节点文献

乳腺癌保乳术后影像引导全乳靶区确定的研究

Image-guided Determination of the Whole Breast Target after Breast-conserving Surgery

【作者】 徐敏

【导师】 于金明;

【作者基本信息】 山东大学 , 临床医学(专业学位), 2015, 博士

【摘要】 第一部分勾画者及勾画标准对保乳术后全乳靶区勾画的影响目的:通过对比乳腺癌保乳术后不同勾画者勾画的全乳靶区的差别,探讨勾画标准在靶区勾画中的作用。方法:入组12例早期乳腺癌保乳手术后拟行术后放疗的患者,按照治疗体位行CT模拟定位,得到的CT图像传送至三维治疗计划系统,由5位勾画者依据个人认识分别在12套图像上勾画全乳靶区,随后由资深放疗医师和影像科医师共同商讨制定勾画标准,而后由5位勾画者参照勾画标准重复勾画12位患者的全乳靶区,比较参照勾画标准前靶区体积在勾画者之间的差异、参照勾画标准后靶区体积在勾画者之间的差异以及参照勾画标准前后每位勾画者自己所勾画靶区在体积和空间结构上差异。结果:参照勾画标准前5位不同勾画者所勾画CTV之间差异有统计学意义(P=0.009),参照勾画标准后5位勾画者所勾画CTV之间差异无统计学意义(P=0.209),其中3位勾画者在参照勾画标准后对自己之前所勾画的全乳靶区做了较大调整,前后体积差异有显著性(P<0.05)。5位勾画者制定标准后勾画的靶区对制定标准前的靶区的包含度范围为0.645-0.917,勾画者之间差异显著(P=0.000);5位勾画者标准前后靶区的匹配指数范围为0.593-0.830,勾画者之间差异显著(P=0.000)。结论:由于影像学分辨率的限制和勾画者主观认识上的差异,全乳靶区勾画者之间的有明显差异,加强对医师的培训,制定统一的勾画标准是十分必要的。第二部分基于CT图像乳腺癌保乳术后全乳靶区勾画方法的研究目的:对比不同勾画方法所勾画全乳靶区在体积及空间位置上的差异,探讨全乳靶区勾画的合理方法。方法:选择15例早期乳腺癌保乳术后拟行放疗的患者,按照治疗体位行3DCT模拟定位,定位前依据临床触诊在乳腺周围放置环形金属丝,扫描图像传送至三维治疗计划系统,由资深放疗医生分别基于体表标记所界定乳腺解剖范围、触诊标记所界定乳腺解剖范围和CT图像显示腺体范围勾画全乳靶区,并分别定义为CTVan、CTVpa和CTVg1,比较3种方法所勾画全乳靶区的体积大小及相互包含度(DI)和适形指数(CI)。结果:CTVan与CTVpa间体积差异没有统计学意义(P=0.08),而CTVan与CTVgl间、CTVpa与CTVg1间差异均有统计学意义(P=0.000,P=0.000)。CTVan与CTVpa的CI (0.644±0.122)显著大于CTVan与CTVgl及CTVpa与CTVgl的CI(0.264±0.1080和0.328±0.115)(P=0.000,P=0.000);CTVpa与CTVan相互间的D1分别为0.890-±0.08和0.709±0.144, CTVgl对CTVan和CTVpa的DI分别为0.994±0.005和0.989±0.008。CTVpa与CTVan内、外、上、下边界差值分别为:3.35±7.23mm、5.57±13.37mm、1.75±11.62mm而且上界与内界差值与的体积呈显著负相关11.25±4.07mm,结论:三种方法勾画的全乳靶区之间具有明显差异,其中基于CT显示的腺CTVpa体勾画的靶区明显小于另外两个靶区。基于触诊标记所界定乳腺范围并参考体(P=0.005, P=0.015)。表解剖标记进行全乳靶区勾画是相对合理的。

【Abstract】 Part One:The influence of delineator and delineating criteria on the delineation of the whole breast target after breast-conserving surgeryOBJECTIVE:To explore The influence of delineator and delineating standard on the delineation of the whole breast target after breast-conserving surgeryMETHODS:Twelve breast cancer patients with stage of T1N0M0 were choosed,for the patients, CT simulation images were got, silver clips were marked in the cavity and gross targe formed by seroma in the cavity were visible. Five delineators delineated respectively the whole breast target named as CTV, whereafter, in accordance with the criteria formulated for breast target delineation, the delineator repeat to delineate the CTV for the every patient.The differences between the volumes of CTV delineated by the different delineators with or without the criteria for the target determination were compared.RESULTS:The difference of the CTV delineated by the different delineators was statistically significant (P=0.009) without the criteria.,not statistically significant (P=0.029) with the criteria, and there was statistical significance between the volumes of CTV outlined by the same delineator with and without the criteria in three of the five delineator (P<0.05).The DI range of the target before criteria to the after was 0.645~0.917, and the difference among the five delineators was statistically significant. The MI range of the target was 0.593-0.830, and the difference among the five delineators was statistically significant.CONCLUSION:For variation of the subjective understanding, the whole breast targets have significant diference among doctors. Application of the criteria of target delineation is the key to eliminate population difference of the delineators to outline the target.Part two:Comparison of the whole breast target volume delineated according to surface marks, palpation and glandular breast tissue on CT images after breast-conserving surgeryObjectives:To compare the different methods in delineating the whole breast target volume and to explore the contouring criteria after breast-conserving surgery.Materials and methods:In 15 patients with breast cancer after breast-conserving surgery, the whole breast target was delineated in 3DCT simulation images each by three different methods. The target delineated according to the anatomical landmarks were named CTVan, according to breast palpation named CTVpa,and according to the mammary gland showing in CT images named CTVgl. The volumes of CTVan, CTVpa and CTVgl, The degree of inclusion(DI) and conformal index (CI)between the targets were measured.Results:The mean volumes of CTVan, CTVpa and CTVg1 were 792.229±282.246 mm3,618.331±295.903 mm3 and 196.825±117.618 mm3 respectively. The difference between CTVan and CTVpa had no statistical significance (P=0.08), the difference between CTVan and CTVgl, the difference between CTVpa and CTVgl both had statistical significance (P=0.000, P=0.000). CI between CTVan and CTVpa (0.644±0.122) was larger than CI between CTVan and CTVg1 (0.264±0.108) and CI between CTVpa and CTVgI (0.328±0.115) (P<0.05). DI between mutual CTVpa and CTVan (were 0.890±0.08 and 0.709±0.144), DI of CTVgl to CTVanand to CTVpa were 0.994±0.005 and 0.989±0.008。between mutual CTVgl and CTVan (0.994±0.005,0.264±0.109) and DI between mutual CTVgl and CTVpa (0.989±0.008, 0.329±0.115) had all statistical significance (P=0.006,0.000,0.000).The boundary difference of CTVan and CTVpa were:3.35±7.23mm, 5.57±13.37mm,1.75±11.62mm, and 11.25±4.07mm respectively. The cranial and medial boundary have negative correlation with CTVpa (P=0.005, P=0.015).Conclusions:Combination of the breast palpation and anatomical marks may be helpful in delineating the whole breast target volume was relatively reasonable at present.

  • 【网络出版投稿人】 山东大学
  • 【网络出版年期】2016年 04期
节点文献中: