基于惩罚交叉熵极小的正电子放射成像技术

描述了一种正电子放射的成像技术，这种方法以下列问题的解作为待估放射浓度参数的估计值：极小化实测光子计数数据与待估浓度图像矢量的投影间的交叉熵，并同时满足近似光子总数恒定及非负性约束。作者形式地推导出一种不动点迭代算法用来解上述优化问题。基于模拟及实际数据对此算法进行了测试并与一些标准算法做了对比，实验结果显示了新算法的有效性。

An Approach to Positron Emission Tomography Based on Penalized Cross-Entropy Minimization

This paper describes an approach to positron emission tomography, which aims to minimize the cross-entropy between the detected photon coincidence counts and the projection due to image vector, subject to the almost conservation in total tube counts and nonnegativity constraints. The authors derive formally a fixe d-point iterative algorithm to solve the resulting optimization problem. The algorithm is applied to both simulated and real data,and compared with some standard algorithms. The results demonstrate its effectiveness.