前言：本书介绍

Preface

For much of the past century, projection radiography has been the workhorse in the diagnostic imaging clinic. Tomosynthesis, which introduces depth information to the x-ray radiographic image with little or no increase in radiation dose, could potentially replace projection radiography as we move further into the twenty-first century. This book, Tomosynthesis Imaging, offers the most comprehensive resource to date for this new emerging imaging technology.
Digital tomosynthesis imaging is a novel quasi-three-dimensional x-ray imaging modality that has been primarily developed during the past two decades, owing to the availability of large-area digital x-ray detectors. The tomosynthesis image is reconstructed from a sequence of projection images acquired from a limited angle x-ray scan; therefore, conceptually, tomosynthesis might be considered as limited-angle CT. Because of the limited angle acquisition, resolution in the reconstructed volume is not isotropic. The resolution in image planes parallel to the detector surface is similar to the native detector resolution, but the resolution perpendicular to the detector surface direction is substantially worse, and depends on the scan arc length and on the size of the detail being imaged.
Tomosynthesis imaging is being actively investigated for use in a variety of clinical tasks. Currently, tomosynthesis breast imaging is at the forefront, having received approval for clinical use in Europe and Canada in 2008, and FDA approval in the United States in 2011. Although conventional mammography has been very successful in reducing the breast cancer mortality rate, its sensitivity and specificity are less than desirable, especially for women with dense breast tissue. By providing tomographic information, breast tomosynthesis promises to greatly improve visualization of important diagnostic features. In addition to breast imaging, the detection of lung nodules by chest tomosynthesis and detection of hairline fractures with tomosynthesis skeletal imaging are also being investigated. The number of clinical applications for tomosynthesis imaging will, most likely, increase in the future.
This book provides an in-depth understanding of the tomosynthesis image formation process that will allow readers to tailor tomosynthesis systems for new clinical applications. The characteristics of the tomosynthesis reconstructed volume depend strongly on system design parameters; therefore, it is important to gain an understanding of the underlying factors and their effects on the reconstructed volume. This book provides an in-depth coverage of system design considerations, as well as image reconstruction strategies. It also describes the current state of clinical applications of tomosynthesis, including imaging of the breast and chest, as well as its use in radiotherapy. While use of tomosynthesis imaging for these clinical applications is at an early stage, they illustrate the merits of tomosynthesis imaging and its breadth of potential uses.
This book is written for clinicians and researchers. With breast tomosynthesis being approved for clinical use in several countries, medical institutions are faced with the decision to add tomosynthesis to their suite of imaging devices. Clinicians as well as other hospital personnel involved in the purchase and use of clinical tomosynthesis systems should find this book helpful in understanding the principles of tomosynthesis. This book may also be used as classroom teaching material or in workshops to improve understanding of tomosynthesis images, regardless of clinical application.
The book is divided into five sections. Section I introduces tomosynthesis imaging with a historical perspective (the principle of tomosynthesis dates back to work by Ziedsdes des Plantes in 1929). Section II discusses imaging system design considerations, including acquisition parameters, system components, optimization, and modeling schemes. The purpose of this section is to acquaint the reader with the flexibility of tomosynthesis acquisition and the impact of physical factors in the various schemes. Section III reviews image reconstruction algorithms that have been developed for tomosynthesis, including filtered back-projection methods that are used in most clinical systems, as well as advanced iterative methods that have the potential to reduce artifacts and improve image quality. Section IV describes system evaluation methodologies, including radiologist performance studies and assessment using mathematical model observer assessment. Finally, Section V is dedicated to current clinical applications, which include breast, chest and therapy applications, and concludes with a discussion of future directions for tomosynthesis.
The goal of this book is to cover the fundamentals of tomosynthesis imaging. As tomosynthesis and its applications are evolving rapidly, clinical trials to assess the clinical performance of tomosynthesis imaging are ongoing, and we hope that our readers will join us in eagerly awaiting the outcome of these trials.
EDITORS
Ingrid Reiser
Chicago, Illinois
Stephen J. Glick
Worcester, Massachusetts

前言

在过去的一个世纪里，投影放射治疗技术一直是临床诊断成像的主力军。随着我们进入二十一世纪，断层合成这种将深度信息引入X射线图像且几乎不增加剂量的方法，有可能取代目前的投影放射治疗技术。本书《断层合成成像》，为这一新兴的成像技术提供了迄今为止最全面的资源。
得益于大尺寸数字X射线检测器的应用，数字断层成像这种一种新型的准三维X射线成像方法，在过去二十年得到了迅速的发展。断层合成的图像是依据有限角度X射线扫描所采集的一系列投影图像重建的；因此，从概念上讲，断层合成可能被视为有限角度CT。由于采集角度的限制，重建体积中的分辨率不是各向同性的。平行于探测器表面的成像平面的分辨率与探测器原始分辨率相似，但是垂直于探测器表面的分辨率要差很多，这取决于扫描的弧长和成像物体的尺寸。
断层合成成像正被积极研究用于各种临床任务中。目前，乳腺断层合成成像技术位于最前沿，并且2008年在欧洲和加拿大获得了临床使用的批准，2011年获得了美国FDA的批准。虽然传统的乳腺X摄影在降低乳腺癌的死亡率方面已经非常成功，但其敏感性和特异性并不理想，特别是对于乳腺组织致密的女性来说。通过提供断层信息，乳腺断层有望大大提高重要诊断特征的可视化。除了乳腺之外，过胸部断层检测肺结节以及通过骨骼断层检测细微的骨折也正在被研究。断层合成成像的临床应用未来可能会大大增大。
本书对断层合成图像的形成过程提供了一个深入的理解，使得读者可以为新的临床应用量身定制断层合成系统。断层合成重建体的特征与系统设计参数密切相关，因此，了解根本因素及其对重建体积的影响十分重要。本书提供了一个深入的覆盖系统设计的考虑因素和图像重建策略。同时也描述了断层合成目前临床应用情况，包括乳腺和胸部成像，以及其在放疗中的应用。尽管断层合成成像的临床应用还处于早期阶段，但它证明了断层合成成像的价值及其应用潜力。
本书是为临床医生和研究人员写的。随着乳腺断层合成在好几个国家被批准用于临床，医疗机构面临着将断层合成新增入其成像设备组件中的决定。临床医生和其他参与购买及使用临床断层成像系统的医院员工应该发现，本书对于理解断层成像的原理很有帮助。不论临床应用如何，这本书也可以用作课堂教学材料或在研讨会中用于提高对断层合成图像的理解。
本书分为五个部分。第一章从历史的角度介绍了断层合成成像（断层合成成像的原理可以追溯到1929年 Ziedsdes des Plantes的研究）。第二章讨论了成像系统的设计考虑，包括采集参数，系统组成，优化以及建模方案。本章节的目的是让读者了解到断层合成采集中的灵活性以及各种方案中物理因素的影响。第三章回顾了断层合成中的图像重建算法，包括在大多数系统中使用的滤波反投影方法。以及具有减少伪影提高图像质量的高级迭代方法。第四章描述了系统评估方法，包括放射科医生的绩效研究和使用数学模型的观察者评估。最后，第五章讲解了目前的临床应用，包括乳腺，胸部和治疗的应用，并以讨论未来断层合成的发展方向结尾。
本书的目标是涵盖断层合成成像的基本理论方法。随着断层合成及其应用的快速发展，用于评估临床表现的临床试验正在进行中，并且，我们希望我们的读者能加入我们，急切的期待这些试验结果。
编辑
Ingrid Reiser
Chicago, Illinois
Stephen J. Glick
Worcester, Massachusetts