4DCT与T_(2)导航项MRI融合定位对腹部肿瘤放疗的初步观察

Preliminary observation of 4DCT and T_(2) navigation MRI fusion localization in radiotherapy of abdominal tumor

目的 提出一种基于MRI导航序列的腹部肿瘤立体定向放疗(SBRT)新策略,验证分析图像配准误差和剂量投送精度以及流程的可行性和准确性。方法 回顾性纳入2021-11-01-2022-02-28山东省肿瘤医院行MRI引导SBRT治疗的肝癌患者12例。按入组时间平均分为2组。模体选用6种运动模式,分试验组和对照组。试验组采用基于导航项的MRI定位序列进行定位,应用导航项与四维CT(4DCT)各相位配准获取最一致相位CT进行计划制定。对照组应用常规T加权序列进行定位,采用自由呼吸状态三维CT(3DCT)进行计划制定。均行在线计划流程。结果 模体摆位数据显示,基于自由呼吸状态的配准误差相对于基于导航项的配准误差,X方向(中位数值±四分位距)分别为(0.285±0.080)、(0.180±0.070) cm,Z=2.562,P=0.010;Y方向分别为(0.545±0.110)、(0.245±0.130) cm,Z=2.882,P=0.004;Z方向分别为(0.145±0.030)、(0.100±0.040) cm,Z=2.402,P=0.016。实际治疗患者的配准误差,试验组中位数值相较于对照组更小,且Y方向差异有统计学意义;X方向(中位数值±四分位距)分别为(0.580±0.160)、(0.310±0.160) cm,Z=2.242,P=0.025;Y方向分别为(1.350±1.100)、(0.310±0.160) cm,Z=2.882,P=0.004;Z方向分别为(0.270±0.060)、(0.190±0.070) cm,Z=2.661,P=0.008。比较2组模体实验实际测量剂量与计划系统计算剂量差异无统计学意义,P=0.631。结论 针对不同运动模式,基于MR导航项的图像引导放疗有较好的图像稳定性和较小的配准误差,在临床上具有高可行性。

Objective To propose a new strategy of stereotactic radiotherapy(SBRT) for abdominal tumors based on MRI navigation sequence, and to verify and analyze the image registration error, dose delivery accuracy and the feasibility and accuracy of the proposed process.Methods A total of 12 patients with liver cancer who were treated with MRI guided SBRT in Shandong Cancer Hospital from November 1,2021 to February 28,2022 were enrolled retrospectively.According to the average time of joining the group, they were divided into two groups.Six kinds of exercise modes were selected and divided into experimental group and control group.The experimental group used MRI positioning sequence based on navigation items for positioning,and applied navigation items and four-dimensional CT(4DCT)phase registration to obtain the most consistent phase CT for planning.In the control group,routine T_(2)-weighted sequence was used for localization,and free breathing three-dimensional CT(3DCT)was used for planning.Both groups went through the online planning workflow.Results:The positioning data of the phantom showed that the registration error based on the free breathing state was(0.285±0.080)cm in the X direction(median value±quartile),compared with the registration error based on navigation item(0.180±0.070)(Z=2.562,P=0.010);The registration errors in the Y direction were(0.545±0.110),and(0.245±0.130)cm,respectively(Z=2.882,P=0.004).The registration errors in the Z direction were(0.145±0.030)and(0.100±0.040)cm,respectively(Z=2.402,P=0.016).As for registration error of actual treated patients,compared with the control group,the median registration error of the experimental group was smaller than that of the control group,and the difference in Y direction was statistically significant;the registration errors in the X direction(median value±quartile)were(0.580±0.160)and(0.310±0.160)cm(Z=2.242,P=0.025);those in the Y direction were(1.350±1.100)and(0.310±0.160)cm,respectively(Z=2.882,P=0.004).Those in the Z direction were(0.270±0.060)and(0.190±0.070)cm,respectively(Z=2.661,P=0.008).There was no significant difference between the actual dose measured by the phantom experiment and the dose calculated by the planning system between the two groups.Conclusion According to different motion modes,the image guided radiotherapy based on MRI navigation item has better image stability and smaller registration error,so it is highly feasible in clinic.