Analysis of EPID Transmission Fluence Maps Using Machine Learning Models and CNN for Identifying Position Errors in the Treatment of GO Patients.
Analysis of EPID Transmission Fluence Maps Using Machine Learning Models and CNN for Identifying Position Errors in the Treatment of GO Patients.
复制标题
DOI：
10.3389/fonc.2021.721591
复制
发表时间：
2021
影响因子：
4.7
通讯作者：
Bai S
中科院分区：
文献类型：
--
作者：
Dai G;Zhang X;Liu W;Li Z;Wang G;Liu Y;Xiao Q;Duan L;Li J;Song X;Li G;Bai S
目的寻找一种合适的方法来分析电子射野成像设备（EPID）透射注量图，以识别Graves眼病（GO）患者体内剂量监测中的位置误差。模拟了40个GO患者放疗计划输送到人体头部体模时左右（LR）、前后（AP）和上下（SI）方向上的位置误差，包括0 mm、2 mm和4 mm误差，并采集了EPID传输注量图。计算剂量差异（DD）和结构相似性（SSIM）图，以量化注量图的变化。利用DD图的放射组学特征的三种类型的机器学习（ML）模型（ML 1型号），SSIM地图的功能（ML 2型号），以及DD和SSIM映射的功能（ML 3模型）作为输入用于执行三种类型的位置误差分类，即等中心误差的二进制分类（类型1），LR，SI，和AP方向误差（类型2），以及组合的LR、SI和AP方向误差的八元素分类（类型3）。卷积神经网络（CNN）也被用来分类使用DD和SSIM地图作为输入的位置误差。性能最好的ML 1模型是XGBoost，其在1型、2型-LR、2型-AP、2型-SI和3型分类中分别实现了0.889、0.755、0.778、0.833和0.532的准确度。最好的ML 2模型是XGBoost，其精度分别为0.856、0.731、0.736、0.949和0.491。最好的ML 3模型是线性判别分类器（LDC），其准确率分别为0.903，0.792，0.870，0.931和0.671。CNN的分类精度分别为0.925、0.833、0.875、0.949和0.689。使用组合DD和SSIM图的ML模型和CNN可以分析EPID传输注量图，以识别GO患者治疗中的位置错误。需要进一步的大样本研究来提高CNN的准确性。
To find a suitable method for analyzing electronic portal imaging device (EPID) transmission fluence maps for the identification of position errors in the in vivo dose monitoring of patients with Graves’ ophthalmopathy (GO). Position errors combining 0-, 2-, and 4-mm errors in the left-right (LR), anterior-posterior (AP), and superior-inferior (SI) directions in the delivery of 40 GO patient radiotherapy plans to a human head phantom were simulated and EPID transmission fluence maps were acquired. Dose difference (DD) and structural similarity (SSIM) maps were calculated to quantify changes in the fluence maps. Three types of machine learning (ML) models that utilize radiomics features of the DD maps (ML 1 models), features of the SSIM maps (ML 2 models), and features of both DD and SSIM maps (ML 3 models) as inputs were used to perform three types of position error classification, namely a binary classification of the isocenter error (type 1), three binary classifications of LR, SI, and AP direction errors (type 2), and an eight-element classification of the combined LR, SI, and AP direction errors (type 3). Convolutional neural network (CNN) was also used to classify position errors using the DD and SSIM maps as input. The best-performing ML 1 model was XGBoost, which achieved accuracies of 0.889, 0.755, 0.778, 0.833, and 0.532 in the type 1, type 2-LR, type 2-AP, type 2-SI, and type 3 classification, respectively. The best ML 2 model was XGBoost, which achieved accuracies of 0.856, 0.731, 0.736, 0.949, and 0.491, respectively. The best ML 3 model was linear discriminant classifier (LDC), which achieved accuracies of 0.903, 0.792, 0.870, 0.931, and 0.671, respectively. The CNN achieved classification accuracies of 0.925, 0.833, 0.875, 0.949, and 0.689, respectively. ML models and CNN using combined DD and SSIM maps can analyze EPID transmission fluence maps to identify position errors in the treatment of GO patients. Further studies with large sample sizes are needed to improve the accuracy of CNN.
登录
查看更多内容
影响因子：
3.5
作者：
Li WT;Ma J;Shende N;Castaneda G;Chakladar J;Tsai JC;Apostol L;Honda CO;Xu J;Wong LM;Zhang T;Lee A;Gnanasekar A;Honda TK;Kuo SZ;Yu MA;Chang EY;Rajasekaran MR;Ongkeko WM
通讯作者：
Ongkeko WM
影响因子：
3.8
作者：
Peng, Jiayuan;Shi, Chengyu;Cai, Bin
通讯作者：
Cai, Bin
影响因子：
3.7
作者：
Li, Yinghui;Zhu, Jinhan;Liu, Xiaowei
通讯作者：
Liu, Xiaowei
影响因子：
2.6
作者：
Park, J. M.;Kim, K.;Ha, S. W.
通讯作者：
Ha, S. W.
影响因子：
3.8
作者：
Potter, Nicholas J.;Mund, Karl;Yan, Guanghua
通讯作者：
Yan, Guanghua