Abstract:Focused on the issue that U-Net has a poor segmentation effect for small-volume pulmonary nodules, a segmentation method based on deep transfer learning was proposed, and Block Superimposed Fine-Tuning (BSFT) strategy was used to assist the segmentation of pulmonary nodules. Firstly, convolutional neural network was used to learn the feature information of large natural image datasets. Then, the learned features were transferred to the network for the segmentation of small pulmonary nodule image datasets. From the last sampling layer of the network, the network was released block by block and fine-tuned until the network completed the superimposition of the last layer. Finally, the similarity coefficient of Dice was quantitatively analyzed to determine the optimal segmentation network. The experimental results show that the Dice value of BSFT on LUNA16 pulmonary nodule open dataset reaches 0.917 9, which is obviously better than that of the mainstream pulmonary nodule segmentation algorithms.
马金林, 魏萌, 马自萍. 基于深度迁移学习的肺结节分割方法[J]. 计算机应用, 2020, 40(7): 2117-2125.
MA Jinlin, WEI Meng, MA Ziping. Pulmonary nodule segmentation method based on deep transfer learning. Journal of Computer Applications, 2020, 40(7): 2117-2125.
[1] ZHU Z,DONG S,YU C,et al. A text hybrid clustering algorithm based on HowNet semantics[J]. Key Engineering Materials,2011, 474/475/476/476:2071-2078. [2] 张花齐, 王光磊, 李艳, 等. CT图像肺结节的全自动算法研究[J]. 激光杂志,2019,40(4):59-63.(ZHANG H Q,WANG G L,LI Y,et al. Automated algorithm for CT image of pulmonary nodules[J]. Laser Journal,2019,40(4):59-63.) [3] 孙申申, 李宏, 侯欣然, 等. 基于EM和Mean-shift的肺结节分割[J]. 中国图象图形学报,2009,14(10):2016-2022.(SUN S S, LI H,HOU X R,et al. Pulmonary nodule segmentation based on EM and Mean-shift[J]. Journal of Image and Graphics,2009,14(10):2016-2022.) [4] ARMATO S G II,GIGER M L,MORAN C J,et al. Computerized detection of pulmonary nodules on CT scans[J]. RadioGraphics, 1999,19(5):1301-1311. [5] KANAZAWA K,KAWATA Y,NIKI N,et al. Computer-aided diagnosis for pulmonary nodules based on helical CT images[J]. Computerized Medical Imaging and Graphics,1998,22(2):157-167. [6] MIWA T,KAKO J-I,YAMAMOTO S,et al. Automatic detection of lung cancers in chest CT images by the variable N-quoit filter[J]. Systems and Computers in Japan,2002,33(1):53-63. [7] MESSAY T,HARDIE R C,TUINSTRA T R. Segmentation of pulmonary nodules in computed tomography using a regression neural network approach and its application to the lung image database consortium and image database resource initiative dataset[J]. Medical Image Analysis,2015,22(1):48-62. [8] ZHAO J,JI G,XIA Y,et al. Cavitary nodule segmentation in computed tomography images based on self-generating neural networks and particle swarm optimisation[J]. International Journal of Bio-Inspired Computation,2015,7(1):62-67. [9] MONKAM P,QI S,XU M,et al. CNN models discriminating between pulmonary micro-nodules and non-nodules from CT images[J]. BioMedical Engineering OnLine,2018,17(1):No. 96. [10] SHELHAMER E,LONG J,DARRELT. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(4):640-651. [11] CUI S,MAO L,XIONG S. Brain tumor automatic segmentation using fully convolutional networks[J]. Journal of Medical Imaging and Health Informatics,2017,7(7):1641-1647. [12] SHAIKH M,ANAND G,ACHARYA G,et al. Brain tumor segmentation using dense fully convolutional neural network[C]//Proceedings of the 3rd International MICCAI Brainlesion Workshop. Cham:Springer,2017:309-319. [13] RONNEBERGER O,FISCHER P,BROX T. U-Net:convolutional networks for biomedical image segmentation[C]//Proceedings of the 18th International Conference on Medical Image Computing and Computer-Assisted Intervention,LNCS 9351. Cham:Springer,2015:234-241. [14] TONG G,LI Y,CHEN H,et al. Improved U-NET network for pulmonary nodules segmentation[J]. Optik, 2018, 174:460-469. [15] LIU H,CAO H,SONG E M. A cascaded dual-pathway residual network for lung nodule segmentation in CT images[J]. Physica Medica,2019,63:112-121. [16] 张声超. 基于特征融合的肺结节检测关键技术研究[D]. 广州:华南理工大学,2018:33-43. (ZHANG S C. Research on key techniques of pulmonary nodule detection based on feature fusion[D]. Guangzhou:South China University of Technology,2018:33-43.) [17] PAN S J,YANG Q. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering,2010,22(10):1345-1359. [18] OQUAB M,BOTTOU L,LAPTEV I,et al. Learning and transferring mid-level image representations using convolutional neural networks[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2014:1717-1724. [19] 杨涵方, 周向东. 基于深度稀疏辨别的跨领域图像分类[J]. 计算机工程,2018,44(4):310-316. (YANG H F,ZHOU X D. Cross domain image classification based on deep sparse discrimination[J]. Computer Engineering,2018,44(4):310-316.) [20] 李浩波, 祝忠明, 范治中. 基于深度学习的腺体分化程度自动分类[J]. 信息技术与信息化,2019(6):93-94. (LI H B,ZHU Z M,FAN Z Z. Automatic classification of glandular differentiation based on deep learning[J]. Information Technology and Informatization,2019(6):93-94. [21] 李淼, 王敬贤, 李华龙, 等. 基于CNN和迁移学习的农作物病害识别方法研究[J]. 智慧农业,2019,1(3):46-55.(LI M, WANG J X,LI H L,et al. Method for identifying crop disease based on CNN and transfer learning[J]. Smart Agriculture,2019,1(3):46-55.) [22] 徐胜舟, 程时宇. 基于全卷积神经网络迁移学习的乳腺肿块图像分割[J]. 中南民族大学学报(自然科学版),2019,38(2):278-284. (XU S Z,CHENG S Y. Breast mass image segmentation based on transfer learning of fully convolutional neural networks[J]. Journal of South-Central University for Nationalities (Natural Science Edition),2019,38(2):278-284.) [23] TAJBAKHSH N,SHIN J Y,GURUDU S,et al. Convolutional neural networks for medical image analysis:full training or fine tuning?[J]. IEEE Transactions on Medical Imaging,2016,35(5):1299-1312. [24] 褚晶辉, 吴泽蕤, 吕卫, 等. 基于迁移学习和深度卷积神经网络的乳腺肿瘤诊断系统[J]. 激光与光电子学进展,2018,55(8):No. 081001. (CHU J H,WU Z R,LYU W, et al. Breast cancer diagnosis system based on transfer learning and deep convolutional neural networks[J]. Laser & Optoelectronics Progress,2018,55(8):No. 081001.) [25] SHIN H C,ROTH H R, GAO M,et al. Deep convolutional neural networks for computer-aided detection:CNN architectures,dataset characteristics and transfer learning[J]. IEEE Transactions on Medical Imaging,2016,35(5):1285-1298. [26] SIMONYAN K,ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL].[2018-12-01]. https://arxiv.org/pdf/1409.1556.pdf. [27] HE K,ZHANG X,REN S,et al. Deep residual learning for image recognition[C]//Proceedings of the 2016 IEEE Computer Vision and Pattern Recognition. Piscataway:IEEE,2016:770-778. [28] SZEGEDY C,VANHOUCKE V,IOFFE S,et al. Rethinking the inception architecture for computer vision[C]//Proceedings of the 2016 IEEE Computer Vision and Pattern Recognition. Piscataway:IEEE,2016:2818-2826. [29] HUANG G,LIU Z,VAN DER MAATEN L,et al. Densely connected convolutional networks[C]//Proceedings of the 2017 IEEE Computer Vision and Pattern Recognition. Piscataway:IEEE, 2017:2261-2269. [30] KINGMA D P,BA J L. Adam:a method for stochastic optimization[EB/OL].[2018-12-01]. https://arxiv.org/pdf/1412.6980.pdf. [31] WANG S,ZHOU M,LIU Z,et al. Central focused convolutional neural network:developing a data-driven mode for lung nodule segmentation[J]. Medical Image Analysis,2017,40:172-183. [32] MILLETARI F,NAVAB N,AHMADI S A. V-Net:fully convolutional neural networks for volumetric medical image segmentation[C]//Proceedings of the 4th International Conference on 3D Vision. Piscataway:IEEE,2016:565-571.