An efficient estimation method for reducing the axial intensity drop in circular cone-beam CT.
An efficient estimation method for reducing the axial intensity drop in circular cone-beam CT.
复制标题
一种减少圆锥束CT轴向强度下降的有效估计方法。
DOI：
10.1155/2008/242841
复制
发表时间：
2008
影响因子：
7.6
通讯作者：
Fahrig R
中科院分区：
文献类型：
--
作者：
Zhu L;Starman J;Fahrig R
圆形锥束（CB）扫描的重建算法已被广泛研究的文献。由于测量的数据不足，因此对于这样的几何形状不可能进行精确的重建。如果重建算法（例如标准FDK算法）假定缺失数据为零，则所产生的CB伪影的主要类型是沿轴向方向沿着的强度下降。许多算法已被提出来提高图像质量时，面对这个问题的数据丢失，然而，一个有效的和计算效率的算法的发展仍然是一个重大挑战。在这项工作中，我们提出了一种新的方法来估计未测量的数据，并减少强度下降的文物。通过Grangeat一阶导数在Radon空间中分析每个CB投影。假设CB投影是从一个平行光束的几何形状，我们提取这些数据，驻留在氡空间的未测量区域。然后，这些数据被用于平行射束几何结构中以计算校正项，该校正项与Hu的校正项一起被添加到FDK结果以形成最终重建。然后对附加项的计算进行更多的近似，并且非常有效地实现了最终公式。该算法的性能进行了评估，使用计算机模拟分析幻影。重建结果与其他现有算法的比较表明，该算法在减少轴向强度下降伪影方面具有上级性能，计算效率高。
Reconstruction algorithms for circular cone-beam (CB) scans have been extensively studied in the literature. Since insufficient data are measured, an exact reconstruction is impossible for such a geometry. If the reconstruction algorithm assumes zeros for the missing data, such as the standard FDK algorithm, a major type of resulting CB artifacts is the intensity drop along the axial direction. Many algorithms have been proposed to improve image quality when faced with this problem of data missing; however, development of an effective and computationally efficient algorithm remains a major challenge. In this work, we propose a novel method for estimating the unmeasured data and reducing the intensity drop artifacts. Each CB projection is analyzed in the Radon space via Grangeat's first derivative. Assuming the CB projection is taken from a parallel beam geometry, we extract those data that reside in the unmeasured region of the Radon space. These data are then used as in a parallel beam geometry to calculate a correction term, which is added together with Hu's correction term to the FDK result to form a final reconstruction. More approximations are then made on the calculation of the additional term, and the final formula is implemented very efficiently. The algorithm performance is evaluated using computer simulations on analytical phantoms. The reconstruction comparison with results using other existing algorithms shows that the proposed algorithm achieves a superior performance on the reduction of axial intensity drop artifacts with a high computation efficiency.