论文部分内容阅读
充分考虑调强放疗计划系统产生的计划剂量数据中存在较多梯度边缘点的特点,提出了一种基于梯度特征的调强放疗计划剂量插值算法(TDAGI)。算法首先改进了传统Canny边缘探测方法,并根据计划剂量数据平面的梯度信息来获取该平面上各梯度边缘点和非梯度边缘点;对梯度边缘点追踪梯度剖面并计算其所对应的锐度,对非梯度边缘点计算自定义的偏离系数。使用锐度和偏离系数,通过TDAGI所构造的分布函数得到以该数据平面上每一点为双三次插值中心点时对应的双三次插值核的系数,对每一个待插值点进行特定的双三次插值,从而得到每一个待插值点的计划剂量数据。结果表明TDAGI与传统双三次插值算法以及目前治疗计划系统中广为使用的双线性插值算法相比,在减小误差的同时,具有梯度保持方面的优势,因此TDAGI对于调强放疗计划剂量插值有一定的实际应用意义。
Considering the characteristics of more gradient edge points in the planned dose data generated by IMRT planning system, a dose-enhanced IMRT dose-interpolation algorithm (TDAGI) based on gradient features was proposed. The algorithm first improves the traditional Canny edge detection method and obtains the gradient edge points and non-gradient edge points on the plane according to the gradient information of the planned dose data plane. The gradient edge points are traced to the gradient profile and the corresponding sharpness is calculated, Calculate custom deviation coefficients for non-gradient edge points. Using the sharpness and the deviation coefficient, the coefficients of the bicubic interpolation kernels with each point in the data plane as a bicubic interpolation center point are obtained through the distribution function constructed by TDAGI, and a specific bicubic interpolation is performed for each point to be interpolated , So as to obtain the planned dose data for each point to be interpolated. The results show that compared with the traditional bi-cubic interpolation algorithm and the widely used bilinear interpolation algorithm in the current treatment planning system, the TDAGI has the advantages of reducing the error and maintaining the gradient. Therefore, There is a certain practical significance.