National Digital Switching System Engineering & Technological R & D Centre, Zhengzhou, Henan, China.
J Xray Sci Technol. 2012;20(1):69-77. doi: 10.3233/XST-2012-0319.
In circular cone-beam computed tomography (CT), although the minimum data filtered-backprojection (MD-FBP) algorithm has many significant applications, such as handling super-short scan problem, its reconstruction efficiency is limited by the heavy calculation of backprojection. In this paper, aiming at the image reconstruction of flat region in a super-short scan, an improved method based on MD-FBP algorithm is developed using an integral operation with fixed integral interval during the implementation of backprojection, which has an improvement in reconstruction efficiency and parallel performance compared with the original MD-FBP algorithm. It is found that if the thickness of the flat region is less than 0.0349 R (R is the scanning radius), the uncertainty of the method can be ignored. When the thickness of reconstructed region is a little fat, it can also be reconstructed by increasing the scanning radius befittingly. The results of numerical simulation and real data experiments have demonstrated the correctness and merits of the proposed method.
在环形锥束 CT 中,最小数据滤波反投影(MD-FBP)算法虽然有许多重要的应用,例如处理超短扫描问题,但由于反投影的计算量很大,其重建效率受到限制。本文针对超短扫描中的平区域图像重建,提出了一种改进的方法,即在反投影过程中使用固定积分间隔的积分运算,与原始的 MD-FBP 算法相比,该方法在重建效率和并行性能方面都有所提高。结果表明,如果平区域的厚度小于 0.0349R(R 为扫描半径),则可以忽略该方法的不确定性。当重建区域的厚度稍大时,也可以通过适当增加扫描半径来进行重建。数值模拟和实际数据实验的结果验证了所提出方法的正确性和优点。