中文摘要：

目的 提出一种基于四分位距值的调强放射治疗（IMRT）计划靶区及危及器官剂量体积直方图（DVH）差异分析方法 .方法回顾分析22例宫颈癌IMRT计划,将靶区和危及器官（膀胱、直肠和左右股骨头）5组DVH曲线从Pinnacle3计划系统中导出,对每组DVH曲线求出平均DVH曲线和四分位距值曲线.结果 计划靶区DVH在54.03 Gy处差异最大,四分位距值为6.95%,处方剂量包绕的靶区体积为（96.43±1.63）%.膀胱DVH在17.24 Gy处差异最大,四分位距值为14.62%,V40和V30分别为（32.79±7.06）%和（56.47±9.94）%.直肠DVH在35.92 Gy处差异最大,四分位距值为19.94%,V40和V30分别为（30.17±10.80）%和（58.16±11.99）%.膀胱与直肠的四分位距值差异有统计学意义（z=-6.59,P〈0.05）.左侧股骨头DVH在16.06 Gy处差异最大,四分位距值为31.47%.右侧股骨头DVH在17.47 Gy处差异最大,四分位距值为32.82%,左右股骨头间的四分位距值间差异无统计学意义（P〉0.05）.结论 四分位距值曲线可以分析靶区和危及器官DVH曲线的变化趋势,为自动计划优化参数设置提供指导.

英文摘要：

Objective To study the dose volume histogram （ DVH ） difference of target and organs at risk in intensity-modulated radiation therapy （ IMRT ） planning of cervical cancer based on the interquartile range. Methods A total of 22 cervical cancer patients′ IMRT plans were retrospectively analyzed. Five groups of DVH curves of PTV and OARs （ bladder, rectum, left and right femoral heads） were derived from the Pinnacle3 planning system. The mean DVH curves and interquartile range of volume （ IQRvol ） as a function of dose were calculated from each group′s DVH curve. Results The maximum volume variation for PTV reached at IQRvol（54. 03 Gy） =6. 95%, the volume of the target reached by the prescription dose was （ 96. 43 ± 1. 63 ）%. The maximum volume variation for bladder reached at IQRvol （17. 24 Gy） = 14. 62%. V40 and V30 for bladder were （ 32. 79 ± 7. 06 ）% and （ 56. 47 ± 9. 94 ）%, respectively. Rectal DVH showed the largest difference at 35. 92 Gy, with the quartile of 19. 94%, the V40 and V30 for rectum were （30. 17 ± 10. 80）% and （58. 16 ± 11. 99）%, respectively. The difference in the bladder and the rectum was statistically significant （z = -6. 59, P 〈 0. 05）. The maximum volume variation of left femoral head reached at IQRvol（16. 06 Gy） =31. 47%. The maximum volume variation of right femoral head was reached at IQRvol （17. 47 Gy） =32. 82%. There was no significant difference in IQRvol between right and left femoral head （P〉0. 05）. Conclusions Quartile range curve of DVHs can be used to analyze the deviations of DVH curves of PTV and OARs, and provide guidance for automatic planning optimization parameter setting.