Madsen M T
Department of Radiology, University of Iowa, Iowa City 52242, USA.
Radiographics. 1995 Jul;15(4):975-91. doi: 10.1148/radiographics.15.4.7569142.
The article explores the fundamentals of emission computed tomography (CT) from a nonmathematical approach. Tomographic images reveal the internal distributions of radioactivity in three-dimensional objects, and thus allows anatomic localization and improves contrast. Tomography requires a stable distribution of radionuclides, uniform detector response, an accurate center of rotation, and a complete set of projections. In emission CT, a large number of measurements, called projections, are collected at various angles about the patient during the examination. This information is organized by the angles of acquisition into a stack, called a sinogram. Each projection is modified by applying a reconstruction filter (eg, ramp or windowed reconstruction filters). These modified projections are backprojected to form the transverse tomographic images. The quality of tomographic images generated from filtered backprojection depends on the underlying assumptions about the projections. Typical artifacts that result from violations of these assumptions include motion, uniformity, and attenuation artifacts. In addition, an inaccurate center of rotation, insufficient angular sampling, and errors in selection of pixel size can result in poor-quality reconstructed images.
本文从非数学角度探讨发射型计算机断层扫描（CT）的基本原理。断层图像揭示了放射性物质在三维物体中的内部分布，从而实现解剖定位并提高对比度。断层扫描需要放射性核素的稳定分布、均匀的探测器响应、精确的旋转中心以及完整的投影集。在发射型CT中，检查期间会围绕患者在各个角度收集大量测量值，称为投影。这些信息按采集角度组织成一个堆栈，称为正弦图。每个投影通过应用重建滤波器（如斜坡或加窗重建滤波器）进行修改。这些修改后的投影进行反投影以形成横向断层图像。由滤波反投影生成的断层图像质量取决于关于投影的基本假设。违反这些假设导致的典型伪影包括运动伪影、均匀性伪影和衰减伪影。此外，旋转中心不准确、角度采样不足以及像素大小选择错误会导致重建图像质量不佳。
