Advertisement

DPA-DenseBiasNet: Semi-supervised 3D Fine Renal Artery Segmentation with Dense Biased Network and Deep Priori Anatomy

  • Yuting He
  • Guanyu YangEmail author
  • Yang Chen
  • Youyong Kong
  • Jiasong Wu
  • Lijun Tang
  • Xiaomei Zhu
  • Jean-Louis Dillenseger
  • Pengfei Shao
  • Shaobo Zhang
  • Huazhong Shu
  • Jean-Louis Coatrieux
  • Shuo Li
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11769)

Abstract

3D fine renal artery segmentation on abdominal CTA image targets on the segmentation of the complete renal artery tree which will help clinicians locate the interlobar artery’s corresponding blood feeding region easily. However, it is still a challenging task that no one has reported success due to the large intra-scale changes, large inter-anatomy variation, thin structures, small volume ratio and limitation of labeled data. Hence, in this paper, we propose a novel semi-supervised learning framework named DPA-DenseBiasNet for 3D fine renal artery segmentation. The dense biased connection method is presented for multi-receptive field feature maps merging and implicit deep supervision [5] which enable the network to adapt to large intra-scale changes and improve its training process. The dense biased network (DenseBiasNet) is designed based on this method. We develop deep priori anatomy (DPA) for semi-supervised learning of thin structures. Differ from other semi-supervised methods, it embeds priori anatomical features to segmentation network which avoids inaccurate results sensitive to thin structures as optimizing targets, so that the network achieves generalization of different anatomies with the help of unlabeled data. Only 26 labeled and 118 unlabeled images were used to train our framework and it achieves satisfactory results on the testing dataset. The mean centerline voxel distance is 1.976 which reduced by 3.094 compared to 3D U-Net. The results illustrate that our framework has great prospects in the diagnosis and treatment of kidney disease.

Notes

Acknowledgements

This research was supported by the National Key Research and Development Program of China (2017YFC0107903), National Natural Science Foundation under grants (31571001, 61828101), the Short-Term Recruitment Program of Foreign Experts (WQ20163200398), Key Research and Development Project of Jiangsu Province (BE2018749) and Southeast University-Nanjing Medical University Cooperative Research Project (2242019K3DN08).

References

  1. 1.
    Cancer stat facts: Kidney and renal pelvis cancer. https://seer.cancer.gov/statfacts/html/kidrp.html. Accessed 12 Feb 2019
  2. 2.
    Bai, W., et al.: Semi-supervised learning for network-based cardiac MR image segmentation. In: Descoteaux, M., Maier-Hein, L., Franz, A., Jannin, P., Collins, D.L., Duchesne, S. (eds.) MICCAI 2017. LNCS, vol. 10434, pp. 253–260. Springer, Cham (2017).  https://doi.org/10.1007/978-3-319-66185-8_29CrossRefGoogle Scholar
  3. 3.
    Baur, C., Albarqouni, S., Navab, N.: Semi-supervised deep learning for fully convolutional networks. In: Descoteaux, M., Maier-Hein, L., Franz, A., Jannin, P., Collins, D.L., Duchesne, S. (eds.) MICCAI 2017. LNCS, vol. 10435, pp. 311–319. Springer, Cham (2017).  https://doi.org/10.1007/978-3-319-66179-7_36CrossRefGoogle Scholar
  4. 4.
    Çiçek, Ö., Abdulkadir, A., Lienkamp, S.S., Brox, T., Ronneberger, O.: 3D U-Net: learning dense volumetric segmentation from sparse annotation. In: Ourselin, S., Joskowicz, L., Sabuncu, M.R., Unal, G., Wells, W. (eds.) MICCAI 2016. LNCS, vol. 9901, pp. 424–432. Springer, Cham (2016).  https://doi.org/10.1007/978-3-319-46723-8_49CrossRefGoogle Scholar
  5. 5.
    Lee, C.Y., Xie, S., Gallagher, P., Zhang, Z., Tu, Z.: Deeply-supervised nets. Eprint Arxiv, pp. 562–570 (2014)Google Scholar
  6. 6.
    Li, J., Lo, P., Taha, A., Wu, H., Zhao, T.: Segmentation of renal structures for image-guided surgery. In: Frangi, A.F., Schnabel, J.A., Davatzikos, C., Alberola-López, C., Fichtinger, G. (eds.) MICCAI 2018. LNCS, vol. 11073, pp. 454–462. Springer, Cham (2018).  https://doi.org/10.1007/978-3-030-00937-3_52CrossRefGoogle Scholar
  7. 7.
    Ljungberg, B., et al.: Eau guidelines on renal cell carcinoma: 2014 update. Eur. Urol. 67(5), 913–924 (2015)CrossRefGoogle Scholar
  8. 8.
    Milletari, F., Navab, N., Ahmadi, S.A.: V-net: fully convolutional neural networks for volumetric medical image segmentation. In: 2016 Fourth International Conference on 3D Vision (3DV), pp. 565–571. IEEE (2016)Google Scholar
  9. 9.
    Nie, D., Gao, Y., Wang, L., Shen, D.: ASDNet: attention based semi-supervised deep networks for medical image segmentation. In: Frangi, A.F., Schnabel, J.A., Davatzikos, C., Alberola-López, C., Fichtinger, G. (eds.) MICCAI 2018. LNCS, vol. 11073, pp. 370–378. Springer, Cham (2018).  https://doi.org/10.1007/978-3-030-00937-3_43CrossRefGoogle Scholar
  10. 10.
    Petru, B., Elena, Ş., Dan, I., Klara, B., Radu, B., Constantin, D.: Morphological assessments on the arteries of the superior renal segment. Surg. radiol. Anat. 34(2), 137–144 (2012)CrossRefGoogle Scholar
  11. 11.
    Shao, P., et al.: Laparoscopic partial nephrectomy with segmental renal artery clamping: technique and clinical outcomes. Eur. Urol. 59(5), 849–855 (2011)CrossRefGoogle Scholar
  12. 12.
    Shao, P., et al.: Precise segmental renal artery clamping under the guidance of dual-source computed tomography angiography during laparoscopic partial nephrectomy. Eur. Urol. 62(6), 1001–1008 (2012)CrossRefGoogle Scholar
  13. 13.
    Taha, A., Lo, P., Li, J., Zhao, T.: Kid-net: convolution networks for kidney vessels segmentation from CT-volumes. arXiv preprint arXiv:1806.06769 (2018)
  14. 14.
    Vincent, P., Larochelle, H., Bengio, Y., Manzagol, P.A.: Extracting and composing robust features with denoising autoencoders. In: Proceedings of the 25th International Conference on Machine Learning, pp. 1096–1103. ACM (2008)Google Scholar
  15. 15.
    Zhu, X.J.: Semi-supervised learning literature survey, Technical report. University of Wisconsin-Madison Department of Computer Sciences (2005)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Yuting He
    • 1
  • Guanyu Yang
    • 1
    • 4
    Email author
  • Yang Chen
    • 1
    • 4
  • Youyong Kong
    • 1
    • 4
  • Jiasong Wu
    • 1
    • 4
  • Lijun Tang
    • 3
  • Xiaomei Zhu
    • 3
  • Jean-Louis Dillenseger
    • 2
    • 4
  • Pengfei Shao
    • 5
  • Shaobo Zhang
    • 5
  • Huazhong Shu
    • 1
    • 4
  • Jean-Louis Coatrieux
    • 2
  • Shuo Li
    • 6
  1. 1.LIST, Key Laboratory of Computer Network and Information Integration (Southeast University), Ministry of EducationNanjingChina
  2. 2.Univ Rennes, Inserm, LTSI - UMR1099RennesFrance
  3. 3.Department of RadiologyThe First Affiliated Hospital of Nanjing Medical UniversityNanjingChina
  4. 4.Centre de Recherche en Information Biomédicale Sino-Français (CRIBs)StrasbourgFrance
  5. 5.Department of UrologyThe First Affiliated Hospital of Nanjing Medical UniversityNanjingChina
  6. 6.Department of Medical BiophysicsUniversity of Western OntarioLondonCanada

Personalised recommendations