中文摘要：

目的测试基于患者解剖图像和矩阵探测器在线测量进行重建（RDBMOM）的调强放疗（IMRT）三维剂量验证系统的准确性，评估其临床应用可行性。方法分别在体模设计规则野和非规则野测试计划，同时以指形电离室和二维电离室阵列测量各测试例的点剂量和平面剂量分布，评估RDBMOM系统剂量重建精度。选择2例鼻咽癌IMRT计划做RDBMOM验证，分析验证结果的临床应用意义。结果与指形电离室点测量结果比较，RDBMOM系统对各测试例的重建剂量偏差均〈1％（3cm×3cm小野除外），IMRT测试例的重建剂量偏差最大为2．12％。与电离室阵列测量比较，RDBMOM重建的测量平面内离轴剂量分布曲线符合良好，两者比较的1通过率（3％／3mm）为94．56％～100％。2例IMRT计划的RDBMOM验证结果整体γ通过率〉99％；计划靶体积γ通过率〉98％且D95误差〈0．4％，腮腺和晶体平均剂量的最大误差分别达2．97％和59．58％。结论测试系统剂量重建精度可满IMRT验证要求，并能给出与患者解剖结构相关的体积剂量误差与误差位置等信息，有利于评估其对临床的影响。

英文摘要：

Objective To test a three-dimensional dose verification system, which reconstructing dose to anatomy based on modeling and online measurements （RDBMOM） , and to evaluate the accuracy and feasibility of its application in clinical intensity-modulated radiotherapy （IMRT） quality assurance. Methods Phantom plans of regular and irregular fields were selected for the testing. All test plans were implemented and the dose distributions were measured using the thimble ion-chamber and two-dimensional ion-chamber array, the accuracy of RDBMOM were then evaluated by comparing the corresponding results. Two practical treated nasopharyngeal carcinoma IMRT plans were verified with RDBMOM and the clinic significancy were valued. Results Compared with measurements of the thimble ion-chamber, deviations of RDBMOM were within 1% in all tested cases except small field of 3 cm ~ 3 cm. The largest deviation of reconstructed dose in IMRT cases was 2. 12%. The dose profile reconstructed by RDBMOM coincided with the measurement using two-dimensional ion-chamber array. The γ rates （3%/3 mm） were 94. 56% - 100%. The RDBMOM verification of IMRT cases shown that the γ rate 〉 99% in total and 〉 98% in planning target volume, deviation in D95 〈 0. 4%, but the largest deviations in mean dose of the parotids and lens were 2. 97% and 59.58% respectively. Conclusions Accuracy of the tested system satisfies the demand of IMRT dose verification. RDBMOM is able to provide information of volumetric dosimetry and anatomical location of dose error, which is benefit for evaluating the clinical value of verification results.