[1]张丽媛,贾楠,郑晓娜,等.基于人工神经网络多叶准直器叶片位置偏差预测[J].中国医学物理学杂志,2021,38(12):1495-1501.[doi:DOI:10.3969/j.issn.1005-202X.2021.12.008]
　ZHANG Liyuan,JIA Nan,ZHENG Xiaona,et al.Prediction of multi-leaf collimator leaf positional deviations based on artificial neural network[J].Chinese Journal of Medical Physics,2021,38(12):1495-1501.[doi:DOI:10.3969/j.issn.1005-202X.2021.12.008]
点击复制

基于人工神经网络多叶准直器叶片位置偏差预测()
分享到：

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
38卷
期数:
2021年第12期
页码:
1495-1501
栏目:
医学放射物理
出版日期:
2021-12-24

文章信息/Info

Title:
Prediction of multi-leaf collimator leaf positional deviations based on artificial neural network
文章编号:
1005-202X（2021）12-1495-07
作者:
张丽媛; 贾楠; 郑晓娜; 张博; 李松丽; 韩全乡
郑州市中心医院放射治疗科， 河南 郑州 450052
Author(s):
ZHANG Liyuan;  JIA Nan;  ZHENG Xiaona;  ZHANG Bo;  LI Songli;  HAN Quanxiang
Keywords: dynamic intensity-modulated radiotherapy multi-leaf collimator log file artificial neural network positional deviation
关键词:
动态调强放疗; 多叶准直器; 日志文件; 人工神经网络; 位置偏差
Keywords:
Department of Radiation Oncology;  Zhengzhou Central Hospital;  Zhengzhou 450052;  China
分类号:
R312；R811.1
DOI:
DOI:10.3969/j.issn.1005-202X.2021.12.008
文献标志码:
A
摘要:
目的：探讨人工神经网络方法利用加速器日志文件预测动态调强放疗（dIMRT）计划多叶光栅位置偏差。方法：选择瓦里安Trilogy 50例dIMRT动态日志文件，TrueBeam和Edge各30例dIMRT轨迹日志文件。日志提取叶片计划位置、剂量、机架角度、铅门位置、叶片间距、叶片速度及运动参数等14个特征为输入，日志记录叶片实际位置为输出，隐藏层为22个神经元的3层神经网络模型。训练集包括70%数据，验证集和测试集包括30%数据，均方误差代价函数评估模型性能。结果：叶片速度是输入特征与位置偏差最相关特征，皮尔逊相关系数0.7以上。运动和静息叶片位置平均绝对偏差存在显著性差别。测试集Trilogy、TrueBeam和EDGE预测叶片位置最大均方误差分别小于9×10-5、3×10-5和3×10-5 mm2，预测位置与实际位置极为接近。模型预测当前叶片和其它序号叶片位置平均绝对偏差具有显著性差异（P<0.001）。结论：基于加速器日志文件人工神经网络模型可以对瓦里安dIMRT计划叶片位置进行预测。
Abstract:
Abstract: Objective To predict the multi-leaf collimator (MLC) leaf positional deviations during dynamic intensity-modulated radiotherapy (dIMRT) by log file analysis and artificial neural network (ANN) method. Methods The dIMRT DynaLog files from Varian Trilogy (50 cases) and the dIMRT Trajectory log files from TrueBeam and Edge (30 cases of each) were retrieved. A 3-layer ANN model with hidden layer of 22 neurons was developed for each MLC leaf, with 14 features extracted from the log files, such as leaf planned positions, dose, gantry angle, jaw position, leaf gap, leaf velocity and leaf motion status, as input parameters, and the delivered leaf position recorded by log files as output. The proposed model was trained on 70%, validated and tested on 30% of the total data. Mean square error (MSE) was taken as the cost function to evaluate the model performance. Results The leaf velocity was the most relevant input feature to positional deviation, with a Pearson correlation coefficient greater than 0.7. Significant differences were found in the mean absolute error (MAE) between moving and resting leaf. The maximum MSE in predicting the leaf positions on test set were less than 9×10-5, 3×10-5 and 3×10-5 mm2 for Trilogy, TrueBeam and EDGE, respectively and the predicted leaf positions closely matched the actual positions during the treatment delivery. There was significant difference in MAE of the model in predicting the positions of current leaf and other leaves (P<0.001). Conclusion The proposed log file-based ANN model is capable of predicting the Varian MLC leaf position during dIMRT.

相似文献/References:

[1]陈炳周,路俊哲,侯 氢.不同设计方式的调强逆向治疗计划比较[J].中国医学物理学杂志,2015,32(03):391.[doi:10.3969/j.issn.1005-202X.2015.03.020]
[2]王矫健,徐利明,全红,等.多叶准直器叶片到位偏差对调强放射治疗剂量的影响[J].中国医学物理学杂志,2013,30(03):4121.[doi:10.3969/j.issn.1005-202X.2013.03.009]
[3]张富利,刘清智,王雅棣,等.基于配置不同类型MLC叶片加速器的宫颈癌术后调强放疗计划评估[J].中国医学物理学杂志,2013,30(05):4383.[doi:10.3969/j.issn.1005-202X.2013.05.010]
[4]石锦平,石俊田,何宝贞,等.容积调强放射治疗多叶准直器质量保证方法[J].中国医学物理学杂志,2015,32(05):724.[doi:doi:10.3969/j.issn.1005-202X.2015.05.025]
　[J].Chinese Journal of Medical Physics,2015,32(12):724.[doi:doi:10.3969/j.issn.1005-202X.2015.05.025]
[5]张富利,蒋华勇,许卫东,等.基于不同多叶准直器的胸膜间皮瘤容积旋转调强放疗剂量学比较[J].中国医学物理学杂志,2015,32(05):635.[doi:doi:10.3969/j.issn.1005-202X.2015.05.005]
[6]张富利,许卫东,蒋华勇,等.不同类型多叶准直器对左侧乳腺癌保乳术后容积旋转调强放疗技术剂量分布的影响[J].中国医学物理学杂志,2017,34(7):726.[doi:10.3969/j.issn.1005-202X.2017.07.015]
　[J].Chinese Journal of Medical Physics,2017,34(12):726.[doi:10.3969/j.issn.1005-202X.2017.07.015]
[7]王锐濠,张书旭,田允鸿,等.胸段食管癌不同调强放疗技术的剂量学差异分析[J].中国医学物理学杂志,2017,34(8):766.[doi:DOI:10.3969/j.issn.1005-202X.2017.08.003]
[8]叶淑敏,滕建建,石锦平,等. MLC叶片系统误差对鼻咽癌VMAT和IMRT计划剂量影响的比较[J].中国医学物理学杂志,2019,36(10):1139.[doi:DOI:10.3969/j.issn.1005-202X.2019.10.005]
　YE Shumin,TENG Jianjian,SHI Jinping,et al. Comparison of dosimetric impacts of MLC systematic errors on VMAT and IMRT plans for nasopharyngeal carcinoma[J].Chinese Journal of Medical Physics,2019,36(12):1139.[doi:DOI:10.3969/j.issn.1005-202X.2019.10.005]
[9]马天斌,蒋振东,俞海东,等.两种准直器宽度对鼻咽癌放疗中的剂量学差异影响[J].中国医学物理学杂志,2021,38(3):302.[doi:DOI:10.3969/j.issn.1005-202X.2021.03.006]
　MA Tianbin,JIANG Zhendong,YU Haidong,et al.Multileaf collimators of two different leaf widths in radiotherapy of nasopharyngeal carcinoma: a dosimetric comparison[J].Chinese Journal of Medical Physics,2021,38(12):302.[doi:DOI:10.3969/j.issn.1005-202X.2021.03.006]

备注/Memo

备注/Memo:
【收稿日期】2021-06-13 【基金项目】河南省高等学校重点科研项目（21A320066） 【作者简介】张丽媛，硕士，主管技师，研究方向：医学物理，E-mail: liyuannana0306@163.com 【通信作者】韩全乡，副主任医师，研究方向：肿瘤放射治疗，E-mail: hanquanxiang1@163.com
更新日期/Last Update: 2021-12-24