目的 探索瓦里安Rapidplan优化模块在胸上段食管癌容积旋转调强计划(VMAT)中的可行性及潜在优势.方法 选取50例患者已执行放疗的胸上段食管癌VMAT计划,导入Rapidplan数据库,建立自动优化模型.另选10例胸上段食管癌VMAT计划对该模型进行独立验证.比较Rapidplan自动优化计划(RP)和传统人工计划(RG)的靶区和危及器官剂量学.结果 Rapidplan可预测剂量分布,在靶区、心脏和肺受量相当的情况下,降低脊髓的受量.RP计划靶区的均质性指数(HI)差于RG计划,适形指数(CI)优于RG计划(P＜0.05).RP计划的脊髓外扩结构的最大剂量优于RG计划(P＜0.05),总肺的V20、V5和RG计划差异无统计学意义(P＞0.05).RP计划优化效率优于RG计划(P＜0.05),执行效率与RG计划差异无统计学意义(P＞0.05).结论 Rapidplan自动优化模块能够很好地应用于胸上段食管癌VMAT计划.

Objective:To evaluate the usability of Gafchromic HD-V2 film for dose dosimetry in the ultra-high dose-rate (UD) electron beam from a modified medical linac, and to investigate the response between the energy and dose-rate dependence to the film.Methods:The HD-V2 film was utilized to measure the average dose-rate of the UD electron beam. The measured result was compared with those by advanced Markus chamber and alanine pellets. And characteristics of the UD electron beam were also measured by HD-V2 film. Energy dependence of HD-V2 film at three beam energies (6 MV X-ray, 9 MeV and 16 MeV electron beam) was investigated by obtaining and comparing the calibration curves based on the clinical linear accelerator in the dose range of 10-300 Gy. The dose-rate dependence of HD-V2 film was also studied by varying the dose rate among 0.03 Gy/s, 0.06 Gy/s and 0.1 Gy/s, and range of 100-200 Gy/s.Results:The measured average maximum dose-rate of 9 MeV UD electron beam at source skin distance (SSD) 100 cm was approximately 121 Gy/s using HD-V2 film, consistent with the results by advanced Markus chamber and alanine pellets. The measured percentage depth dose (PDD) curve parameters of the UD electron beam were similar to the conventional 9 MeV beam. The off-axis dose distribution of the UD electron beam showed the highest central axis, and the dose was gradually decreased with the increase of off-axis distance. The energy dependence of HD-V2 film had no dependency of 6 MV and 9, 16 MeV while measuring the dose in the range from 20 to 300 Gy. The HD-V2 film had no significant dose-rate dependency at the dose rate of 0.03 Gy/s, 0.06 Gy/s and 0.1 Gy/s for the clinical linear accelerator. Likewise, there was also no dose-rate dependence in the range 100-200 Gy/s in the modified machine.Conclusion:HD-V2 film is suitable for measuring ultra-high dose rate electron beam, independent of energy and dose rate.

Radiomics and dosiomics as two kinds of imaging features are widely used for machine learning-based prognosis prediction in adaptive radiotherapy. Feature selection and modeling are two main components in the radiotherapy prognosis prediction pipeline. So far, few studies have considered both the stability and discrimination ability of the features at the stage of feature selection. Also, in the modeling phase, to fuse radiomics and dosiomics features, most works have only directly concatenated radiomics or dosiomics features as inputs into a model, which may omit the complementary information across different omics. Additionally, overfitting is a common issue when the training data is not enough or contains noises. To solve these problems, in this study, we have developed a novel machine learning-based pipeline and applied it to predict radiation pneumonitis for stage III non-small cell lung cancer patients under medical Internet of Things. The contributions contain the following: in the feature selection phase, a decision criterion which considers both the feature stability and feature discrimination is developed to determine appropriate feature selection methods; in the modeling phase, we have developed a non-sparse multi-kernel learning method with manifold regularization for multi-omics fusion, which can fully explore patterns from both radiomics and dosiomics features and reduce overfitting coincidently. Experimental results show that the decision criterion works effectively for feature selection method selection. Compared to direct feature concatenation, the proposed multi-kernel fusion strategy performs better. Moreover, manifold regularization can alleviate the overfitting problem.

Purpose: This study aimed to evaluate the utility of a new plan feature (planomics feature) for predicting the results of patient-specific quality assurance using the head and neck (H&N) volumetric modulated arc therapy (VMAT) plan.Methods: One hundred and thirty-one H&N VMAT plans in our institution from 2019 to 2021 were retrospectively collected. Dosimetric verification for all plans was carried out using the portal dosimetry system integrated into the Eclipse treatment planning system based on the electronic portal imaging devices. Gamma passing rates (GPR) were analyzed using three gamma indices of 3%/3 mm, 3%/2 mm, and 2%/2 mm with a 10% dose threshold. Forty-eight conventional features affecting the dose delivery accuracy were used in the study, and 2,476 planomics features were extracted based on the radiotherapy plan file. Three prediction and classification models using conventional features (CF), planomics features (PF), and hybrid features (HF) combining two sets of features were constructed by the gradient boosting regressor (GBR) and Ridge classifier for each GPR of 3%/3 mm, 3%/2 mm, and 2%/2 mm, respectively. The absolute prediction error (APE) and the area under the curve (AUC) were adopted for assessing the performance of prediction and classification models.Results: In the GPR prediction, the average APE of the models using CF, PF, and HF was 1.3 ± 1.2%/3.6 ± 3.0%, 1.7 ± 1.5%/3.8 ± 3.5%, and 1.1 ± 1.0%/4.1 ± 3.1% for 2%/2 mm; 0.7 ± 0.6%/2.0 ± 2.0%, 1.0±1.1%/2.2 ± 1.8%, and 0.6 ± 0.6%/2.2 ± 1.9% for 3%/2 mm; and 0.4 ± 0.3%/1.2 ± 1.2%, 0.4±0.5%/1.3 ± 1.0%, and 0.3±0.3%/1.2 ± 1.1% for 3%/3 mm, respectively. In the regression prediction, three models give a similar modeling performance for predicting the GPR. The classification results were 0.67 ± 0.03/0.66 ± 0.07, 0.77 ± 0.03/0.73 ± 0.06, and 0.78 ± 0.02/0.75 ± 0.04 for 3%/3 mm, respectively. For 3%/2 mm, the AUCs of the training and testing cohorts were 0.64 ± 0.03/0.62 ± 0.07, 0.70 ± 0.03/0.67 ± 0.06, and 0.75 ± 0.03/0.71 ± 0.07, respectively, and for 2%/2 mm, the average AUCs of the training and testing cohorts were 0.72 ± 0.03/0.72 ± 0.06, 0.78 ± 0.04/0.73 ± 0.07, and 0.81 ± 0.03/0.75 ± 0.06, respectively. In the classification, the PF model has a better classification performance than the CF model. Moreover, the HF model provides the best result among the three classifications models.Conclusions: The planomics features can be used for predicting and classifying the GPR results and for improving the model performance after combining the conventional features for the GPR classification.

Objective:To build a systemic and automatic importing scheme for importing CT images and structures into the treatment planning systems (TPSs) of Eclipse and Monaco.Methods:Based on two TPSs of Eclipse and Monaco, the files of CT images and structures were automatically transported from OAR auto-delineation system to the importing directory of these two TPSs using batch script in Windows system. Following the standard importing procedures of these two TPSs, the automatically importing script of CT images and structures were developed using the application of UiBot. Finally, the CT images and structures were imported into these two TPSs opportunely.Results:By comparing the importing time using script and manual methods, the script not only achieved auto-importing CT images and structures into TPSs, but also yielded almost the same efficiency to manual method. The number of imaging layers in most patients was between 130 and 180, and the average manual and automatic importing time within this interval was 76 s and 75 s.Conclusions:Automatic scripts can be developed by using the automation function of UiBot combined with the actual problems of radiotherapy and repeated workflow. The efficiency of radiotherapy work can be significantly improved. Manual and time costs can be saved. It provides a novel alternative for the automation of radiotherapy procedures.

Objective:Based on the AAPM-TG218 report, the dose verification of intensity-modulated radiotherapy (IMRT) plans were classified to understand the current status, establish the process and determine the limits of dose verification in our hospital.Methods:Different combinations of tumor locations, accelerators, treatment planning systems and verification devices in our hospital were verified and compared to determine the tolerance limits and action limits of each combination. The measurement requirement was adopted according to the AAPM-TG218 report, and 80 cases were selected for each measurement. The measurement procedures were implemented based upon the AAPM-TG218 report and clinical experience of our hospital.Results:The clinical action limits of IMRT plans in our hospital could meet the recommended range of the AAPM-TG218 report, and the tolerance limits were slightly lower than the AAPM-TG218 report′s recommendation (93.94% for 3%/2 mm). The measurement of verification devices was related to the sensitivity. The tolerance limits measured by EPID were higher than ArcCHECK, especially when the dose/distance requirements were more stringent (94.12% and 92.03% for 3%/2 mm, P=0.074; 86.82% and 74.61% for 2%/2 mm, P=0.017). Conclusion:Through the AAPM-TG218 report, the work flow of IMRT dose verification and the limit range are established in our hospital, providing guidance for subsequent clinical dosimetric measurement.

Objective:To validate the accuracy of physical model of in-vivo 3D dose verification based on electronic portal imaging device (EPID) using the phantom and preliminarily analyze the clinical application.Methods:Two phantoms (uniform and non-uniform phantoms) were involved in this study. The system of in-vivo 3D dose verification based on EPID was employed to acquire the images of square fields (SF) and combined fields of intensity-modulated radiotherapy (CFIMRT). The physical model of different media was constructed using the system. The factor of γ passing rate under different dose/distance criteria was statistically compared. For clinical cases, the dose-volume histograms were adopted to analyze the dose distribution of target volume and organs at risk (OARs).Results:For the SF in the uniform phantom, the average γ passing rate (3%/3 mm) was (97.49±1.11)%, and (94.06±5.11)% for the SF in the non-uniform phantom ( P>0.05). No statistical significance was noted in IMRT using different delivery methods (all P>0.05). For clinical cases, the average γ passing rate (3%/2 mm) was (97.96±1.84)% in the pre-treatment dose verification, and (90.51±6.96)%(3%/3 mm) for the in-vivo 3D dose verification. For clinical cases, significant dose deviation was observed in OARs with small size and large volume changes. Conclusion:The in-vivo 3D dose verification model based on EPID can be effectively applied in inter-fraction dose verification, providing technical support for adaptive radiotherapy in clinical practice.

目的:对ArcCheck的剂量学特性进行测量,分析ArcCheck的剂量学特性曲线,探讨ArcCheck能否满足临床使用的要求.方法:分别测量ArcCheck的重复性、能量响应、角度响应和剂量率响应.结果:归一后ArcCheck的重复性为0.9906±0.0090.能量响应分析表明经过直线回归后的常数项为-0.2575,标准差为0.1215;回归系数为1.4643,标准差为5.3687E-4.角度响应的曲线符合探测器表面的模体材料厚度变化.剂量率的变化对探头的测量结果影响很小.结论:ArcCheck模体探测器的剂量学特性满足临床使用的要求,可以对容积弧形调强计划进行验证.

目的:建立一种算法,从低空间分辨力二维电离室矩阵测得的预埋有金球的穿过模体的剂量分布中读取金球投影位置坐标.方法:(1)利用Octavius 729二维电离室矩阵测得穿过埋有金球的模体的剂量分布;(2)采用一种插值算法将得到的低空间分辨力的剂量分布转换成高空间分辨力的剂量分布;(3)在得到的高空间分辨力的剂量分布上自编算法读取模体内金球投影位置坐标;(4)将得到的金球投影位置坐标与用高空间分辨力EPID图像得到的金球投影位置坐标进行比较.结果:在模体中共有3个金球,在机架角为0°条件下进行测量,计算得到3个金球的投影位置坐标,与金球实际投影位置坐标的最大偏差为2.5 mm,最小偏差为0.1mm.结论:该插值算法及所编软件可以用于低空间分辨力二维电离室矩阵读取预埋于体内金球的投影位置识别,且软件操作方便快捷,结果可靠.