Rawool-Sullivan Mohini W, Sullivan John P, Tornga Shawn R, Brumby Steven P
MS B230, IAT-1, Los Alamos National Laboratory, Los Alamos, NM 87545, USA.
Appl Radiat Isot. 2008 Dec;66(12):1986-91. doi: 10.1016/j.apradiso.2008.06.040. Epub 2008 Jul 6.
Compton imaging is used to predict the location of gamma-emitting radiation sources. The X and Y coordinates of the source can be obtained using a back-projected image and a two-dimensional peak-finding algorithm. The emphasis of this work is to estimate the source-to-detector distance (Z). The algorithm presented uses the solid angle subtended by the reconstructed image at various source-to-detector distances. This algorithm was validated using both measured data from the prototype Compton imager (PCI) constructed at the Los Alamos National Laboratory and simulated data of the same imager. Results show this method can be applied successfully to estimate Z, and it provides a way of determining Z without prior knowledge of the source location. This method is faster than the methods that employ maximum likelihood method because it is based on simple back projections of Compton scatter data.
康普顿成像用于预测伽马射线辐射源的位置。利用反投影图像和二维峰值查找算法可获取源的X和Y坐标。这项工作的重点是估计源到探测器的距离（Z）。所提出的算法使用了在不同源到探测器距离下重建图像所对的立体角。该算法通过洛斯阿拉莫斯国家实验室建造的康普顿成像原型（PCI）的实测数据以及同一成像器的模拟数据进行了验证。结果表明，该方法可成功应用于估计Z，并且提供了一种无需事先知晓源位置就能确定Z的方法。该方法比采用最大似然法的方法更快，因为它基于康普顿散射数据的简单反投影。
