Journal of Jilin University(Engineering and Technology Edition) ›› 2022, Vol. 52 ›› Issue (10): 2419-2427.doi: 10.13229/j.cnki.jdxbgxb20210278

Previous Articles    

Detecetion of lung nodule based on mask R-CNN and contextual convolutional neural network

Xiao-ying PAN1,2(),De WEI1,2,Yi-zhe ZHAO1,3   

  1. 1.School of Computer Science and Technology，Xi'an University of Posts and Telecommunications，Xi'an 710121，China
    2.Key Laboratory of Network Data Analysis and Intelligent Processing of Shaanxi Province，Xi'an 710121，China
    3.School of Information Science and Technology，Northwest University，Xi'an 710129，China
  • Received:2021-04-01 Online:2022-10-01 Published:2022-11-11

RICH HTML

 15   

PDF (PC)

129

Abstract:

In order to improve the effect of early lung cancer diagnosis， a pulmonary nodule detection algorithm based on deep learning architecture is proposed. A Mask R-CNN pulmonary nodule detection module based on V-Net & R-CNN， and a false-positive attenuation network for pulmonary nodules based on multi-scale and context are designed. The algorithm first uses V-Net for lung nodule localization， and then uses 3D R-CNN for false positive attenuation to determine whether the candidate region is a real lung nodule. Among them， the Inception network is introduced to design a V-Net multi-scale module to realize the localization of pulmonary nodules. A 3D R-CNN pulmonary nodule detection model was designed， and the lung nodule localization results were fused to determine the candidate regions of false-positive pulmonary nodules. The false positive attenuation network classification model designed by the design realizes the false positive judgment of lung nodules and improves the detection effectiveness. The experimental results showed that the algorithm achieves a FROC value of 0.97， which was 5% higher than that of the existing algorithms， indicating that the algorithm has good clinical significance for early detection and diagnosis of lung cancer.

Key words: computer application, lung nodule detection, convolutional neural network, recurrent neural network, multi-scale feature fusion, context convolution

CLC Number: 

  • TP391

Fig.1

3D Inception multi-scale block"

Fig.2

Multi-scale 3D V-Net model"

Fig.3

Workflow of R-CNN algorithm"

Fig.4

Conversion of 2D lung nodule candidate bounding box to 3D candidate bounding box"

Fig.5

False-positive reduction model"

Fig.6

Contextual false-positive reduction model"

Table 1

Model performance of lung nodule false-positive reduction model"

学习率DropoutF1?Score准确率/%
1e-40.50.96496.4
1e-40.80.97193.6
1e-60.50.97998.3
1e-60.80.96896.3

Fig.7

Training model of lung nodule false-positive reduction"

Fig.8

Segmentation results of the lung parenchyma"

Fig.9

Lung nodule segmentation results obtained by multiscale 3D V-Net"

Fig.10

Results of selective search in 2D lung slices"

Fig.11

Line chart for model performance of lung nodule false-positive reduction models"

1 邹小农. 中国肺癌流行病学[J]. 中华肿瘤防治杂志, 2007, 14(12): 881-883.
Zou Xiao-nong. Epidemiology of lung cancer in China[J]. China cancer Prev Trea, 2007, 14(12): 881-883.
2 杨德昌, 杨拴盈. 肺癌诊断水平的现状与进展[J]. 中华结核和呼吸杂志, 2001, 24(8): 450-460.
Yang De-chang, Yang Shuan-ying. Current status and progress of diagnosis of lung cancer[J]. Chinese Journal of Tuberculosis and Respiratory Disease, 2001, 24(8): 450-460.
3 范雪丽, 时高峰, 杜煜, 等. 多层螺旋 CT 扫描在早期肺癌鉴别诊断中的应用[J]. 临床合理用药, 2017, 10(34): 114-116.
Fan Xue-li, Shi Gao-feng, Du-yu, et al. Application of multislice spiral CT scan in differential diagnosis of early lung cancer[J]. Rational Clinical Drug Use, 2017, 10(34): 114-116.
4 曹捍波, 王梅, 王和平. MSCT 对孤立性肺结节 (≤ 2cm) 胸膜凹陷征的诊断及鉴别诊断价值[J]. 医学影像学, 2017, 27(8): 1471-1474.
Cao Han-bo, Wang Mei, Wang He-ping. Value of MSCT in the diagnosis and differential diagnosis of solitary pulmonary nodules (≤ 2 cm) with pleural depression[J]. Medical Radioglaciology, 2017, 27(8): 1471-1474.
5 Ozdemir O, Woodward B, Berlin A A. Propagating uncertainty in multi-stage Bayesian convolutional neural networks with application to pulmonary nodule detection[EB/OL]. [2021-02-26]. arXiv preprint:1712.00497,2017.
6 Tan M, Deklerck R, Jansen B, et al. A novel computer‐aided lung nodule detection system for CT images [J]. Medical Physics, 2011, 38(10): 5630-5645.
7 Setio A A A, 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.
8 Wang B, Qi G J, Tang S, et al. Automated pulmonary nodule detection: High sensitivity with few candidates[C]∥International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: Springer, 2018: 759-767.
9 王伟胜. 基于 CT 影像的肺癌计算机辅助诊断关键技术研究[D]. 长沙: 湖南大学计算机学院, 2016.
Wang Wei-sheng. Research on key techniques of computer-aided diagnosis of lung cancer based on CT images[D]. Changsha: Department of Computer Science, Hunan University, 2016.
10 Szegedy C, Ioffe S, Vanhoucke V, et al. Inception-v4, inception-resnet and the impact of residual connections on learning[C]∥ Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence. San Francisco, USA, 2017: 4278-4284.
11 Ronneberger O, Fischer P, Brox T. U-net: convolutional networks for biomedical image segmentation [C]∥International Conference on Medical Image Computing and Computer-assisted Intervention, Cham: Springer, 2015: 234-241.
12 Cheng J Z, Ni D, Chou Y H, et al. Computer-aided diagnosis with deep learning architecture: applications to breast lesions in US images and pulmonary nodules in CT scans[J]. Scientific Reports, 2016, 6: No. 24454.
13 Buty M, Xu Z Y, Gao M C, et al. Characterization of lung nodule malignancy using hybrid shape and appearance features[C]∥International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: Springer, 2016: 662-670.
14 Ioffe S, Szegedy C. Batch normalization: accelerating deep network training by reducing internal covariate shift[C]//International Conference on Machine Learning, Lille, France, 2015: 448-456.
15 Srivastava N, Hinton G, Krizhevsky A, et al. Dropout: a simple way to prevent neural networks from overfitting[J]. The Journal of Machine Learning Research, 2014, 15(1): 1929-1958.
16 Ding J, Li A, Hu Z, et al. Accurate pulmonary nodule detection in computed tomography images using deep convolutional neural networks[C]∥International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: Springer, 2017: 559-567.
17 Girshick R, Donahue J, Darrell T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, USA,2013: 580-587.
18 Uijlings J R R, van de Sande K E, Gevers T, et al. Selective search for object recognition[J]. International Journal of Computer Vision, 2013, 104(2): 154-171.
19 Dou Q, Chen H, Jin Y, et al. Automated pulmonary nodule detection via 3d convnets with online sample filtering and hybrid-loss residual learning[C]∥International Conference on Medical Image Computing and Computer-Assisted Intervention, Cham: Springer, 2017: 630-638.
20 Bandos A I, Rockette H E, Song T, et al. Area under the free-response ROC curve (FROC) and a related summary index[J]. Biometrics, 2009,65(1):247-256.
[1] Feng-feng ZHOU,Hai-yang ZHU. SEE： sense EEG⁃based emotion algorithm via three⁃step feature selection strategy [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(8): 1834-1841.
[2] Tian BAI,Ming-wei XU,Si-ming LIU,Ji-an ZHANG,Zhe WANG. Dispute focus identification of pleading text based on deep neural network [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(8): 1872-1880.
[3] Xuan-jing SHEN,Xue-feng ZHANG,Yu WANG,Yu-bo JIN. Multi⁃focus image fusion algorithm based on pixel⁃level convolutional neural network [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(8): 1857-1864.
[4] Fu-heng QU,Tian-yu DING,Yang LU,Yong YANG,Ya-ting HU. Fast image codeword search algorithm based on neighborhood similarity [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(8): 1865-1871.
[5] Gui-he QIN,Jun-feng HUANG,Ming-hui SUN. Text input based on two⁃handed keyboard in virtual environment [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(8): 1881-1888.
[6] Jun WANG,Yan-hui XU,Li LI. Data fusion privacy protection method with low energy consumption and integrity verification [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(7): 1657-1665.
[7] Feng-feng ZHOU,Yi-chi ZHANG. Unsupervised feature engineering algorithm BioSAE based on sparse autoencoder [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(7): 1645-1656.
[8] Ming-hua GAO,Can YANG. Traffic target detection method based on improved convolution neural network [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(6): 1353-1361.
[9] Huai-jiang YANG,Er-shuai WANG,Yong-xin SUI,Feng YAN,Yue ZHOU. Simplified residual structure and fast deep residual networks [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(6): 1413-1421.
[10] Yao-long KANG,Li-lu FENG,Jing-an ZHANG,Fu CHEN. Outlier mining algorithm for high dimensional categorical data streams based on spectral clustering [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(6): 1422-1427.
[11] Wen-jun WANG,Yin-feng YU. Automatic completion algorithm for missing links in nowledge graph considering data sparsity [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(6): 1428-1433.
[12] Xue-yun CHEN,Xue-yu BEI,Qu YAO,Xin JIN. Pedestrian segmentation and detection in multi-scene based on G-UNet [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(4): 925-933.
[13] Shi-min FANG. Multiple source data selective integration algorithm based on frequent pattern tree [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(4): 885-890.
[14] Da-xiang LI,Meng-si CHEN,Ying LIU. Spontaneous micro-expression recognition based on STA-LSTM [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(4): 897-909.
[15] Ming LIU,Yu-hang YANG,Song-lin ZOU,Zhi-cheng XIAO,Yong-gang ZHANG. Application of enhanced edge detection image algorithm in multi-book recognition [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(4): 891-896.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
Copyright © Journal of Jilin University(Engineering and Technology Edition), All Rights Reserved.
Tel: +86-431-85095297　Fax: +86-431-85094074　E-mail: xbgxb@jlu.edu.cn
Powered by Beijing Magtech Co. Ltd