Ratib Osman
Department of Radiology and Department of Molecular and Medical Pharmacology, Ahmanson Biological Imaging Center, David Geffen School of Medicine, University of California at Los Angeles, Los Angeles, California 90095-1721, USA.
J Nucl Med. 2004 Jan;45 Suppl 1:46S-55S.
The advent of multimodality imaging scanners combining PET and CT has led to a new paradigm in image display and presentation that raises new challenges in workstation interpretation software, image navigation, and communication. The essence of multimodality imaging is the ability to overlay imaging information from different modalities in a visually compelling fashion. This is accomplished by combining functional and anatomic data into multidimensional views using color-encoding techniques that provide visual clues on the spatial distribution of image data.
Combined PET/CT scanners provide spatially registered images from the two modalities acquired simultaneously in a single imaging session. Special reconstruction software and image display programs are required to rescale the native images from different spatial resolution into orthogonal or oblique reformatted planes in which data from PET images are color coded and superimposed on corresponding anatomic CT images. The color overlay technique allows the user to visually identify areas of high tracer activity and determine the underlying anatomic structure. Because of the multidimensional nature of the data, visualization requires interactive multidimensional navigation techniques that allow the viewer to move the visualization planes through three spatial directions and two additional dimensions. The fourth dimension is the continuum blend from PET to CT fusion, and the fifth is the dynamic range of the CT images that can be adjusted to display different tissue characteristics, such as bones, soft tissue, and lungs. Software tools currently available are often relatively complex, requiring the user to perform cumbersome maneuvers and time-consuming image manipulation to navigate through all dimensions and obtain adequate image settings and plane positioning for diagnostic interpretation of the image data. Moreover, the ability to convey these images to referring physicians is usually limited because of the lack of adequate viewing software. Distribution of results is usually performed instead through static "snapshots" of the fused images generated by the interpreting radiologist. The ability of the referring physician to navigate through the set of multimodality image data is thus limited.
The wider adoption of multimodality PET/CT imaging techniques in routine clinical use will depend heavily on the development of more adequate image display and navigation tools that allow interpreting physicians to navigate easily and efficiently through multiple dimensions of data. Distribution of results to referring physicians and care providers also requires new tools for interactively reviewing the multimodality data, and current static images obtained from fused image data remain inadequate for proper visualization of the true content of images.
结合正电子发射断层扫描（PET）和计算机断层扫描（CT）的多模态成像扫描仪的出现，带来了图像显示和呈现的新范式，这在工作站解读软件、图像导航和通信方面提出了新挑战。多模态成像的本质是能够以视觉上引人注目的方式叠加来自不同模态的成像信息。这是通过使用颜色编码技术将功能和解剖数据组合成多维视图来实现的，该技术为图像数据的空间分布提供视觉线索。
PET/CT联合扫描仪在单次成像过程中同时提供来自两种模态的空间配准图像。需要特殊的重建软件和图像显示程序，将来自不同空间分辨率的原始图像重新缩放至正交或斜向重新格式化平面，其中PET图像数据进行颜色编码并叠加在相应的解剖CT图像上。颜色叠加技术允许用户在视觉上识别高示踪剂活性区域并确定潜在的解剖结构。由于数据的多维性质，可视化需要交互式多维导航技术，使观察者能够在三个空间方向和另外两个维度上移动可视化平面。第四个维度是从PET到CT融合的连续混合，第五个维度是CT图像的动态范围，可进行调整以显示不同的组织特征，如骨骼、软组织和肺部。目前可用的软件工具通常相对复杂，要求用户执行繁琐的操作和耗时的图像处理，以在所有维度上导航并获得用于图像数据诊断解读的适当图像设置和平面定位。此外，由于缺乏合适的查看软件，将这些图像传达给转诊医生的能力通常有限。结果的分发通常通过解读放射科医生生成的融合图像的静态“快照”来进行。因此，转诊医生浏览多模态图像数据集的能力受到限制。
多模态PET/CT成像技术在常规临床应用中的更广泛采用将在很大程度上取决于开发更合适的图像显示和导航工具，使解读医生能够轻松、高效地在数据的多个维度中导航。向转诊医生和护理人员分发结果也需要用于交互式查看多模态数据的新工具，并且从融合图像数据获得的当前静态图像对于正确可视化图像的真实内容仍然不足。
