[1]周正东,李剑波,辛润超,等.基于带孔U-net神经网络的肺癌危及器官并行分割方法[J].东南大学学报(自然科学版),2019,49(2):231-236.[doi:10.3969/j.issn.1001-0505.2019.02.005]
 Zhou Zhengdong,Li Jianbo,Xin Runchao,et al.Parallel segmentation method for organs at risk in lung cancer based on dilated U-net neural network[J].Journal of Southeast University (Natural Science Edition),2019,49(2):231-236.[doi:10.3969/j.issn.1001-0505.2019.02.005]
点击复制

基于带孔U-net神经网络的肺癌危及器官并行分割方法()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
49
期数:
2019年第2期
页码:
231-236
栏目:
计算机科学与工程
出版日期:
2019-03-20

文章信息/Info

Title:
Parallel segmentation method for organs at risk in lung cancer based on dilated U-net neural network
作者:
周正东1李剑波12辛润超12涂佳丽12贾俊山1魏士松1
1南京航空航天大学机械结构力学及控制国家重点实验室, 南京 210016; 2南京航空航天大学核科学与工程系, 南京 210016
Author(s):
Zhou Zhengdong1 Li Jianbo12 Xin Runchao12 Tu Jiali12 Jia Junshan1 Wei Shisong1
1State Key Laboratory of Mechanics and Control of Mechanical Structures, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
2Department of Nuclear Science and Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
关键词:
肺癌放射治疗 深度学习 带孔卷积 图像分割
Keywords:
lung cancer radiotherapy deep learning dilated convolution image segmentation
分类号:
TP391
DOI:
10.3969/j.issn.1001-0505.2019.02.005
摘要:
为了提高肺癌放疗计划危及器官勾画的精度和效率,提出了一种基于带孔U-net神经网络的肺癌放疗计划危及器官肺及心脏的并行分割方法.首先,构建了肺窗、心脏窗以及纵膈窗下的三通道伪彩色图像数据集,将图像数据集分成训练集、验证集以及测试集;然后,搭建了带孔U-net神经网络,利用训练集和验证集对其进行训练和参数调优;最后,利用测试集对训练后的带孔U-net神经网络进行图像分割性能评价,并与U-net神经网络及3种传统图像分割算法进行比较.实验结果表明,带孔U-net神经网络分割性能最优,可有效地完成肺及心脏的自动并行分割,提高勾画效率,分割结果与人工勾画结果相当.
Abstract:
To improve the accuracy and efficiency of the delineation of organs at risk in lung cancer radiotherapy planning, a parallel segmentation method for heart and lung based on the dilated U-net neural network for lung cancer radiotherapy planning was proposed. First, the three-channel pseudo color image data set was constructed with lung window, heart window and mediastinal window, and it was split into the training set, the validation set and the test set. Then, the dilated U-net neural network was built, trained and optimized by the training set and the validation set. Finally, the segmentation performance of the dilated U-net neural network was evaluated by the test data set, and compared with those of the U-net neural network and three traditional image segmentation algorithms. The experimental results show that the dilated U-net neural network has the best segmentation performance. It can effectively achieve the automatic parallel segmentation of lung and heart. The delineation efficiency can be improved and the segmentation result is approximate to the manual segmentation result.

参考文献/References:

[1] 陈万青, 孙可欣, 郑荣寿, 等. 2014年中国分地区恶性肿瘤发病和死亡分析[J]. 中国肿瘤, 2018, 27(1): 1-14.DOI:10.11735/j.issn.1004-0242.2018.01.A001.
Chen W Q, Sun K X, Zheng R S, et al. Report of cancer incidence and mortality in different areas of China, 2014[J]. China Cancer, 2018, 27(1): 1-14.DOI:10.11735/j.issn.1004-0242.2018.01.A001. (in Chinese)
[2] 陈首英, 刘福林, 庞志刚, 等. 肺癌病人5年生存率及生存因素分析[J]. 预防医学文献信息, 2004, 10(1): 1-3.DOI:10.3969/j.issn.1672-9153.2004.01.001.
Chen S Y, Liu F L, Pang Z G, et al. Analysis about survival rate and survival factor of the patients with lung cancer in 5 years[J]. Literature and Information on Preventive, 2004, 10(1): 1-3.DOI:10.3969/j.issn.1672-9153.2004.01.001. (in Chinese)
[3] Than J C M, Noor N M, Rijal O M, et al. Enhanced automatic lung segmentation using graph cut for interstitial lung disease[C]//2014 IEEE Conference on Biomedical Engineering and Sciences(IECBES). Kuala Lumpur, Malaysia, 2014: 17-21. DOI:10.1109/IECBES.2014.7047479.
[4] Than J C M,Noor N M, Rijal O M, et al. Lung segmentation for HRCT thorax images using radon transform and accumulating pixel width[C]//2014 IEEE Region 10 Symposium. Kuala Lumpur, Malaysia, 2014: 157-161. DOI:10.1109/TENCONSpring.2014.6863016.
[5] Gao H, Dou L, Chen W, et al. The applications of image segmentation techniques in medical CT images[C]//30th Chinese Control Conference. Yantai, China, 2011: 3296-3299.
[6] Kumar S.Automatic lung nodule segmentation using autoseed region growing with morphological masking(ARGMM)and feature ex-traction through complete local binary pattern and microscopic information pattern[J].Euromediterranean Biomedical Journal,2015,10(5):99-119.
[7] Rebouças Filho P P, Cortez P C, da Silva Barros A C, et al. Novel and powerful 3D adaptive crisp active contour method applied in the segmentation of CT lung images[J]. Medical Image Analysis, 2017, 35: 503-516. DOI:10.1016/j.media.2016.09.002.
[8] Qiang Y, Zhang X H, Ji G H, et al. Automated lung nodule segmentation using an active contour model based on PET/CT images[J]. Journal of Computational and Theoretical Nanoscience, 2015, 12(8): 1972-1976. DOI:10.1166/jctn.2015.4216.
[9] Zhang W L, Li R J, Deng H T, et al. Deepconvolutional neural networks for multi-modality isointense infant brain image segmentation[J]. NeuroImage, 2015, 108: 214-224. DOI:10.1016/j.neuroimage.2014.12.061.
[10] 郭树旭,马树志,李晶,等.基于全卷积神经网络的肝脏CT影像分割研究[J].计算机工程与应用,2017,53(18):126-131. DOI:10.3778/j.issn.1002-8331.1611-0523.
Guo S X, Ma S Z, Li J, et al. Fully convolutional neural network for liver segmentation in CT image[J]. Computer Engineering and Applications, 2017, 53(18): 126-131. DOI:10.3778/j.issn.1002-8331.1611-0523. (in Chinese)
[11] Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation[M]//Ronneberger O, Fischer P, Brox T, eds. Lecture notes in computer science. Cham: Springer International Publishing, 2015: 234-241. DOI:10.1007/978-3-319-24574-4_28.
[12] Yu F, Koltun V. Multi-scale context aggregation by dilated convolutions[C]//2016 ICLR. Sam Juan, Puerto Rico, 2016:1-9.
[13] Ioffe S, Szegedy C. Batch normalization: Accelerating deep network training by reducing internal covariate shift[C]// Proceedings of the 32nd International Conference on Machine Learning. Lille, France, 2015:448-456.

相似文献/References:

[1]程旭,张毅锋,刘袁,等.基于深度特征的目标跟踪算法[J].东南大学学报(自然科学版),2017,47(1):1.[doi:10.3969/j.issn.1001-0505.2017.01.001]
 Cheng Xu,Zhang Yifeng,Liu Yuan,et al.Object tracking algorithm based on deep feature[J].Journal of Southeast University (Natural Science Edition),2017,47(2):1.[doi:10.3969/j.issn.1001-0505.2017.01.001]
[2]李林超,曲栩,张健,等.基于特征级融合的高速公路异质交通流数据修复方法[J].东南大学学报(自然科学版),2018,48(5):972.[doi:10.3969/j.issn.1001-0505.2018.05.029]
 Li Linchao,Qu Xu,Zhang Jian,et al.Missing value imputation method for heterogeneous traffic flow data based on feature fusion[J].Journal of Southeast University (Natural Science Edition),2018,48(2):972.[doi:10.3969/j.issn.1001-0505.2018.05.029]

备注/Memo

备注/Memo:
收稿日期: 2018-09-18.
作者简介: 周正东(1969—),男,博士,副教授,zzd_msc@nuaa.edu.cn.
基金项目: 国家自然科学基金资助项目(51575256)、江苏省重点研发计划(社会发展)重点资助项目(BE2017730)、重庆市产业类重点研发资助项目(重大主题专项项目)(cstc2017zdcy-zdzxX0007)、江苏高校优势学科建设工程资助项目.
引用本文: 周正东,李剑波,辛润超,等.基于带孔U-net神经网络的肺癌危及器官并行分割方法[J].东南大学学报(自然科学版),2019,49(2):231-236. DOI:10.3969/j.issn.1001-0505.2019.02.005.
更新日期/Last Update: 2019-03-20