Error detection and classification in patient-specific IMRT QA with dual neural networks
Error detection and classification in patient-specific IMRT QA with dual neural networks
复制标题
DOI：
10.1002/mp.14416
复制
发表时间：
2020-08-13
期刊：
影响因子：
3.8
通讯作者：
Yan, Guanghua
中科院分区：
文献类型：
--
作者：
Potter, Nicholas J.;Mund, Karl;Yan, Guanghua
尽管γ分析是调强放射治疗（IMRT）患者特异性质量保证（QA）的标准度量，但它有两个缺点：（a）它对小但临床相关的误差缺乏灵敏度（B）它没有提供有效的方法来分类误差源。本文提出了一种双神经网络方法，用于实现患者特定调强放射治疗QA中的同步错误检测和分类。方法对于一对剂量分布，提取低剂量梯度区的剂量差直方图（DDH）和高剂量梯度区的两个有符号一致性距离（sDTA）图（一个在x方向，一个在y方向）。分别设计了人工神经网络（ANN）和卷积神经网络（CNN）来分析DDH和两个sDTA图。人工神经网络进行了训练，以检测和分类六类剂量测定误差：不正确的多叶准直器（MLC）传输（+/- 1%）和四种类型的监视器单元（MU）缩放误差（+/- 1%和+/- 2%）。CNN被训练来检测和分类七类空间误差：不正确的有效源尺寸、1 mm MLC叶组超程或欠程、2 mm单个MLC叶超程或欠程以及设备未对准误差（在x或y方向上1 mm）。使用内部平面剂量计算软件模拟引入误差和噪声的测量。两个网络都用13个调强放射治疗计划（共88个野）进行了训练和验证。一个五重交叉验证技术被用来评估其准确性。结果DDH和sDTA图谱具有明显的特征性。神经网络完美地识别了所有四种类型的MU缩放误差，并且对于无误差、MLC透射率增加、MLC透射率减少的类别的特定准确度分别为98.9%、96.6%和94.3%。对于CNN，最大的混淆发生在1 mm MLC组超程类和x方向类中的1 mm器件对准误差之间，这使得具体准确度分别下降到90.9%和92.0%。2 mm单个MLC叶片下行程类别的特定准确度为93.2%，因为它将5.7%的类别错误分类为无错误（假阴性）。特异性准确率在95%以上。ANN和CNN的五倍总体准确率分别为98.3 +/- 0.7%和95.6% +/- 1.5%。结论DDH图和sDTA图均适用于IMRT QA中的错误分类。提出的双神经网络方法实现了同时错误检测和分类具有良好的精度。它可以与伽马分析一起使用，以潜在地将IMRT QA范式从被动通过/失败分析转变为主动错误检测和根本原因识别。
Purpose Despite being the standard metric in patient-specific quality assurance (QA) for intensity-modulated radiotherapy (IMRT), gamma analysis has two shortcomings: (a) it lacks sensitivity to small but clinically relevant errors (b) it does not provide efficient means to classify the error sources. The purpose of this work is to propose a dual neural network method to achieve simultaneous error detection and classification in patient-specific IMRT QA. Methods For a pair of dose distributions, we extracted the dose difference histogram (DDH) for the low dose gradient region and two signed distance-to-agreement (sDTA) maps (one in x direction and one in y direction) for the high dose gradient region. An artificial neural network (ANN) and a convolutional neural network (CNN) were designed to analyze the DDH and the two sDTA maps, respectively. The ANN was trained to detect and classify six classes of dosimetric errors: incorrect multileaf collimator (MLC) transmission (+/- 1%) and four types of monitor unit (MU) scaling errors (+/- 1% and +/- 2%). The CNN was trained to detect and classify seven classes of spatial errors: incorrect effective source size, 1 mm MLC leaf bank overtravel or undertravel, 2 mm single MLC leaf overtravel or undertravel, and device misalignment errors (1 mm in x- or y direction). An in-house planar dose calculation software was used to simulate measurements with errors and noise introduced. Both networks were trained and validated with 13 IMRT plans (totaling 88 fields). A fivefold cross-validation technique was used to evaluate their accuracy. Results Distinct features were found in the DDH and the sDTA maps. The ANN perfectly identified all four types of MU scaling errors and the specific accuracies for the classes of no error, MLC transmission increase, MLC transmission decrease were 98.9%, 96.6%, and 94.3%, respectively. For the CNN, the largest confusion occurred between the 1-mm-MLC bank overtravel class and the 1-mm-device alignment error in x-direction class, which brought the specific accuracies down to 90.9% and 92.0%, respectively. The specific accuracy for the 2-mm-single MLC leaf undertravel class was 93.2% as it misclassified 5.7% of the class as being error free (false negative). Otherwise, the specific accuracy was above 95%. The overall accuracies across the fivefold were 98.3 +/- 0.7% and 95.6% +/- 1.5% for the ANN and the CNN, respectively. Conclusions Both the DDH and the sDTA maps are suitable features for error classification in IMRT QA. The proposed dual neural network method achieved simultaneous error detection and classification with excellent accuracy. It could be used in complement with the gamma analysis to potentially shift the IMRT QA paradigm from passive pass/fail analysis to active error detection and root cause identification.