【作者】 梅欣；

【导师】 傅小龙；

【作者基本信息】 复旦大学 ， 肿瘤学， 2011， 博士

【摘要】 第一部分膀胱癌千伏级锥形束CT (kVCBCT)与模拟定位CT图像配准方法学的研究目的：在MIM软件中,探讨膀胱癌放疗中,千伏级锥形束CT (kVCBCT)与模拟定位CT图像配准的最佳方法。材料和方法：选择2008年10月至2010年12月在我院行根治性放疗的膀胱癌患者12例,分别选取其模拟定位CT图像及第一次放射治疗前的kVCBCT图像进行配准。采用MIM图像处理软件中的自动配准方式,分别运行刚性辅助配准(Rigid Assisted)、运行基于边框辅助配准(Box-based Alignment)及运行基于勾画配准(Contour-based Alignment)。由同一名医生在三个不同时刻对12例患者的模拟定位CT图像与第一次治疗前kVCBCT生成的CT图像之间进行三种方法的配准。分别记录三种方法配准两幅图像中心的坐标在LR、SI及AP三个方向上需要移动的距离值和所用时间。由另外一名医生和一名物理师做为评价者进行独立评价,采用单盲打分的办法进行三种方法配准质量的评分。结果：由同一名医生在三个不同时间对12例患者进行三种不同方法的配准,共得到两幅图像中心的坐标距离配对数据及所用配准时间108组。结果显示运行基于边框辅助配准及运行基于勾画配准,其配准后其图像中心坐标移动距离之间的差距基本上在1mm以内,而刚性辅助配准结果移动范围略有增大。基于勾画配准方法所用时间最长,时间在44.81秒,基于勾画配准方法得到图像配准质量评分最高(达到72分)。结论：在MIM医学图像处理分析软件中,基于勾画配准方法是KVCBCT和模拟定位CT之间配准质量最好的方法,所用时间在可接受范围内。第二部分膀胱癌放射治疗中膀胱内在靶体积(ITV)的量化测定目的：应用kVCBCT与模拟定位CT所采集到膀胱癌放疗前和放疗后图像,量化确定膀胱ITV变化和变化的规律。材料和方法：选取2008年10月至2010年12月在我院行根治性放疗的膀胱癌患者12例。放射治疗前行模拟定位CT扫描。治疗期间,每例患者第一周每天治疗前、后各行一次kVCBCT扫描,自第二周起,每周选择一天在治疗前、后各行一次kVCBCT扫描,不进行在线引导体位校正。在MIM医学图像处理软件中分别勾画模拟定位CT和kVCBCT膀胱的轮廓,在相互配准后比较每次放疗前与模拟定位时膀胱体积变化；比较每次放疗后与放疗前的膀胱体积变化；比较从第二次治疗及随后每次治疗得到的膀胱合并部分体积与第一次治疗时膀胱体积相比的变化情况。结果：本部分研究共获取12副模拟定位CT图像,232副kVCBCT生成的CT。全部12例模拟定位CT所显示的膀胱大小平均值：104.68±41.32ml,KVCBCT所显示的每次放疗前和放疗后的膀胱体积平均值分别为：104.75±42.16ml和113.63±51.59ml。每次治疗前膀胱体积变化的标准差除以其均值(变异系数)在0.2以内的有7例,5例患者的体积发生了20%以上的改变。有6例其治疗前膀胱体积的平均值与该患者在模拟定位CT时膀胱体积存在显著变化(P值<0.05)。有6例患者其放疗前后的体积变化有显著性差异(P值<0.05),有10例患者放疗后的平均体积大于其治疗前的膀胱体积。考虑到每天治疗前后膀胱体积存在变化的可能,我们将每次治疗前后的膀胱体积合并到一起做为该次治疗时的膀胱体积,分别比较后面的治疗与第一次治疗,发现其中有8例患者在随后的治疗中其膀胱的变化情况与第一次治疗相比有显著性差异(P值<0.05)。结论：膀胱癌患者在放射治疗期间,膀胱的体积变化较大,如何精确确定膀胱ITV是重要的。第三部分膀胱ITV精确确定的方法学研究目的：将患者kVCBCT得到的膀胱轮廓叠加所形成的膀胱总体积作为膀胱ITV的金标准,来探讨依据放疗中多少天的KVCBCT所获得叠加膀胱体积能够个体化确定膀胱的ITV,并且评价基于模拟定位CT的膀胱均匀外扩和各方向不均匀外扩生成ITV与金标准的差异性。材料和方法：选取2008年10月至2010年12月在我院行根治性放疗的膀胱癌患者12例。放射治疗前行模拟定位CT扫描,治疗期间,每例患者在第一周的每天治疗前、后各行一次kVCBCT扫描,自第二周起,每周选择一天在治疗前、后各行一次kVCBCT扫描,不进行在线引导体位校正。根据勾画得到的每幅kVCBCT影像得到的12例患者的膀胱总体体积VCBCT-Total。依次将每一天(获取kVCBCT的治疗日)的kVCBCT图像叠加得到每次叠加的总轮廓VCBCT-Total-#。分别计算每一天得到的VCBCT-Total-#占VCBCT-Total的百分比。勾画全部模拟定位CT中的膀胱轮廓做为CTV, CTV勾画后在该软件中外扩10mm做为ITVB。分别记录每个患者的得到的VCBCT-Total和VITVB,适形指数CI、VCBCT-ITVB-EX (CBCT的ITV中位于均匀外扩轮廓外部分的体积)、V ITVB-CBCT-SF (VITVB内VCBCT-Total覆盖区域以外的部分)。勾画模拟定位CT膀胱轮廓做为CTV,依次在头脚方向脚侧(Inferior)外扩10mm,头侧(Superior)外扩20mm,左右方向上向左侧(Left)外扩11mm,右侧(Right)外扩8mm,前后方向上向前侧(Anterior)外扩20mm,向后侧(Posterior)外扩14mm得到VITVC分别记录每个患者的得到的VCBCT-Total和VITVC CI、VCBCT-ITVC-EX和VITVC-CBCT-SF。结果：全部12例患者的膀胱叠加总体积变化为：191.49±75.80.24m1(108.63-350.60m1)。12例患者在治疗第一周的D4、D5其V%CBCT-Total的均值均超过了90%,其中D5的百分比可信区间下限也接近90%,只有1例患者结果距离90%较远。均匀外扩10mm后的ITV体积变化范围：300.14±79.23ml(197.80-437.57m1)。其外扩10mmm后的ITV的适形指数的变化范围为：0.52±0.06(0.38～0.06), V ITVB-CBCT-SF的体积为126.63±36.77ml(72.31～194.96m1)。不均匀外扩后得到ITV的体积在412.30±102.82ml(272.77-566.36m1)之间,适形指数的变化范围为：0.44±0.08(0.29～0.57),V ITVC-CBCT-SF的体积为225.14±53.51ml(148.46-326.34m1)。与均匀外扩10mm生成膀胱ITV的计划B相比,采用不均匀外扩的膀胱ITV设计的计划C中,VEX其百分比明显减少,结果有显著性差异(P=0.00),而VSp占其PTV体积的百分比明显增加,结果有显著性差异(P=0.00)。结论：根据我们利用kVCBCT得到的膀胱体积和空间变化的资料可以发现,前五次的膀胱叠加体积用来做为膀胱癌个体化ITV勾画设计治疗方案是具有可行性的。单纯采用模拟定位CT的膀胱勾画CTV外扩10mm的膀胱的ITV有可能造成治疗期的膀胱的漏照,并存在较大范围正常组织的照射体积。CTV不均匀外扩的ITV对膀胱的覆盖较好,也存在较大范围的正常组织的照射体积。第四部分个体化确定膀胱ITV对膀胱癌放疗剂量学影响的研究目的：分别用个体化确定,均匀和不均匀外放所确定的膀胱ITV来设计满足各自肿瘤和正常组织剂量学要求的放疗计划A、B、C。比较三个计划中个体化确定的ITV以及其周边正常组织接受到的放疗剂量的差异性。材料和方法：选取2008年10月至2010年12月在我院行根治性放疗的膀胱癌患者12例。利用第三部分得到的三个不同ITV,分别外扩0.8mm得到相应的PTVA(个体化确定ITV后的PTV)、PTVB(均匀外放ITV得到的PTV)、PTVC(不均匀外放ITV得到的PTV),采用逆向束流调强技术分别进行上述三种PTVA、PTVB、PTVC和PTV2(肿瘤区域加量照射)的计划设计。所设计得到计划均能满足各自外放的PTV和正常组织剂量要求,然后将个体化确定的PTVA套入到三个计划中,观察在三个计划中95%PTVA和99%PTVA所接受的剂量,记录95%PTVA体积接受的最低剂量；99%PTVA体积接受的最低剂量,99%PTVA的剂量达不到处方剂量的95%被认为存在不可接受肿瘤内漏照；并观察各自计划中95%等剂量线(41.8Gy等剂量线)与PTVA的体积差值以及该体积内平均剂量(反映了正常组织多照射体积和剂量),分别命名为VA (4180-PTVA),. VB (4180-PTVA)和VC (4180-PTVA)的体积和收到的平均剂量。结果：三组计划均采用IMRT计划设计后满足肿瘤和正常组织剂量学的要求。将PTVA套入计划A中,95%PTVA内的最低剂量范围为44.17Gy±0.21Gy, 99%PTVA内的最低剂量为43.42Gy±0.58Gy,均符合剂量学要求。将PTVA套入计划B中95%PTVA的最低剂量范围为42.66Gy±4.33Gy,99%PTVA内的最低剂量范围为35.72Gy±9.93Gy (15.54～44.13Gy),有7例患者存在不可接受的漏照(99% PTVA接受的最低剂量低于41.8Gy)。将PTVA套入计划C中,有4例患者存在不可以接受的漏照射(99% PTVA接受的最低剂量低于41.8Gy)。与计划A中12例病人均能满足肿瘤最低剂量学要求相比,计划B中7例达不到95%最低剂量要求,其出现漏照的患者数与计划A相比有显著性差异(P=0.005),计划C中4例达不到95%最低剂量要求,结果无显著性差异(P=0.093)。各自计划中95%处方剂量等剂量线(41.8Gy等剂量线)与PTVA的体积差值以及该体积内平均剂量在计划A中为97.47±39.18ml,其平均剂量为42.99±0.32Gy；计划B中,VB (4180-PTVA)的平均体积为390.98±119.67ml,其平均剂量为44.41+0.29Gy;计划C中,VC (4180-APTV)的体积变化范围为471.57±124.93ml,其平均剂量为44.56±0.22Gy。将计划A中的VA (4180-PTVA)分别与计划B的VB (4180-PTVA)和计划C的VC (4180-PTVA)进行配对分析,其结果计划A与计划B相比,41.8Gy(处方44Gy的95%剂量)等剂量线轮廓内未包括PTVA的部分体积两者有显著性差异(P=0.000)；计划A与计划C相比,41.8Gy等剂量线内PTVA轮廓外的体积两者也有显著性差异(P=0.000)。对VA(4180-PTVA)分别于VB(4180-PTVA)和VC(4180-PTVA)的平均剂量进行配对分析,其结果计划B中VB(4180-PTVA)平均剂量高于计划A的VA(4180-PTVA),两者有显著性差异(P=0.000),；计划C中VC(4180-PTVA)所受平均剂量高于VA(4180-PTVA),两者也有显著性差异(P=0.000)。结论：无论是均匀外扩还是不均匀外放所生成ITV均存在肿瘤漏照射和正常组织过多被照射,这些提示膀胱ITV个体化确定是非常有价值的。更多还原

【Abstract】 IEstablishment of the optimal image alignment method for kilovoltage cone-beam CT (kVCBCT) and simulation CT in bladder cancerObjective:To compare the repeatability, accuracy and time cost in three different alignment methods used kilovoltage cone beam CT (kVCBCT) and simulation CT of bladder cancer in MIM software.Materials and Methods:From October 2008 to December 2010,12 patients with bladder cancer who received radiotherapy in our hospital were included in this study. Images of simulation CT before radiotherapy and those of kVCBCT before the first irradiation were aligned with the automatic registration method, respectively, including Rigid Assisted Alignment, Box-based Alignment and Contour-based Alignment in MIM software. All the images were aligned by the same physician at three different moments. The distances which the center coordinates of the two series CT images moved in the alignments in the LR, SI, and AP directions, and the time spent in each method were recorded. All the results were evaluated with a single-blind approach by another physician and a physicist independently.Results:108 data of the 12 patients were obtained in three different alignment methods at three different moments. The results showed the maximum motion distances were less than lmm with both the Contour-based Alignment and the Box-based Alignment, while the motion ranges were larger in the Rigid Assisted alignment. The Contour-based Alignment used the longest time,44.81 seconds on average. It’s indicated the Contour based alignment might be the optimal method for alignment in our study.Conclusion:In the MIM software, the Contour-based Alignment is the precise method, and the time cost is within an acceptable range. IIQuantitative determination of Internal target volume (ITV) of bladder during the irradiation of bladder cancerObjective:To observe the differences on the volume of bladder between the simulation CT and the kVCBCT scans, as well as each kVCBCT image, for the interfraction and intrafraction differences in bladder during irradiation of bladder cancer.Materials and Methods:From October 2008 to December 2010,12 patients of bladder cancer in our hospital with radical radiotherapy were selected. All the patients received simulation CT scan before radiotherapy. During the radiotherapy, each patient took a kVCBCT scan before and after irradiation every day at the first week, and repeated kVCBCT scans before and after irradiation only once a week at any day since the second week, without on-line kVCBCT position correction. All image data were sent to the MIM software, and the outline of the bladder was sketched. After alignment of the compared images, the bladder volumes were compared between simulation CT and each kVCBCT before irradiation,, and between the before and after treatment kVCBCT. In the meanwhile, the changes of overlapped bladder volume in the kVCBCT images scanned before and after the first and subsequent irradiation were observed.Results:With a total of 12 simulation CT images and 232 kVCBCT images, the average bladder volume in 12 simulation CT scans was 104.68±41.32ml,the change of bladder volume of 12 patients was 108.82±48.73ml(41.96-292.60ml)by kVCBCT images. In 7 patients the change range of bladder volumes before each irradiation waswithin 20% relative to its mean value (coefficient of variation), but 5 patients had a variance more than 20%. Average bladder volume before each irradiation changed significantly in comparison (P<0.05) with the volume of the simulation CT in 10 patients. Volumes of intrafraction treatment significantly different (P<0.05) in 6 patients, while volumes of interfraction treatment altered significantly (P<0.05)in 8 patients.Conclusion:bladder volume changes obviously during treatment, contouring an ITV according to the outline of the bladder is necessary. The accurate determination of the bladder ITV for bladder cancerObjective:Using the total volume of the bladder by superimposed contours of all kVCBCT as a combined index, to assess the the feasibility of contouring ITV outlines according to the volumes of bladder of the first five treatments as a adaptive individual bladder cancer therapy; To evaluate two methods of generating ITV the external isotropic expansion of bladder CTV contoured based a simulation CT, as well as anisotropy margin expansion.Materials and Methods:From October 2008 to December 2010,12 patients of bladder cancer received radiotherapy in our hospital were selected. All the patients received simulation CT scan before radiotherapy. During the radiotherapy, each patient took a kVCBCT scan before and after irradiation every day at the first week, and repeated kVCBCT scans before and after irradiation only once a week at any day since the second week, without on-line kVCBCT position correction. Overall volume of the bladder (VCBCT-Total) under the outline image of each kVCBCT of each patient was obtained. Overlap two new-received bladder volumes of kVCBCT at each kVCBCT treatment day, and accumulate the volume (VCBCT-Total-#) and calculate its percentage of VCBCT-Total followed by subsequent kVCBCT treatment day. Outlining all the bladders in the simulator CT scans as CTV, and expanding uniform a 10 mm margin as ITVB. VCBCT-Total, VITVB, conformal index (CI), VCBCT-ITVB-EX (the exceed part of CBCT’s ITV beyond contour of ITVB), VITVB-CBCT-SF (the superfluous part of VITVB within the contour of ITVB beyond the VCBCT-Total coverage area) of each patient were recorded,. Sketching the margin of bladders in the simulator CT scans as the CTV, CTV-to-ITV margins of 10 mm inferior,20 mm superior,11 mm left,8 mm right,20 mm anterior and 14 mm posterior were required to contour VITVC.VCBCT-Total and VITVC, CI, VCBCT-ITVC-EX and VITVC-CBCT-SF wereResults:The total volume of superimposed bladder of 12 patients was 191.49±75.80.24ml (108.63～350.60ml). The average of V% of its CBCT-Total’s volume in D4, D5 of the first week were more than 90%, the lower confidence interval of the percentage of D5 is close to 90%, and only 1 patient results far away from 90%. Homogeneous volume expansion ITV after 10mm ranged from 300.14±79.23ml (197.80～437.57ml), CI range:0.52±0.06 (0.38～0.06), VCBCT-ITVB-EX was ranged from 126.63±36.77ml (72.31～194.96ml). The volume of non-uniform outside expanded ITV was 412.30±102.82ml (272.77～566.36ml) between the CI ranges from:0.44±0.08 (0.29～0.57), and VCBCT-ITVC-EX:225.14±53.51ml (148.46 326.34ml). Compared with 10mm external uniform expansion (plan B), the non-uniform expansion of the bladder (plan C) had a percentage of VEX significantly reduced, and the results were significantly different (P= 0.00); the percentage of VSP accounts for PTV significantly increased, resulting in significant differences (P= 0.00).Conclusion:From the variation of bladder volume and spatial position in kVCBCT, it is feasible to establish the ITV of individual bladder cancer treatment with the total volume of the bladder by superimposed contours of the first five kVCBCT scans. Only using 10 mm expansion of the simulator CT to contour ITV may cause a bladder leakage area during the treatment period, and a big range of normal tissue irradiated volume. External non-uniform expansion of the ITV can cover the bladder area better, but a large range of normal tissue irradiated volume is also existed. The study about radiation dosimetry affected by individual identified bladder ITV in bladder cancerObjective:The radiation treatment plans, Plan A, Plan B and Plan C were designed separately by an adaptive individual bladder ITV, an uniform or non-uniform expansion bladder ITV. Radiation dose received by the bladder ITV and surrounding normal tissues in these three different plans were compared.Materials and Methods:From October 2008 to December 2010,12 patients with bladder cancer who received radiotherapy in our hospital were included in this study. 0.8mm outside expansion of three different ITV obtained from partⅢ, corresponding PTVA, PTVB and PTVC was established. All the plans were designed with reverse beam IMRT technique. PTV2 was defined from a tumor boost irradiation volume. Make sure all treatment plans were generated to prescription dose on PTVs and limiting doses on organs at risk according to standard, then copy PTVA to Plan B and C to observe the dose of 95% PTVA and 99% PTVA on the three plans. Dosage of 99% PTVA lower than 95% of prescription dose was considered as an existing of unacceptable radiation leakage. The lowest dose received on 95% and 99% PTVA volume was registered, the volume beyond PTVA but within the 41.8Gy isodose(95% isodose) was defined as VA(4180-PTVA),VB(4180-PTVA)and VC(4180-PTVA), and their average dosage reflected the extra irradiation volume and dose of normal tissues was recorded.Results:All the three plans were designed achieving the planned requirements. In Plan A, lowest dosage on 95% and 99% of PTVA were 44.17Gy±0.21Gy and 43.42Gy±0.58Gy, both were in line with required dose. Copy PTVA to Plan B,95% and 99% of PTVA received 42.66Gy±4.33Gy and 35.72Gy±9.93Gy, unacceptable irradiation leakage occurred in 7 patients (Dosage on 99% of PTVA was less than 41.8Gy). Copy PTVA to Plan C,95% and 99% of PTVA got a dosage of 44.21Gy±1.17Gy,41.06Gy±6.16Gy on average. There were 4 unacceptable irradiation leakages in plan C. The number of irradiation leakage had a significant difference between Plan A and Plan B (P=0.005), but was not statistically different between Plan A and Plan C (P=0.093). The volume of VA(4180-PTVA),、VB(4180-PTVA) and VC(4180-PTVA) was 97.47±39.18ml,390.98±119.67ml and 471.57±124.93ml (Plan A vs. Plan B, P=0.000; Plan A vs. Plan C,P=0.000), the average dosage of them was 42.99±0.32Gy, 44.41±0.29Gy and 44.56±0.22Gy (Plan A vs. Plan B, P=0.000; Plan A vs. Plan C, P=0.000). Conclusion:A radiation treatment plan based on an isotropic or an anisotropic expansion ITV exists unacceptable irradiation leakage and excessive irradiation of normal tissues surrounding the bladder, so determination of an individual bladder ITV in bladder cancer treatment planning is valuable.更多还原