) ventricle and right (RA) and left (LA) atria are well seen. The fetus (b) has an atrioventricular septal defect (AVSD) (*) and the fetus c) an Ebstein’s anomaly where the tricuspid valve has a lower insertion in the RV (arrow)

(a)

(b)

(c)

Fig. 20.14: STIC volume of the four-chamber view in surface mode rendering in a normal heart (a) showing both right (RV) and left (LV) ventricle and right (RA) and left (LA) atria. In comparison the fetus in (b) has a hypoplastic left heart syndrome (HLHS) with a small LV and the fetus (c) a hypoplastic RV in tricuspid atresia with ventricular septal defect (TA+VSD).

(a)

(b)

Fig. 20.15: STIC volume and minimum mode rendering as demonstrated in a fetus (a) with transposition of the great arteries (TGA), with the aorta (Ao) arising from the right (RV) and the pulmonary artery (PA) from the left ventricle (LV). Fetus (b) has a double outlet right ventricle (DORV) and both Ao and PA are seen to arise from the RV.

20.4 Fetal heart in 3D/4D volume rendering 

(a)

 265

(b)

Fig. 20.16: STIC in a fetus with transposition of the great arteries (TGA), with the aorta (Ao) arising from the right ventricle (RV) and the pulmonary artery (PA) displayed in surface mode (a) and in inversion mode (b).

(a)

(b)

Fig. 20.17: Color-Doppler STIC volume in glass-body mode rendering. The left and right panels reveal that, depending on the position of the projection line (arrows), a different result can be obtained. In the left panel (a), the green line is placed under the origin of the aorta (upper left) and only the four-chamber-view in diastole is visualized here. In the right panel (b), the projection line has been placed over the great vessels (upper right) and in the rendered image one recognizes both the fourchamber-view in the background and the great vessels in the front.

266 

 20 STIC and 3D/4D Fetal Echocardiography

Fig. 20.18: A similar display of a normal heart as seen in Fig. 20.17, but here with the use of the HD-live flow tool with a light source. The light-dark effect with shadowing has the ability to increase the perception of a 3D effect. Compare with images of abnormal hearts in Figs. 20.20–20.22.

(a)

(b)

(c)

Fig. 20.19: A fetus with an atrioventricular septal defect in systole (a) with closed valves, in diastole with open valves (b) with the defect (*) clearly visible. In (c) with color Doppler and HD-live mode, blood flow is illustrated streaming from both atria into the ventricles across the large central defect (*); right and left atrium (RA, LA), right and left ventricle (RV, LV).

20.4 Fetal heart in 3D/4D volume rendering 

 267

Fig. 20.20: STIC with color Doppler and glass-body mode rendering with HD-live flow with a view from left lateral on the great vessels. In a normal finding (left), the crossing of the aorta (AO) and pulmonary artery (PA) is clearly visible in the lateral view merging into the descending aorta (AOD). The fetus in the center has a pulmonary atresia with a reverse flow in the tortuous Ductus arteriosus (DA). On the right, we see a fetus with hypoplastic left heart syndrome; the reverse flow is clearly visible in the tiny aortic isthmus.

Fig. 20.21: STIC with color Doppler and glass-body mode rendering with HD-live flow with a view from the mediastinum on the great vessels in a fetus with a right aortic arch. In the image on the left, the trachea (arrow) can be observed between both the aortic arch (Ao) on the right and the pulmonary artery (PA) on the left on a plane in color Doppler. In the 3D display on the right image, the spatial demonstration of the course of the great vessels better illustrates the course of the vessels merging into the descending aorta.

268 

 20 STIC and 3D/4D Fetal Echocardiography

Fig. 20.22: STIC with color Doppler and glass-body mode rendering (left) and in combination with HD-live flow (right) in two fetuses with a transposition of the great arteries (curved arrows). See the explanation in Fig. 20.16.

B-flow are not only ideal for visualizing large, but also small vessels, such as the pulmonary arteries and veins. In our experience, however, we have found that 3D/4D in combination with B-flow is more complicated to use as compared with other modes and we prefer to use the previously described HD-live flow with 3D or STIC for the spatial visualization of tiny vessels. The calculation of volumes using VOCAL or Sono-AVC offers interesting applications for the calculation of ejection fraction and other volumes, but it is still mainly used in research units rather than in actual clinical practice.

20.5 Conclusions The 3D/4D examination of the heart has revolutionized fetal echocardiography. The significant advantage lies both in the spatial visualization of a heart with the great arteries as well as in the offline manipulation of heart volumes in order to virtually reconstruct any needed sectional plane. To extend the use of 3D and STIC on the fetal heart, efforts should be made in facilitating the acquisition and the compression of volumes and in the improvement on the automatic detection of landmarks within a cardiac volume in order to effectively use a software like the Sono-VCAD on routine scan.

21 3D in Early Pregnancy 21.1 Background The widespread introduction of first trimester nuchal translucency screening between 11 and 14 weeks’ gestation has led to an increased interest in ultrasound screening in early gestation. The use of high-resolution transabdominal and transvaginal transducers has opened a new time window in the diagnosis of fetal malformations in early gestation. From the first sonographic evidence of cardiac activity and until 14 weeks of gestation, the brain, heart, face, extremities and other organs can be examined during early development. During this period, the 3D visualization of the whole fetus is possible using surface and other rendering modes, which offer additional imaging possibilities, as discussed later in this chapter. The fetus can be visualized transabdominally (Fig. 21.1a) but a better resolution is achieved with transvaginal ultrasound (Fig. 21.1b). With the exception of figures 21.1a and 21.2, all other images in this chapter were acquired using 3D transvaginal ultrasound. As discussed in other chapters, different rendering modes can also be applied in early gestation (Fig. 21.2).

21.2 3D volume rendering in early gestation The use of the surface mode is the most commonly used 3D rendering mode in early gestation, as it makes the optimal visualization of the developing embryo and fetus possible. Images acquired using 3D surface mode of the embryo are currently similar to photographic images and drawings from embryology as demonstrated in Fig. 21.3.

(a)

(b)

Fig. 21.1: 3D surface mode providing a picture of the complete fetus at 12 weeks’ gestation by transabdominal (a) and transvaginal (b) examination. The image on the right has a higher resolution.

270 

 21 3D in Early Pregnancy

Fig. 21.2: A fetus after transabdominal examination with surface mode and different rendering modes. The modes used from left to right are: gradient light, HD-live mode, HD-live mode with silhouette, HD-live mode with silhouette with back light source.

(a)

(b)

(c)

(d)

Fig. 21.3: Development of the embryo between 7 and 10 weeks’ gestation with increasing crownrump length from 16 mm (a), 21 mm (b), 29 mm (c) to 36 mm (d).

As early as 11 weeks’ gestation, the integrity of the fetus along with the proportions of the head, trunk, extremities and other details can be reliably demonstrated. Figures 21.4 and 21.5 illustrate fetuses between 11 and 13 weeks’ gestation. Severe anomalies affecting the body surface can be immediately recognized in 3D by clinicians and patients as well, but caution is recommended when relying solely in 3D image before a comprehensive evaluation is obtained in 2D imaging. Figures 21.6 and 21.12 present examples of normal fetuses and fetuses with thickened nuchal translucency, omphalocele, spina bifida, facial anomalies, and arm and leg malformations. Care is advised in assessing the gender in 3D ultrasound in early gestation, as male and female genitalia can appear similar and thus lead to erroneous predictions. 3D ultrasound plays a critical role in

21.2 3D volume rendering in early gestation 

 271

Abb. 21.4: Different fetuses between 11 and 13 weeks’ gestation examined transvaginally with 3D surface mode and gradient light display.

Fig. 21.5: Different fetuses between 11 and 13 weeks’ gestation in 3D surface mode and HD-live mode display

272 

 21 3D in Early Pregnancy

Fig. 21.6: Neck region (arrows) in surface mode in three different fetuses. Left: Normal appearing neck. Middle: thickened nuchal translucency, Right: Nuchal hygroma. The fetus in the middle image had a rare chromosomal aberration and the fetus on the right, Turner’s syndrome.

Fig. 21.7: Surface mode in two fetuses at 12 weeks on the left with a closed anterior abdominal wall (arrow) and on the right with an omphalocele (arrow).

Fig. 21.8: Two fetuses with omphalocele (long arrow). The fetus on the left has a normal looking hand, while the fetus on the right exhibits the typical finding of a radius aplasia, in both cases at high-risk for the presence of a trisomy 18.

21.2 3D volume rendering in early gestation 

 273

Fig. 21.9: Back region of two fetuses at 12 weeks in surface mode. The fetus on the left has a normal looking back, while the fetus on the right has an open spina bifida with myelomeningocele.

(a)

(b)

(c)

(d)

(e)

(f)

Fig. 21.10: A fetal face in normal fetuses at 12–13 weeks’ gestation (a–c) and in abnormal head and face (d-f). The aspect of the anterior face with eyes, nose, mouth and ears is well recognized in the upper images (a–c). In the lower panel, the abnormal fetuses are well recognized with anencephaly (d), with facial anomaly in holoprosencephaly with hypotelorism, cebocephaly and low position of the ear (e) and in (f) in a fetus with facial dysmorphism with cleft lip and palate.

ruling out major fetal malformations in early gestation in pregnant women with a history of prior severe fetal malformations. In anomalies with fluid accumulation in the body, 3D surface mode can also be used in combination with increased transparency mode for a better demonstration of the severity of the lesion, as illustrated in Figs. 11.5 and 21.13. In multiple pregnancies, fetuses can be visualized well along with surrounding structures. Monochorionic and dichorionic twin pregnancies demonstrate different thickness of the amniotic membranes and can be well differentiated,

274 

(a)

 21 3D in Early Pregnancy

(b)

(c)

(d)

Fig. 21.11: Fetal hand in surface mode between 11 and 13 weeks. The images reveal: (a) normal hand, (b) hand with lower arm in radius aplasia, (c) brachydactyly (short fingers) in a case of autosomal dominant inheritance from the mother, (d) the absence of the hand in a fetus with trisomy 21.

Fig. 21.12: Legs in surface mode; left: Fetus with normal legs, middle: Fetus with abnormal leg in context of caudal regression, right: Distal defect of the leg in a fetus with femur-fibula-ulna complex.

but the diagnosis is more reliable performed in 2D ultrasound with the lambda- and T-signs. 3D examples of abnormal twin pregnancies, such as a TRAP sequence or conjoined twins are presented in Fig. 21.16 and can be diagnosed in one glance. Maximum mode is infrequently applied in early gestation due to the reduced level of ossification in the fetal skeleton and the rare diagnosis of skeletal disorders. Figure 21.17 provides an example of maximum mode with demonstration of the spine in a normal and abnormal fetus. One of the interesting applications for 3D sonography in the embryonic and early fetal period appears to be the demonstration of brain structures under normal and abnormal conditions (see Chapter 15). While minimum mode is rarely applied in these conditions, the inversion mode can be used to visualize the intracerebral ventricular system in early gestation (Fig. 21.18). Other tools used include Sono-AVC or the new silhouette technique (Fig. 21.19) (also refer to Chapter 11), with a potential for more clinical applications in the future.

21.2 3D volume rendering in early gestation 

 275

Fig. 21.13: Fetuses with fluid accumulation in the body. Upper panel: Fetus with megacystis and dilated abdomen and the right image demonstrates an opening of the abdomen with Magicut, revealing a dilated bladder (arrow). In the lower panel to the left, one recognizes an intrahepatic cyst in this fetus displayed in a transparency silhouette mode, while the right image presents the cyst after being opened with Magicut (arrow).

Fig. 21.14: Dichorionic diamniotic twin pregnancy at 10 weeks with a thick separating membrane (arrows) between both cavities displayed in surface mode and silhouette (compare with Fig. 21.15).

276 

 21 3D in Early Pregnancy

Fig. 21.15: Monochorionic diamniotic twin pregnancy at 11 weeks with a thin separating membrane between both cavities displayed in surface mode with silhouette (compare with Fig. 21.14.).

Fig. 21.16: Discordant monochorionic twin pregnancies at 11 weeks. Left: The image presents an acardiac twin (arrow) in a TRAP sequence, where TRAP stands for Twin-Reverse-Arterial-Perfusion. Right: Typical 3D surface mode image of thoracopagus as one type of conjoined twins.

21.2 3D volume rendering in early gestation 

 277

Fig. 21.17: Fetal spine in maximum mode left in a normal fetus at 13 weeks and in the right figure in a fetus at 12 weeks with deviated spine in a body-stalk anomaly.

Fig. 21.18: Intracerebral ventricular system of a fetus at 9 weeks in orthogonal mode and static VCI (left) and in inversion mode (right); lateral ventricle (LV), 3rd ventricle (3V), Rhombencephalon (Rb).

278 

 21 3D in Early Pregnancy

Fig. 21.19: Intracerebral ventricular system of a fetus at 8 weeks (left) and 9 weeks (right) displayed with silhouette; lateral ventricle (LV), 3rd ventricle (3V), Rhombencephalon (Rb).

21.3 Multiplanar display in early gestation If a high-resolution image is needed in early gestation, especially in 3D multiplanar reconstruction, acquiring the volume using the transvaginal approach is recommended. Given that embryos and fetuses infrequently are positioned and presented in an optimum position in order to visualize all of the anatomic structures in 2D ultrasound, the acquisition of 3D volumes and the reconstruction of planes can be of significant help (see Figs. 5.1–5.4). This can be achieved by a static 3D scan with the multiplanar, tomographic or Omniview reconstruction of one or more section images. This principle was explained earlier in Chapters 5 and 6. Using tomography mode, the examiner is able to present the complete anatomy of the fetus in one display, thus documenting intracranial anatomy, face and eyes, nose and mouth, chest with heart position, stomach, abdominal wall, kidneys and urinary bladder as well as the limbs. Such a reconstruction cannot be always used for the measurement of the nuchal translucency (as illustrated in Fig. 5.4). On the other hand, in the presence of a thickened nuchal translucency or a cystic hygroma, multiplanar reconstruction can provide reliably an image of a midsagittal view to document the severity of the finding. The fetal spine, the limbs, the profile and the internal organs such as lungs, diaphragm, kidneys and others can be well reconstructed in the sectional planes. The brain is probably the best organ to examine, starting at 7 weeks gestation and using multiplanar mode. One can then follow brain development in a step-by-step fashion into the early second trimester. Figures 21.20–21.25 illustrate examples of the use of the multiplanar orthogonal and tomographic mode in early gestation under normal and abnormal conditions and Fig. 21–26 is an example on the use of STIC with color Doppler.

21.3 Multiplanar display in early gestation 

 279

Fig. 21.20: Fetuses with anomalies (arrow) displayed in a reconstructed cross-sectional plane in multiplanar mode and enhanced with VCI-mode. Upper left: A fetus with omphalocele; upper right, a fetus with nuchal hygroma; lower left, a fetus with anencephaly; and lower right, a fetus with megacystis.

Fig. 21.21: A fetus at 12 weeks’ gestation with hydrothorax (*) in 2D (left) and in tomographic mode (right).

280 

 21 3D in Early Pregnancy

Fig. 21.22: Intracranial structures in multiplanar orthogonal mode with normal brain anatomy. Intracranial translucency (*), slim brain stem (double arrow) and two separated thalami (T) are well recognized.

Fig. 21.23: Intracranial structures in multiplanar orthogonal mode in a fetus with an open spina bifida. The posterior fossa is abnormal with a typically thickened brain stem (double arrow) and almost absent cerebrospinal fluid with no typical intracranial translucency (compare with Fig. 21.22).

21.3 Multiplanar display in early gestation 

 281

Fig. 21.24: The tomographic mode of a fetus with holoprosencephaly reveals the absent falx cerebri with the fused ventricles (*) and thalami (T). Compare this with Fig. 21.22 where the thalami are separated.

Fig. 21.25: Maxilla of a normal fetus at 13 weeks’ gestation (upper panel) and the “maxillary gap” (lower panel) in a fetus with a cleft lip and palate demonstrated in multiplanar orthogonal mode in combination with VCI.

282 

 21 3D in Early Pregnancy

Fig. 21.26: Transvaginal STIC acquisition with color Doppler of a heart with 13 weeks. To the left in multiplanar mode and to the right in glass-body mode. The upper left panel presents the diastolic and lower left panel the systolic phase. The right panel reveals the four-chamber view in glass-body mode with the filling of both ventricles.

21.4 Conclusions The 3D/4D examination has revolutionized the examination of the early embryo and fetus. The combination of transvaginal ultrasound and 3D has the main advantage of reconstructing any plane in order to obtain typical views. Limitations of the manipulation of the transvaginal probe can be overcome by combining with multiplanar 3D reconstruction and different volume rendering modes. Images acquired in high-resolution can provide valuable information on the developing embryo and fetus. Brain structures in particular can be studied in their embryologic development. The external view of the fetus in normal and abnormal conditions can reliably be achieved with the surface mode and is ideally for the visualization of the external structures as face, limbs and anterior abdominal wall, back and others. The accurate examination of the embryo and fetus has been tremendously improved since the introduction of 3D ultrasound.

Further literature references and sources Performing a literature search in PubMed end of 2015 with the words “3D, ultrasound, fetal” reveals around 1,000 hits. We found that in such a monography, it is impossible to present a comprehensive literature list especially given the fact that this book has been conceived as a practical book. We hereby provide a short list of some literature sources, including some books and journal articles, which partly or completely discuss both technical as well as clinical aspects of 3D ultrasound.

Books Abu-Rustum RS. A Practical Guide to 3D Ultrasound. London: CRC Press, Taylor & Francis Group, 2014 Abuhamad A, Chaoui R. A Practical Guide to Fetal Echocardiography: Normal and Abnormal Hearts. 3rd ed. Philadelphia: Lippincott-Williams Wilkins, 2015 Gembruch U, Hecher K, Steiner H. Ultraschalldiagnostik in Geburtshilfe und Gynäkologie, 2. Auflage, Heidelberg, Springer-Verlag, 2016 Kurjak A, Azumendi G. The Fetus in Three Dimensions: Imaging, Embryology and Fetoscopy. London: Taylor & Francis, 2007 Levaillant JM, Bault J-P, Benoit B. Pratique de l´ échographie volumique-Echographie obstetricale. Paris: Sauramps Medical, 2008 Levaillant JM, Bault J-P, Benoit B, Couly G. La Face Foetale Normale et Pathologique : Aspects Échographiques. Paris: Sauramps Medical, 2013 Paladini D, Volpe P. Ultrasound of Congenital Fetal Anomalies. London: CRC Press, Taylor & Francis Group, 2014

Articles Abuhamad A, Falkensammer P, Reichartseder F, Zhao Y. Automated retrieval of standard diagnostic fetal cardiac ultrasound planes in the second trimester of pregnancy: a prospective evaluation of software. Ultrasound Obstet Gynecol 2008; 31: 30–36 Abuhamad AZ. Standardization of 3-dimensional volumes in obstetric sonography: a required step for training and automation. J Ultrasound Med 2005; 24: 397–401 Acar P, Dulac Y, Taktak A, Abadir S. Real-time three-dimensional fetal echocardiography using matrix probe. Prenat Diagn 2005; 25: 370–375 Achiron R, Gindes L, Zalel Y, Lipitz S, Weisz B. Three- and four-dimensional ultrasound: new methods for evaluating fetal thoracic anomalies. Ultrasound Obstet Gynecol 2008; 32: 36–43 Benacerraf BR, Shipp TD, Bromley B. How sonographic tomography will change the face of obstetric sonography: a pilot study. J Ultrasound Med 2005; 24: 371–378 Benacerraf BR. Inversion mode display of 3D sonography: applications in obstetric and gynecologic imaging. AJR Am J Roentgenol 2006; 187: 965–971 Benoit B, Chaoui R. Three-dimensional ultrasound with maximal mode rendering: a novel technique for the diagnosis of bilateral or unilateral absence or hypoplasia of nasal bones in secondtrimester screening for Down syndrome. Ultrasound Obstet Gynecol 2005; 25: 19–24 Benoit B, Chaoui R, Heling KS. Static Volume Contrast Imaging (Static VCI): Principle and Clinical applications. GE-White Papers 2009;: 1–11 Benoit B. The value of three-dimensional ultrasonography in the screening of the fetal skeleton. Childs Nerv Syst 2003; 19: 403–409

284 

 Further literature references and sources

Campbell S, Lees C, Moscoso G, Hall P. Ultrasound antenatal diagnosis of cleft palate by a new technique: the 3D “reverse face” view. Ultrasound Obstet Gynecol 2005; 25: 12–18 Chaoui R, Kalache KD, Hartung J. Application of three-dimensional power Doppler ultrasound in prenatal diagnosis. Ultrasound Obstet Gynecol 2001; 17: 22–29 Chaoui R, Heling KS, Karl K. Ultrasound of the fetal veins part 2: Veins at the cardiac level. Ultraschall Med 2014; 35: 302–18–quiz319–21 Chaoui R, Levaillant JM, Benoit B, Faro C, Wegrzyn P, Nicolaides KH. Three-dimensional sonographic description of abnormal metopic suture in second- and third-trimester fetuses. Ultrasound Obstet Gynecol 2005; 26: 761–764 Chaoui R, Heling KS, Kainer F, Karl K. (Fetal Neurosonography using 3-dimensional Multiplanar Sonography)(German). Z Geburtsh Neonatol 2012; 216: 54–62 Chaoui R, Heling K, Karl K. Ultrasound of the Fetal Veins Part 1: The Intrahepatic Venous System. Ultraschall Med 2014; 35: 208–228 Chaoui R, Benoit B. Volume Ultrasound, rendering modes and clinical application. GE-White Papers 2006;: 1–8 Chaoui R, Hoffmann J, Heling KS. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatio-temporal image correlation (STIC). Ultrasound Obstet Gynecol 2004; 23: 535–545 Chaoui R, Nicolaides KH. From nuchal translucency to intracranial translucency: towards the early detection of spina bifida. Ultrasound Obstet Gynecol 2010; 35: 133–138 Chaoui R, Heling KS. Grundlagen der 3D- und 4D-Echokardiographie beim Fetus unter Nutzung der Spatio-Temporal-Image-Correlation(STIC)-Software. Ultraschall Med 2006; 27: 1–7 Chaoui R, Heling KS. Three-dimensional ultrasound in prenatal diagnosis. CurrOpinObstet Gynecol 2006; 18: 192–202 Chaoui R, Rake A, Heling KS. Drei- und vierdimensionale fetale Echokardiographie. Gynäkologe 2006; 39: 15–24 Chaoui R, Heling KS. New developments in fetal heart scanning: Three- and four-dimensional fetal echocardiography. Semin Fetal Neonatal Med 2005; 10: 567–577 DeVore GR, Falkensammer P, Sklansky MS, Platt LD. Spatio-temporal image correlation (STIC): new technology for evaluation of the fetal heart. Ultrasound Obstet Gynecol 2003; 22: 380–387 DeVore GR, Polanco B, Sklansky MS, Platt LD. The “spin” technique: a new method for examination of the fetal outflow tracts using three-dimensional ultrasound. Ultrasound Obstet Gynecol 2004; 24: 72–82 Deng J. Terminology of three-dimensional and four-dimensional ultrasound imaging of the fetal heart and other moving body parts. Ultrasound Obstet Gynecol 2003; 22: 336–344 Espinoza J, Kusanovic JP, Goncalves LF, Nien JK, Hassan S, Lee W, Romero R. A novel algorithm for comprehensive fetal echocardiography using 4-dimensional ultrasonography and tomographic imaging. J Ultrasound Med 2006; 25: 947–956 Espinoza J, Goncalves LF, Lee W, Chaiworapongsa T, Treadwell MC, Stites S, Schoen ML, Mazor M, Romero R. The use of the minimum projection mode in 4-dimensional examination of the fetal heart with spatiotemporal image correlation. J Ultrasound Med 2004; 23: 1337–1348 Espinoza J, Lee W, Comstock C, Romero R, Yeo L, Rizzo G, Paladini D, Vinals F, Achiron R, Gindes L, Abuhamad A, Sinkovskaya E, Russell E, Yagel S. Collaborative study on 4-dimensional echocardiography for the diagnosis of fetal heart defects: the COFEHD study. J Ultrasound Med 2010; 29: 1573–1580 Goncalves LF, Espinoza J, Romero R, Kusanovic JP, Swope B, Nien JK, Erez O, Soto E, Treadwell MC. Four-dimensional ultrasonography of the fetal heart using a novel Tomographic Ultrasound Imaging display. J PerinatMed 2006; 34: 39–55

Further literature references and sources 

 285

Goncalves LF, Romero R, Espinoza J, Lee W, Treadwell M, Chintala K, Brandl H, Chaiworapongsa T. Four-dimensional ultrasonography of the fetal heart using color Doppler spatiotemporal image correlation. J Ultrasound Med 2004; 23: 473–481 Heling KS, Chaoui R. The Use of the Minimum Mode in Prenatal Ultrasound Diagnostics – Possibilities and Limitations. J Turkish-German Gynecol Assoc 2008; 9: 212–216 Karl K, Heling KS, Chaoui R. Ultrasound of the Fetal Veins Part 3: The Fetal Intracerebral Venous System. Ultraschall Med 2016; 37: 6–26 Kim MS, Jeanty P, Turner C, Benoit B. Three-dimensional sonographic evaluations of embryonic brain development. J Ultrasound Med 2008; 27: 119–124 Lee W, Chaiworapongsa T, Romero R, Williams R, McNie B, Johnson A, Treadwell M, Comstock CH. A diagnostic approach for the evaluation of spina bifida by three-dimensional ultrasonography. J Ultrasound Med 2002; 21: 619–626 Lee W, Goncalves LF, Espinoza J, Romero R. Inversion mode: a new volume analysis tool for 3-dimensional ultrasonography. J Ultrasound Med 2005; 24: 201–207 Leibovitz Z, Haratz KK, Malinger G, Shapiro I, Pressman C. Fetal posterior fossa dimensions: normal and anomalous development assessed in mid-sagittal cranial plane by three-dimensional multiplanar sonography. Ultrasound Obstet Gynecol 2014; 43: 147–153 Martinez-Ten P, Perez-Pedregosa J, Santacruz B, Adiego B, Barrón E, Sepulveda W. Three-dimensional ultrasound diagnosis of cleft palate: “reverse face”, “flipped face” or “oblique face” which method is best? Ultrasound Obstet Gynecol 2009; 33: 399–406 Merz E, Abramowicz J, Blaas HG, Deng J, Gindes L, Lee W, Platt LD, Pretorius D, Schild R, Sladkevicius P, Timor-Tritsch I. 3D imaging of the fetal face - Recommendations from the International 3D Focus Group. Ultraschall Med 2012; 33: 175–182 Merz E, Welter C. 2D and 3D Ultrasound in the evaluation of normal and abnormal fetal anatomy in the second and third trimesters in a level III center. Ultraschall Med 2005; 26: 9–16 Michailidis GD, Papageorgiou P, Economides DL. Assessment of fetal anatomy in the first trimester using two- and three-dimensional ultrasound. The British journal of radiology 2002; 75: 215–219 Moeglin D, Talmant C, Duyme M, Lopez AC. Fetal lung volumetry using two- and three-dimensional ultrasound. Ultrasound Obstet Gynecol 2005; 25: 119–127 Paladini D, Vassallo M, Sglavo G, Lapadula C, Martinelli P. The role of spatio-temporal image correlation (STIC) with tomographic ultrasound imaging (TUI) in the sequential analysis of fetal congenital heart disease. Ultrasound Obstet Gynecol 2006; 27: 555–561 Paladini D, Volpe P, Sglavo G, Vassallo M, De Robertis V, Marasini M, Russo MG. Transposition of the great arteries in the fetus: assessment of the spatial relationships of the arterial trunks by four-dimensional echocardiography. Ultrasound Obstet Gynecol 2008; 31: 271–276 Paladini D, Giovanna Russo M, Vassallo M, Tartaglione A. The “in-plane” view of the inter-ventricular septum. A new approach to the characterization of ventricular septal defects in the fetus. Prenat Diagn 2003; 23: 1052–1055 Paladini D, Sglavo G, Masucci A, Pastore G, Nappi C. Role of four-dimensional ultrasound (spatiotemporal image correlation and Sonography-based Automated Volume Count) in prenatal assessment of atrial morphology in cardiosplenic syndromes. Ultrasound Obstet Gynecol 2011; 38: 337–343 Pashaj S, Merz E. Prenatal Demonstration of Normal Variants of the Pericallosal Artery by 3D Ultrasound. Ultraschall Med 2014; 35: 129–133 Pilu G. Three dimensional ultrasound of cranio-facial anomalies. GE-White Papers 2006;: 1–12 Pilu G, Ghi T. Preliminary experience with Advanced Volume Contrast Imaging (VCI) and Omniview in obstetric and gynecologic ultrasound. GE-White Papers 2012;: 1–6

286 

 Further literature references and sources

Pilu G, Segata M, Ghi T, Carletti A, Perolo A, Santini D, Bonasoni P, Tani G, Rizzo N. Diagnosis of midline anomalies of the fetal brain with the three-dimensional median view. Ultrasound Obstet Gynecol 2006; 27: 522–529 Pilu G, Ghi T, Carletti A, Segata M, Perolo A, Rizzo N. Three-dimensional ultrasound examination of the fetal central nervous system. Ultrasound Obstet Gynecol 2007; 30: 233–245 Platt LD, Devore GR, Pretorius DH. Improving cleft palate/cleft lip antenatal diagnosis by 3-dimensional sonography: the “flipped face” view. Journal of Ultrasound in Medicine 2006; 25: 1423–1430 Pooh RK. Neurosonoembryology by three-dimensional ultrasound. Semin Fetal Neonatal Med 2012; 17: 261–268 Ruano R, Benachi A, Aubry MC, Dumez Y, Dommergues M. Volume contrast imaging: A new approach to identify fetal thoracic structures. J Ultrasound Med 2004; 23: 403–408 Sarut Lopez A, Heling KS, Chaoui R. 3D-Ultraschall in der Pränataldiagnostik. Gynäkologische Praxis 2012;: 23–34 Tonni G, Grisolia G, Sepulveda W. Second trimester fetal neurosonography: reconstructing cerebral midline anatomy and anomalies using a novel three-dimensional ultrasound technique. Prenat Diagn 2014; 34: 75–83 Vinals F, Munoz M, Naveas R, Giuliano A. Transfrontal three-dimensional visualization of midline cerebral structures. Ultrasound Obstet Gynecol 2007; 30: 162–168 Volpe P, Campobasso G, Stanziano A, De Robertis V, Di Paolo S, Caruso G, Volpe N, Gentile M. Novel application of 4D sonography with B-flow imaging and spatio-temporal image correlation (STIC) in the assessment of the anatomy of pulmonary arteries in fetuses with pulmonary atresia and ventricular septal defect. Ultrasound Obstet Gynecol 2006; 28: 40–46 Xiong Y, Chen M, Chan LW, Ting YH, Fung TY, Leung TY, Lau TK. Scan the fetal heart by real-time three-dimensional echocardiography with live xPlane imaging. Journal of Maternal-Fetal and Neonatal Medicine 2012; 25: 324–328 Yeo L, Romero R, Jodicke C, Oggè G, Lee W, Kusanovic JP, Vaisbuch E, Hassan S. Four-chamber view and “swing technique” (FAST) echo: a novel and simple algorithm to visualize standard fetal echocardiographic planes. Ultrasound Obstet Gynecol 2011; 37: 423–431

Index A A-plane 16, 17 Abdomen – Biplane 175, 177 – gallbladder 118, 119, 127, 128, 243 – glass-body mode 149, 150, 154 – inversion mode 127, 128, 243 – minimum mode 117–122 – orthogonal mode 17, 18, 76 – portal venous system 128, 149, 150 – stomach 17, 80, 118, 120, 121, 128, 182, 237 – tomographic mode 76–80, 85–88, 249, 250 – VCI 55, 65 – volume, VOCAL 182, 183 Abdomen, anomalies – Ascites 83, 88, 104, 177, 244–246 – Cyst 275 – Double bubble 83, 87, 104, 120, 121, 128, 183, 242, 243, – Ductus venosus anomaly 120, 149 – Gastroschisis 101, 244, 245, 246 – Hydronephrosis 121, 130, 131, 248, 253 – Ileus 87, 242 – Multicystic kidneys 83, 122, 130, 175, 184, 250, 251, 254 – Omphalocele 101, 137, 176, 243, 246, 272 – Polycystic kidneys 249, 250, 254 – Pyelectasia 121, 247 – Situs inversus 120, 242, 246 Acquisition plane 3, 6 Acquisition3D 11, Acquisition, 4D 11 Acquisition, STIC 11, 70, 90, 131, 151, 152, 155, 157, 256–268 Angle, acquisition 3, 6 Arm, see hand, skeleton Artifact 11, 23, 24, 30, 34, 49 B B-Flow 39, 156–161, 256 B-Plane 16, Biplane  13, 162–177, 209 C CNS – Cerebellum 69, 81, 82, 187, 188, 194, 197

– Chiasma opticum 195 – Corpus callosum 54, 57, 59, 60, 62, 67, 82, 141, 165, 166, 190, 191, 193, – Early pregnancy 124, 141, 200–204, 142, 278, 280 – Glass-body mode 152, 153, 196 – Inversion mode 38, 198, 204, 277, 129 – Omniview 68, 69, 73, 200, 202, 203 – Orthogonal mode 63–65, 193–195 – Silhouette 141, 142, 197, 199, 203–205 – Surface mode 105, 197, – Tomographic mode 81–83, 187–195, 201, 281 CNS, anomalies – Agenesis septum pellucidum 83, 123, 130, 141 – Agenesis corpus callosum 151, 165, 167, 189, 192, 193, 194, 196, 199 – Anencephaly 200, 273, 279 – Choroid plexus cyst 167, 197 – Dandy-Walker Malformation 81, 196, 197 – Encephalocele 60, 166, 198, 199 – Holoprosencephaly 123, 127, 189, 197, 198, 201, 204, 230, 281 – Spina bifida 102, 112, 139, 171, 174, 197, 202, 206, 211, 212, 273, – Vein of Galen aneurysm 161, 200 – Ventriculomegaly 81, 82, 105, 129, 142, 197 C-Plane 16, 17 Cardiac, see STIC and heart anomalies Color Doppler 143–155  E Echocardiography, see STIC and heart anomalies Embryo 47, 49, 73, 129, 134, 141, 142, 203, 270, 274, 278, F Face – Ear 219, 233 – Eyes 220, 223, 228, 230, 234 – maximum mode 106, 114, 115, 234, 235, – Nose 228, 230, – surface mode 223–233 – tomographic mode 220–222 Face, anomalies – Clef lip and palate 139, 169, 170, 217, 221, 222, 231, 234, 273, 281

288 

 Index

– Ear anomaly  219, 233 – Eye anomaly 220, 223, 228, 230, 234 – in holoprosencephaly 230, – Lymphangioma 170, 231 – Syndrome 215 ,217, 229, 230, 232, 234 Foot 100, 101, 213, 215 G Gain 43, 93–95 Glass-body mode 143–155, 196, 265–268 H Hand HD-Live 36, 37, 133–136, 226, 275 HD-Live flow, see Glass-body mode Heart, see STIC Heart, anomalies – Atrioventricular septal defect 264, 266, – Azygos vein continuity 60, 149, 150, 155, 244 – Dextrocardia 261 – Double aortic arch 151, 160, 263 – Double outlet right ventricle 264 – Ebstein’s anomaly 264 – Hypoplastic left heart syndrome 60, 151, 264, 267 – Pulmonary atresia 267 – Pulmonary stenosis 262 – Rhabdomyoma 139, 140, 173 – Right aortic arch 90, 151, 159, 267 – Transposition of the great arteries 123, 131, 132, 152174, 262, 264, 265, 268, – Tricuspid atresia 264 – Ventricular septal defect 172, 173, 264 I Init 22, 23 Intersection point  17–20 Intrauterine device (IUD) 55 Inversion mode  125–132 L Light source 40, 41, 46, 47, 135, 137, 225, 251, Lung, normal 38, 52, 53, 67, 83, 84, 118, 119, Lung, anomalies – Congenital diaphragmatic hernia 123, 237, 238 – Cyst 238 – Cystic adenomatoid malformation 239 – Hydrothorax 84, 122, 241

– Hyperechogenic lung 85, 176, 239 – Lung sequestration 85, 176, 241 – Volume, VOCAL 240 M Magicut 43–45, 136 Malformation, see organ Matrix transducer 162–177 Maxillary gap 170, 281 Maximum mode 106–116, 206–218 Microphthalmia  213 Minimum mode 117–124 Multiplanar reconstruction 62–92 N Navigation 17–23 Navigation dot, see intersection dot 17–19 Neurosonography, see brain O Omniview 54–58, 64–74 Orientation 29, 33 R Render box 30, 33, 34, 126 Rotation 19, 23 S Scrolling 19, 22, 23, 75 Skeleton, see maximum mode  Skeletal, anomalies – absent nasal bone 106, 112, 115, 216, 234 – Apert syndrome 100, 214, 216, 217, 229, 230 – Clef lip and palate 112, 137, 139, 167, 169, 170, 216, 217, 220, 221, 231, 273, 281,  – Clubfoot 99, 101, 116, 215, 216 – Craniosynostosis 112, 216, 234 – Feet anomalies 101, 116, 213, 215, 216, 274 – Hand anomalies 214, 215, 216, 230, 274 – Hemivertebra 112, 114, 206, 211, 212 – Skeletal dysplasia 213–216  – Spina bifida 102, 112, 139, 171, 174, 197, 202, 206, 211, 212, 273 Silhouette 133–142, 153, 154, 155, 243, 246, 270, 275, 276 Skull bones 54, 57, 70 Sono-AVC 178, 182–184, 204, 243, 246, 250, 251, 252, Static 3D 11

Index 

STIC 9, 12, 13, 89, 90, 104, 126, 131, 151, 152, 155, 157, 255–268 Surface mode 35–37 T Threshold 41–43, 94–96 Tomography mode 62, 75–92 Translation 19, 22, 23, 75 Twins 102, 274, 276 – conjoint 276 – TRAP 276 V VCI 49–61 VCI-A 58–60, 127, 132

VCI-Omniview 49, 56, 114, VCI, static 49 VOCAL 178–184 Volume box 3–9 Volume Contrast Imaging (see VCI) Volume data sets  15 Volume measurement 178–184 Volume rendering 29–48 X X-Ray mode 38, 53, 56, 66, 124,

 289