[1]陈辛元,易俊林,戴建荣.卷积神经网络放疗计划剂量预测:两种解码器对比[J].中国医学物理学杂志,2020,37(2):153-158.[doi:DOI:10.3969/j.issn.1005-202X.2020.02.004]
 CHEN Xinyuan,YI Junlin,DAI Jianrong.Radiotherapy planning dose prediction using convolutional neural network: a comparison of two kinds of decoders[J].Chinese Journal of Medical Physics,2020,37(2):153-158.[doi:DOI:10.3969/j.issn.1005-202X.2020.02.004]
点击复制

卷积神经网络放疗计划剂量预测:两种解码器对比()
分享到:

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

卷:
37
期数:
2020年第2期
页码:
153-158
栏目:
医学放射物理
出版日期:
2020-03-03

文章信息/Info

Title:
Radiotherapy planning dose prediction using convolutional neural network: a comparison of two kinds of decoders
文章编号:
1005-202X(2020)02-0153-06
作者:
陈辛元易俊林戴建荣
国家癌症中心/国家肿瘤临床医学研究中心/中国医学科学院北京协和医学院肿瘤医院, 北京 100021
Author(s):
CHEN Xinyuan YI Junlin DAI Jianrong
National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, China
关键词:
卷积神经网络解码器放射治疗剂量预测
Keywords:
Keywords: convolutional neural network decoder radiotherapy dose prediction
分类号:
R318;R815
DOI:
DOI:10.3969/j.issn.1005-202X.2020.02.004
文献标志码:
A
摘要:
目的:拟建立卷积神经网络放疗计划三维剂量分布预测模型,比较种两解码器对网络性能的影响。方法:实验数据包括80例早期鼻咽癌的放疗计划,并随机分为训练集(70例)和测试集(10例)。基于VGG16卷积神经网络,分别采用插值和反卷积两种解码器,建立两种网络结构:插值解码VGG16网络(IVGG16)和反卷积解码VGG16网络(DVGG16),用于端到端的放疗计划剂量预测。评价模型准确性指标主要包括外轮廓、靶区及危及器官(OAR)的平均绝对误差(MAE),并分别记录两个网络的训练时间和预测时间。结果:使用两种解码器均可以较准确地预测患者三维剂量分布。IVGG16和DVGG16的外轮廓MAE分别为(5.48±0.46)%和(5.42±0.34)%,差别无统计学意义;靶区的预测值均较准确,MAE均低于2.63%,两种解码器没有统计学意义;两个网络均可以较准确预测OAR的剂量分布。脊髓PRV和甲状腺的MAE,DVGG16较IVGG16分别下降11.8%和15.6%(P=0.029, 0.034),其余OAR的MAE差别无统计学意义。IVGG16和DVGG16模型训练时间分别为14.8 h和24.6 h,每例平均预测时间分别为(6.6±1.0) s和(28.7±3.9) s。结论:采用插值和反卷积两种解码器预测得到的放疗剂量在整体上具有相当的效果。反卷积解码器对部分OAR剂量分布预测准确性略优,但模型训练和预测的效率有待提高。
Abstract:
Abstract: Objective To establish a three-dimensional radiotherapy planning dose distribution prediction model using convolutional neural network, and to compare the effects of two kinds of decoders on network performance. Methods The radiotherapy plans of 80 cases of early nasopharyngeal carcinoma were included in the study, and then were randomly divided into a training set (70 cases) and a test set (10 cases). Based on VGG16 convolutional neural network, interpolation VGG16 network (IVGG16) and deconvolution VGG16 network (DVGG16) were established by two kinds of decodes, namely interpolation and deconvolution. The established networks were used for end-to-end radiotherapy planning dose prediction. The main indicators for the evaluation of model accuracy included the mean absolute errors (MAE) of the outline, tumor target areas and organs-at-risk (OAR). Moreover, the training time and prediction time of two kinds of networks were also recorded. Results The three-dimensional dose distribution could be predicted accurately using both two kinds of decoders. The MAE of the outline obtained by IVGG16 and DVGG16 were (5.48±0.46) % and (5.42±0.34)%, respectively, without statistical significance. The doses of the target areas were accurately predicted by two kinds of decodes, with a MAE lower than 2.63%, and there was no statistical difference between two kinds of decodes. The dose distributions of OAR were accurately predicted by two kinds of networks. However, compared with those obtained by IVGG16, the MAE of spinal cord PRV and thyroid obtained by DVGG16 was decreased by 11.8% and 15.6%, respectively (P=0.029, 0.034), and there was no statistical differences in the MAE of the remaining OAR. The training time of IVGG16 and DVGG16 models were 14.8 h and 24.6 h, respectively, and the average prediction time for each case were (6.6±1.0) s and (28.7±3.9) s, respectively. Conclusion The radiotherapy dose prediction using interpolation or deconvolutional decoders has a satisfactory effect. The deconvolutional decoder has a slightly higher prediction accuracy for some OAR dose distribution, but the efficiency of model training and prediction need to be further improved.

相似文献/References:

[1]陶源,王佳飞,杜俊龙,等.基于卷积神经网络的细胞识别[J].中国医学物理学杂志,2017,34(1):53.[doi:10.3969/j.issn.1005-202X.2017.01.011]
 [J].Chinese Journal of Medical Physics,2017,34(2):53.[doi:10.3969/j.issn.1005-202X.2017.01.011]
[2]刘岩,李幼军,陈萌. 基于固有模态分解和深度学习的抑郁症脑电信号分类分析[J].中国医学物理学杂志,2017,34(9):963.[doi:DOI:10.3969/j.issn.1005-202X.2017.09.021]
 [J].Chinese Journal of Medical Physics,2017,34(2):963.[doi:DOI:10.3969/j.issn.1005-202X.2017.09.021]
[3]张俊,朱金汉,庄永东,等. 基于卷积神经网络CT/CBCT影像质量自动分析[J].中国医学物理学杂志,2018,35(5):557.[doi:DOI:10.3969/j.issn.1005-202X.2018.05.012]
 ZHANG Jun,ZHU Jinhan,ZHUANG Yongdong,et al. Automatic analysis of CT/CBCT image quality based on convolutional neural network[J].Chinese Journal of Medical Physics,2018,35(2):557.[doi:DOI:10.3969/j.issn.1005-202X.2018.05.012]
[4]邓金城,彭应林,刘常春,等. 深度卷积神经网络在放射治疗计划图像分割中的应用[J].中国医学物理学杂志,2018,35(6):621.[doi:DOI:10.3969/j.issn.1005-202X.2018.06.001]
 DENG Jincheng,PENG Yinglin,LIU Changchun,et al. Application of deep convolution neural network in radiotherapy planning image segmentation[J].Chinese Journal of Medical Physics,2018,35(2):621.[doi:DOI:10.3969/j.issn.1005-202X.2018.06.001]
[5]申代友,库洪安,皮红英,等. 基于深度相机的老年跌倒监护系统[J].中国医学物理学杂志,2019,36(2):223.[doi:DOI:10.3969/j.issn.1005-202X.2019.02.019]
 SHEN Daiyou,KU Hongan,PI Hongying,et al. Depth camera-based fall detection system for the elderly[J].Chinese Journal of Medical Physics,2019,36(2):223.[doi:DOI:10.3969/j.issn.1005-202X.2019.02.019]
[6]宫进昌,赵尚义,王远军. 基于深度学习的医学图像分割研究进展[J].中国医学物理学杂志,2019,36(4):420.[doi:DOI:10.3969/j.issn.1005-202X.2019.04.010]
 GONG Jinchang,ZHAO Shangyi,WANG Yuanjun.Research progress on deep learning-based medical image segmentation[J].Chinese Journal of Medical Physics,2019,36(2):420.[doi:DOI:10.3969/j.issn.1005-202X.2019.04.010]
[7]王自强,刘洪运,石金龙,等.基于卷积神经网络的心电图心博识别[J].中国医学物理学杂志,2019,36(8):938.[doi:DOI:10.3969/j.issn.1005-202X.2019.08.015]
 WANG Ziqiang,LIU Hongyun,SHI Jinlong,et al.ECG heartbeat recognition based on convolution neural network[J].Chinese Journal of Medical Physics,2019,36(2):938.[doi:DOI:10.3969/j.issn.1005-202X.2019.08.015]
[8]纪春阳,徐秀林,王燕. 深度神经网络技术在肿瘤细胞识别中的应用[J].中国医学物理学杂志,2019,36(9):1113.[doi:DOI:10.3969/j.issn.1005-202X.2019.09.022]
 JI Chunyang,XU Xiulin,WANG Yan. Application of deep neural network in tumor cell recognition[J].Chinese Journal of Medical Physics,2019,36(2):1113.[doi:DOI:10.3969/j.issn.1005-202X.2019.09.022]
[9]徐航,随力,张靖雯,等.卷积神经网络在医学图像分割中的研究进展[J].中国医学物理学杂志,2019,36(11):1302.[doi:DOI:10.3969/j.issn.1005-202X.2019.11.011]
 XU Hang,SUI Li,ZHANG Jingwen,et al.Progress on convolutional neural network in medical image segmentation[J].Chinese Journal of Medical Physics,2019,36(2):1302.[doi:DOI:10.3969/j.issn.1005-202X.2019.11.011]
[10]蒋家良,罗勇,何奕松,等.特征区域再聚焦提升全卷积神经网络勾画较小靶区准确度[J].中国医学物理学杂志,2020,37(1):75.[doi:DOI:10.3969/j.issn.1005-202X.2020.01.015]
 JIANG Jialiang,LUO Yong,HE Yisong,et al.Feature area refocusing for improving the accuracy of small target area segmentations by fully convolutional networks[J].Chinese Journal of Medical Physics,2020,37(2):75.[doi:DOI:10.3969/j.issn.1005-202X.2020.01.015]

备注/Memo

备注/Memo:
【收稿日期】2019-10-20 【基金项目】国家自然科学基金(11875320);北京市科学技术委员会医药协同科技创新研究(Z181100001918002);中国癌症基金会北京希望马拉松专项基金(LC2018A14) 【作者简介】陈辛元,博士,助理研究员,研究方向:图像引导放疗、人工智能,E-mail: cinya126chen@163.com 【通信作者】戴建荣,E-mail: dai_jianrong@cicams.ac.cn
更新日期/Last Update: 2020-03-03