基于三维密集网络的肺结节检测方法

doi:10.3969/j.issn.0258-8021.2020.01.02

Abstract
Figure/Table
References (28)
Related Citation (5)

Download: PDF (3719 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract To solve the problem that the detection rate of pulmonary nodules is low and require large amounts of calculation by traditional method which based on three-dimensional features, a more efficient three-dimensional-dense-network based method for pulmonary nodules’ detection was proposed. The method firstly involved the introduction of a densely connected unit into the 3D U-Net to construct a 3D Dense U-Net (3D Densely connected U-Net) network for lung nodule detection. Since 3D Dense U-Net replaced the original ordinary convolution layer of 3D with a densely connected block, it facilitated the maximization of the information flows between layers; it not only solved the problem of feature redundancy in traditional stacked networks, but also speeded up the network training. Additionally, the new method retained the fundamental linking method of U-Net to reuse the underlying features, which enabled an effective obtainment of the candidate nodules. On this basis, in order to solve the problem of false nodules in the candidate nodules, in order to more effectively acquire the features of the nodules and improve the ability of the network to identify the nodules, a three-dimensional dense classification network (3D DenseNet) was used to eliminate false-positive nodules. By testing the data collection of Tianchi Medical AI Competition, the total detection sensitivity for lung nodules using the proposed method reached 94.3%; and the sensitivity of nodules less than 10 mm was 91.5%, and the false-positive rate was 5.9%. The experimental results revealed that the proposed method reached greater sensitivity in detecting small nodules, which not only raised the nodules’ detection rate, but also improved the computational efficiency.

Key words： computer-aided diagnosis nodule detection dense connection 3D convolution

Received: 11 January 2019

PACS:

R318

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Wang Shangli
	Jin Gehui
	Xu Liang
	Jin Wei
	Yin Caoqian
	Fu Randi

Cite this article:

Wang Shangli,Jin Gehui,Xu Liang, et al. Method for Detecting Pulmonary Nodules Based on Three-Dimensional Dense Network[J]. Chinese Journal of Biomedical Engineering, 2020, 39(1): 8-18.

URL:

http://cjbme.csbme.org/EN/10.3969/j.issn.0258-8021.2020.01.02 OR http://cjbme.csbme.org/EN/Y2020/V39/I1/8

[1] Chen Wanqing, Sun Kexin, Zheng Rongshou, et al. Cancer incidence and mortality in China, 2014[J]. Chinese Journal of Cancer Research, 2018, 30(1):1-12.
[2] Infante M, Cavuto S, Lutman FR, et al. A randomized study of lung cancer screening with spiral computed tomography: three-year results from the DANTE trial [J]. American Journal of Respiratory & Critical Care Medicine, 2009, 180(5):445-453.
[3] Singh S, Gierada DS, Pinsky P, et al. Reader variability in identifying pulmonary nodules on chest radiographs from the national lung screening trial.[J]. Journal of Thoracic Imaging, 2012, 27(4):249-254.
[4] 吴光耀, 伍建林. 基于CT的计算机辅助检测与诊断肺结节技术研究进展[J]. 中国医学影像技术, 2018, 34(7):1114-1117.
[5] Murphy K, Van GB, Schilham AM, et al. A large-scale evaluation of automatic pulmonary nodule detection in chest CT using local image features and k-nearest-neighbour classification[J]. Medical Image Analysis, 2009, 13(5):757-770.
[6] Jacobs C, Van Rikxoort EM, Twellmann T, et al. Automatic detection of subsolid pulmonary nodules in thoracic computed tomography images[J]. Medical Image Analysis, 2014, 18(2):374-384.
[7] Van Ginneken B, Armato SG, de Hoop B, et al. Comparing and combining algorithms for computer-aided detection of pulmonary nodules in computed tomography scans: The anode09 study[J]. Medical Image Analysis, 2010,14(6): 707-722.
[8] Krizhevsky A, Sutskever I, Hinton GE. Imagenet classification with deep convolutional neural networks[C]//Neural Information Processing Systems. Lake Taboe: Nips Foundation, 2012: 1097-1105.
[9] Setio AA, Ciompi F, Litjens G, et al. Pulmonary nodule detection in CT images: false positive reduction using multi-view convolutional networks[J]. IEEE Transactions on Medical Imaging, 2016, 35(5):1160-1169.
[10] Setio AAA, Traverso A, van Ginneken B, et al. Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the luna16 challenge [J]. Mesical Image Analysis, 2016, 42:1-13.
[11] Zhu Wentao, Liu Chaochun, Wei Fan, et al. DeepLung: Deep 3D dual path nets for automated pulmonary nodule detection and classification[C]// Proceedings of the 2018 IEEE Winter Conference on Applications of Computer Vision (WACV). Lake Tahoe:IEEE, 2018:673-681.
[12] Jin Hongsheng, Li Zongyao, Tong Ruofeng, et al. A deep 3D residual CNN for false positive reduction in pulmonary nodule detection[J]. Medical Physics, 2018, 45(5):2097-2107.
[13] Huang Xiaojie, Shan Junjie, Vaidya V, et al. Lung nodule detection in CT using 3D convolutional neural networks[C]// Proceedings of the IEEE International Symposium on Biomedical Imaging. Melbourne: IEEE, 2017:379-383.
[14] 吕晓琪, 吴凉, 谷宇, 等. 基于三维卷积神经网络的低剂量CT肺结节检测[J]. 光学精密工程, 2018, 26(5):213-220.
[15] 赵鹏飞, 赵涓涓, 强彦,等. 多输入卷积神经网络肺结节检测方法研究[J]. 计算机科学, 2018, 45(1):162-166.
[16] Shelhamer E, Long J, Darrell T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2014, 39(4):640-651.
[17] Ronneberger O, Fischer P, Brox T. U-Net: Convolutional networks for biomedical image segmentation [C]//Medical Image Computing and Computer Assisted Intervention. Cham:Springer 2015:234-241.
[18] Simonyan K, Zisserman A. Very deep convolutional networks for Large-Scale image recognition [J]. Computer Science, 2014, arXiv:1409.1556v6.
[19] Szegedy C, Liu Wei, Jia Yangqing, et al. Going deeper with convolutions[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition(CPVR). Las Vegas: IEEE, 2015:1-9.
[20] He Kaiming, Zhang Xiangyu, Ren Shaoqing, et al. Deep residual learning for image recognition[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas:IEEE, 2016:770-778.
[21] Ji Shuiwang, Xu Wei, Yang Ming, et al. 3D convolutional neural networks for human action recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(1):221-231.
[22] içek, Özgün, Abdulkadir A, Lienkamp SS, et al. 3D U-Net: Learning dense volumetric segmentation from sparse annotation[C]// Medical Image Computing and Computer Assisted Intervention. Cham:Springer, 2016:424-432.
[23] Huang Gao, Liu Zhuang, L. v. Maaten, et al. Densely connected convolutional networks[C]// Proceedings of the 2017 IEEEConference on Computer Vision and Pattern Recognition(CVPR). Las Vegas:IEEE, 2017: 4700-4708.
[24] He Kaiming,Zhang Xiangyu,Ren Shaoqing, et al. Delving deep into rectifiers: Surpassing human-level performance on ImageNet classification[C]//IEEE International Conference on Computer Vision (ICCV). Santiago:IEEE, 2015:1026-1034.
[25] Kingma D, Ba J. Adam: A method for stochastic optimization[J]. Computer Science, 2014, arXiv:1412.6980v9.
[26] 张婷. 基于受限玻尔兹曼机的肺结节检测与诊断模型[D].太原:太原理工大学,2018.
[27] 贺兴怡,龚敬,王丽嘉,等. 基于特征矢量化的肺结节特征选择算法[J].计算机应用研究,2018,35(8):2544-2548.
[28] 严忱君. 基于机器视觉的肺结节初筛算法研究与实现[D].杭州:浙江大学,2018.