Skip to main content
SpringerLink
Log in

A dual-path U-Net for pulmonary vessel segmentation method based on lightweight 3D attention

  • Original Paper
  • Published:
Machine Vision and Applications Aims and scope Submit manuscript

Abstract

In recent years, pulmonary vessel segmentation has aroused widespread interest in medical image analysis. However, the structure of pulmonary vessels is complex and CT images have a lot of noise. Therefore, it is a difficult task to extract pulmonary vessels accurately. In terms of pulmonary vessel segmentation, medical image segmentation methods based on deep learning only utilize single volume data in CT image, but cannot fully fuse multiple volume data, resulting in the accuracy of pulmonary vessel segmentation is low. In order to fully utilize the complementary advantages of multiple volume data, we propose a DS-ResUNet to segment pulmonary vessels in multi-view. The DS-ResUNet uses feature fusion module to fuse 2  mm and 5 mm volume data and makes full use of the detailed vessel textures in 2 mm volume data and coarse vessel structures in 5 mm volume data. By this multi-view fusion module, the accuracy of pulmonary vessel segmentation can be effectively improved. In addition, in order to strengthen the spatial weight of vessels and reduce the model parameters, we design a lightweight 3D axial attention module by separable convolution. To confirm the improved performance, we design some comparison experiments with the state-of-the-art segmentation methods. As a result, our DS-ResUNet has a better performance than other state-of-the-art methods on pulmonary vessel segmentation, but also has fewer parameters.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Similar content being viewed by others

References

  1. Sookpotharom and Supot. The segmentation of dermoscopy images by k-means thresholding

  2. Osuna-Enciso, V., Cuevas, E., Sossa, H.: A comparison of nature inspired algorithms for multi-threshold image segmentation. Expert Syst. Appl. 40(4), 1213–1219 (2013)

    Article  Google Scholar 

  3. Rouhi, R., Jafari, M., Kasaei, S., Keshavarzian, P.: Benign and malignant breast tumors classification based on region growing and CNN segmentation. Expert Syst. Appl. Int. J. 42(3), 990–1002 (2015)

    Article  Google Scholar 

  4. Bertolotto, M., Vo, A.-V., Truong-Hong, L., Laefer, D.F.: Octree-based region growing for point cloud segmentation. ISPRS J. Photogramm. Remote Sens. 104(Jun), 88–100 (2015)

  5. Zhao, Y.Q., Wang, X.H., Wang, X.F., Shih, F.Y.: Retinal vessels segmentation based on level set and region growing. Pattern Recognit. 47(7), 2437–2446 (2014)

    Article  Google Scholar 

  6. Shen T., Wang, Y.: Medical image segmentation based on improved watershed algorithm. In: 2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC), (2018)

  7. Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 39(4), 640–651 (2015)

    Google Scholar 

  8. Ronneberger, O., Fischer, P., Brox, T.: U-net: Convolutional Networks for Biomedical Image Segmentation. Springer, Cham (2015)

    Google Scholar 

  9. Lizarazo, I.: SVM-based segmentation and classification of remotely sensed data. Int. J. Remote Sens. 29(23–24), 7277–7283 (2008)

    Article  Google Scholar 

  10. Park, S., Lee, H.S., Kim, J.: Seed growing for interactive image segmentation using SVM classification with geodesic distance. Electron. Lett. 53(1), 22–24 (2016)

    Article  Google Scholar 

  11. Alush, Amir: Goldberger, Jacob: Hierarchical image segmentation using correlation clustering. IEEE Trans. Neural Netw. Learn. Syst. 27(6), 1358 (2016)

    Article  MathSciNet  Google Scholar 

  12. Burney, S.A., Tariq, H.: K-means cluster analysis for image segmentation. Int. J. Comput. Appl. 96(4), 1–8 (2014)

    Google Scholar 

  13. He, B., Huang, C., Sharp, G., Zhou, S., Hu, Q., Fang, C., Fan, Y., Jia, F.: Fast automatic 3D liver segmentation based on a three-level AdaBoost-guided active shape model. Med. Phys. 43(5), 2421–2434 (2016)

    Article  Google Scholar 

  14. Lupascu, C.A., Tegolo, D., Trucco, E.: FABC: retinal vessel segmentation using AdaBoost. IEEE Trans. Inf. Technol. Biomed. A Publ. IEEE Eng. Med. Biol. Soc, 14(5), 1267–74 (2010)

  15. Abdelhamid, O., Deng, L., Yu D., Jiang, H., Penn, G., Bouvrie, J., Lawrence, S., Giles, C.L., Tsoi, A.C., Back, A.D.: Gradient-based learning applied to document recognition (2013)

  16. Everingham, M., Eslami, S., Van Gool, L., Williams, C., Winn, J., Zisserman, A.: The pascal visual object classes challenge: a retrospective. Int. J. Comput. Vis. 111(1), 98–136 (2015)

    Article  Google Scholar 

  17. Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv, (2014)

  18. He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: IEEE, (2016)

  19. Trivizakis, E., Manikis, G.C., Nikiforaki, K., Drevelegas, K., Constantinides, M., Drevelegas, A., Marias, K.: Extending 2D convolutional neural networks to 3D for advancing deep learning cancer classification with application to MRI liver tumor differentiation. IEEE J. Biomed. Health Inf. 23(3), 923–930 (2018)

    Article  Google Scholar 

  20. Badrinarayanan, V., Kendall, A., Cipolla, R.: Segnet: a deep convolutional encoder-decoder architecture for image segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 39(12), 2481–2495 (2017)

    Article  Google Scholar 

  21. Chen, L.C., Papandreou, G., Kokkinos, I., Murphy, K., Yuille, A.L.: Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs. IEEE Trans. Pattern Anal. Mach. Intell. 40(4), 834–848 (2018)

    Article  Google Scholar 

  22. Zhao, H., Shi, J., Qi, X., Wang, X., Jia, J.: Pyramid scene parsing network. In: IEEE Computer Society, (2016)

  23. Yu, L., Cheng, J.Z., Qi, D., Xin, Y., Heng, P.A.: Automatic 3d cardiovascular MR segmentation with densely-connected volumetric convnets (2017)

  24. Alom, M.Z., Hasan, M., Yakopcic, C., Taha, TM, Asari, V.K.: Recurrent residual convolutional neural network based on u-net (r2u-net) for medical image segmentation (2018)

  25. Wang, B., Qiu, S., He, H.: Dual Encoding U-Net for Retinal Vessel Segmentation. In: Medical Image Computing and Computer Assisted Intervention—MICCAI 2019, 22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Part I (2019)

  26. Wu, H., Zhang, J., Huang, K., Liang, K., Yu, Y.: Fastfcn: rethinking dilated convolution in the backbone for semantic segmentation (2019)

  27. Yisong, H.E., Jiang, J., Hang, Y.U., Yuchuan, F.U., Radiotherapy, D.O., Hospital, W.C., University, S.: Comparison of dice similarity coefficient and Hausdorff distance in image segmentation. Chin. J. Med. Phys. 36(11), 1307–1311 (2019)

    Google Scholar 

Download references

Acknowledgements

We acknowledge the support of the Natural Science Foundation of Zhejiang Province (Grant No. LY22F020001) and the 3315 Plan Foundation of Ningbo (Grant No. 2019B-18-G).

Author information

Authors and Affiliations

  1. Ningbo University, 818 Fenghua Road, Ningbo, 315211, China

    Rencheng Wu, Yu Xin, Yihong Dong & Jiangbo Qian

Authors
  1. Rencheng Wu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Yu Xin
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yihong Dong
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Jiangbo Qian
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Yu Xin.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wu, R., Xin, Y., Dong, Y. et al. A dual-path U-Net for pulmonary vessel segmentation method based on lightweight 3D attention. Machine Vision and Applications 34, 87 (2023). https://doi.org/10.1007/s00138-023-01442-x

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s00138-023-01442-x

Keywords

Search

Navigation