基于EPID的长靶区计划验证测量方法研究

Measurement method for plan verification of long-range target radiotherapy plans based on EPID

目的探讨电子射野影像装置(EPID)板移动之后的数据结果是否存在误差;若存在误差,则进一步探索产生误差的方法,并且探究提出的长靶区计划验证测量方法是否可行。方法选择2021年9月至2022年11月在苏州大学附属第四医院放疗科接受放射治疗(简称放疗)的患者20例,其中男性7例,女性13例;年龄39~79岁,平均年龄57岁;直肠癌11例,宫颈癌9例;冠状位靶区长度17.96~31.86 cm,冠状位靶区平均长度为22.6 cm;靶区体积898.8~2139.2 cm^(3),平均靶区体积1299.5 cm^(3);11例采用容积旋转调强放疗计划,9例采用调强放疗计划;均使用6 MV射线,600 MU/min剂量率;均采用各向异性分析算法和射野剂量(PD)影像预测算法分别计算体剂量和预测图像。设计并开展移动EPID板位置的开放野试验与归零、不归零准直器角度的准直器试验,同步应用MatriXX对计划进行验证,以作为标准验证归零准直器角度对计划通过率的影响,对数据及计划Gamma通过率进行分析。结果研究结果表明,EPID板移位后数据存在较大误差,其修正过程与预测的Profile并不完全一致,但是在曲线趋势上具有一定程度的相似性,并且在关键数据曲线交点上具有高度一致性。针对准直器归零与不归零,执行PD计划验证时二者Gamma通过率具有强相关性(Corr=0.994,P<0.05),差异无统计学意义(t=-1.63,P=0.119)。同时,Bland-Altman一致性分析结果表明无论准直器是否归零,PD计划验证与MatriXX计划验证的计划通过率均具有良好的一致性。结论长靶区计划应用EPID执行计划验证测量时应确保EPID板处于校准零位,若执行计划验证时EPID板移动,得到的结果存在较大误差。在确定执行该计划验证并不会辐照损坏EPID的电路板时,可固定EPID板位置;存在辐照损坏EPID的电路板风险时,可归零准直器角度以执行PD计划验证,所得到的计划通过率可以近似替代不归零准直器角度得到的计划通过率结果。

Objective To explore whether there is error in data results after the movement of electronic portal imaging device(EPID)plate,and study the factors that cause errors,and explore the feasibility of the long-range target plan verification measurement method.Methods From September 2021 to November 2022,20 patients received radiotherapy were enrolled,which included 7 males and 13 females,aged 39-79 years old with mean age of 57 years old.There were 11 cases of rectal cancer and 9 of cervical cancer.The length of coronal target was 17.96-31.86 cm,with mean length of coronal target of 22.6 cm.The target volume was 898.8-2139.2 cm^(3),with mean target volume of 1299.5 cm^(3).Eleven cases were treated with volumetric modulated arc therapy,and 9 were treated with intensity modulated radiotherapy.All of them were treated with 6 MV radiations at dose rate of 600 MU/min.The anisotropic analysis algorithm and portal dosimetry(PD)image prediction algorithm were used to calculate volume dose and predicted image,respectively.The open field test of movement EPID plate and collimator test of zeroing and non-zeroing collimator angles were designed and carried out,and MatriXX was used to verify the plan,which as the standard to verify the influence of zeroing collimator angle on planned pass rate,and the data and planned Gamma pass rate were analyzed.Results The results showed that there was a significant error in data after EPID plate shifted,and correction process was incompletely consistent with predicted Profile,but there was a certain degree of similarity in curve trend,and with high degree of consistency at intersection of key data curve.There was a strong correlation(Corr=0.994,P<0.05)between Gamma passing rate of PD plan verification,and the difference was no statistically significant(t=-1.63,P=0.119)for collimator zeroing and non-zeroing.At the same time,the Bland-Altman consistency analysis results showed that the plan pass rates of PD plan verification and MatriXX plan verification were in good agreement regardless of whether the collimator was zeroed or not.Conclusion It is demonstrated that the application of long-range target plan for EPID verification measurement should ensure that EPID plate is at the calibration zero position.Significant errors may occur in the results obtained if the EPID board is moved during the execution of the plan verification.In determining the verification plan will not irradiate and damage the EPID circuit board,the EPID plate position can be fixed.If there is a risk of radiation damage to EPID circuit board,the collimator angle should be zeroed to perform PD plan verification,and the obtained plan pass rate can approximately replace that of non-zeroing collimator angle plan.