IMPLEMENTING CONE BEAM ALGORITHMS FOR CLINICAL SPECT

单光子发射计算机断层扫描(SPECT)已经成为一种重要的 核医学中的诊断工具，用于心脏、大脑、肝脏的成像， 骨骼和其他器官，因为它是非侵入性的，而且更便宜 而不是其他方式--例如正电子发射断层扫描(PET)。 锥束准直器和SPECT的使用提高了灵敏度 用于对大脑和心脏等小器官进行成像 大视场旋转伽马相机将小器官放大到 一个很大的探测器区域。锥形光束准直器的使用已经证明 与使用平行孔相比，提高了灵敏度和分辨率 准直器，并已导致更好的检测和大小， 例如，心脏病变的扫描时间没有增加。但是，有了 目前的SPECT系统只有一个平面轨道， 锥束层析成像在某些结构中的应用 切片相声。对于像大脑这样的器官，这个系统会显示图像 沿头顶附近旋转轴的伸长(冠状) 以及由于数据不足而产生的矢状观。为了获得足够的数据，它 焦点的扫描轨迹必须在 通过的任何平面至少有一个交点 重建的感兴趣区域。这可能需要修改硬件 然而，即使有足够的数据采样， 不能保证可用的、高效的分析算法。 拟议的研究将开发重建锥形波束的算法。 有足够数据采样的预测。特别强调的是 开发适用于SPECT系统的可修改算法 提供临床系统的额外成本很少或没有额外成本。整体而言 该项目的目标是开发高效的锥形束重建 提供无伪影三维SPECT的算法 用于临床的重建。这一结果将有可能 显著提高核医学的诊断能力 程序。 该项目完成后，将开发和测试软件。这个 研究将为设计制定工程原则和指导方针 获得足够数据用于锥束断层扫描的SPECT系统 获取和处理足够的锥束数据的临床方案。

Single-photon emission computed tomography (SPECT) has become an important diagnostic tool in nuclear medicine for imaging the heart, brain, liver, bones, and other organs because it is non-invasive and is less expensive than other modalities--for example, positron emission tomography (PET). The use of cone-beam collimators with SPECT has increased the sensitivity for imaging small organs like the brain and heart by taking advantage of the large field-of-view rotating gamma camera to magnify small organs onto a large detector area. The use of cone-beam collimators has demonstrated improved sensitivity and resolution over the use of parallel hole collimators, and has resulted in better detection and sizing of, for example, cardiac lesions with no increase in scanning time. However, with present SPECT systems, which have only single planar orbits, the application of cone-beam tomography to some structures shows slice-to- slice cross-talk. For organs like the brain this system shows image elongation along the axis of rotation near the top of the head in coronal and sagittal views due to insufficient data. To achieve sufficient data it is necessary for the scanning trajectory of the focal point to have at least one point of intersection for any plane passing through the reconstructed region of interest. This may require hardware modifications of present SPECT systems; however, even if sufficient data sampling is available, efficient analytical algorithms are not guaranteed. The proposed research will develop algorithms for reconstructing cone-beam projections with sufficient data sampling. Of special emphasis is the development of algorithms applicable to SPECT systems that can be modified with little or no additional cost to present clinical systems. The overall goal of the project is to develop efficient cone-beam reconstruction algorithms that provide artifact-free three-dimensional SPECT reconstructions for clinical use. The results will have the potential to significantly improve the diagnostic capability of nuclear medicine procedures. Upon completion of the project, software will be developed and tested. The research will develop engineering principles and guidelines for the design of SPECT systems that acquire sufficient data for cone-beam tomography and clinical protocols for acquiring and processing sufficient cone-beam data.