摘要目的:比较近距离治疗两种临床常用的剂量计算方法和基于CT影像的蒙特卡罗程序计算的剂量差异，探讨组织非均匀性对宫颈癌近距离治疗剂量评估的影响。方法:回顾性选取2018年1月至2020年6月在安徽省肿瘤医院接受三维近距离治疗的宫颈癌患者11例，分别采用美国医学物理师协会(AAPM)TG43号报告的纯水剂量计算方法(TG43-BT)、快速非均质剂量计算算法Acuros BV(BV-BT)和基于治疗CT影像的EGSnrc蒙特卡罗程序(MC-BT)计算各计划的剂量分布，分析比较3种算法的靶区剂量( D98、 D90和 D50)、剂量区体积( V3 Gy、 V6 Gy、 V9 Gy和 V12 Gy)和危及器官(OARs) D2 cm 3剂量差异。 结果:TG43-BT中HRCTV D90为6.274 Gy，比MC-BT高出了近5%，且TG43-BT对靶区体积剂量和各剂量区体积均存在过高评估；除靶区 D50和高剂量区 V12 Gy外，BV-BT和MC-BT对于靶区剂量计算差异无统计学意义( P>0.05)。此外，BV-BT和MC-BT在直肠、小肠和乙状结肠的 D2 cm 3评估上差异无统计学意义( P>0.05)，而MC-BT膀胱 D2 cm 3为4.609 Gy，比TG43-BT、BV-BT明显偏高。 结论:TG43-BT未考虑组织非均匀性影响，普遍高估了靶区和多数OARs受量；BV-BT计算效率高，且在多数靶区和OARs评估参数上与MC-BT计算基本一致，但在近源位置和被充盈膀胱的剂量计算上存在不足，临床评估时仍需谨慎。

abstractsObjective:To evaluate the impacts of tissue heterogeneity on dose calculation of cervical brachytherapy by comparing the doses calculated by two clinically used dose calculation method and the CT image-based Monte Carlo (MC) method.Methods:This study retrospectively selected 11 patients with cervical cancer treated with 3D brachytherapy in Anhui Provincial Cancer Hospital from January 2018 to June 2020. The dose distribution of each plan was calculated via three methods, dose calculation method described in American Association of Physicist in Medicine(AAPM) Task Group No. 43 Report (TG43-BT), Acuros BV(BV-BT) used to perform accurate dose calculations in high-dose-rate (HDR) brachytherapy with phantom heterogeneity, and CT image-based EGSnrc tool kit used to perform Monte Carlosimulation (MC-BT). The dose volumes( V3 Gy, V6 Gy, V9 Gy, and V12 Gy), target volume doses( D98, D90, D50), D2 cm 3 of organs at risk (OARs) calculated by the three methods were compared. Results:The HRCTV D90obtained by TG43-BT was 6.274 Gy, which was even overestimated by around 5% compared to the result calculated by MC-BT. Meanwhile, TG43-BT overestimated the dose volumesand the target volume doses compared to MC-BT.Except for D50 and V12 Gy, the differences between the doses to tumor calculated by BV-BT and MC-BT were not statistically significant( P>0.05). There was also no significant statistical difference between the D2 cm 3 of rectum, small intestine, and sigmoid calculated by BV-BT and MC-BT ( P>0.05). In contrast, the dose to D2 cm 3 of bladder determined by MC-BT was 4.609 Gy, which was notably higher than those deter mined by TG43-BT and BV-BT. Conclusions:TG43-BT overestimated the doses to tumor targets and most OARs since the effects of tissue heterogeneity were not taken into consideration. BV-BT performed efficient calculation and most of the dose distributionin target volume and OARs obtained by BV-BT were consistent with that calculated by MC-BT. Nevertheless, low accuracy occurred for the regions near the sources and full bladder, which warrants further caution in clinical evaluation.