Advertisement

Scale Normalization Cascaded Dense-Unet for Prostate Segmentation in MR Images

  • Yuxuan Chen
  • Suiyi Li
  • Su YangEmail author
  • Wuyang Luo
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11902)

Abstract

Automated and accurate prostate segmentation technique from magnetic resonance images plays an important role in diagnostic and radiological planning. However, this task faces the challenge of extreme scale variation of prostate glands presented in the slices at different locations of MRI volumes. To alleviate problems arising from scale variation. We propose a cascaded prostate segmentation model that includes three stages: Coarse segmentation, segmentation result refinement, and scale normalization segmentation. Segmentation result refinement can remove the coarse segmentation results that do not contain prostates. More importantly, it normalizes the scale of the prostate region on different slice images of the same nuclear magnetic resonance volume according to the result of the coarse segmentation, thereby making the scale normalization segmentation network obtain scale-invariant magnetic resonance images as input. The experimental results demonstrate that this design can significantly reduce the degradation of segmentation performance arising from large scale variation.

Keywords

Prostate segmentation Scale normalization Cascaded model MRI 

Notes

Acknowledgement

This work is supported by Shanghai Science and Technology Commission (grant No. 17511104203) and NSFC (grant NO. 61472087).

References

  1. 1.
    Vos, P., Barentsz, J., Karssemeijer, N., Huisman, H.: Automatic computer-aided detection of prostate cancer based on multiparametric magnetic resonance image analysis. Phys. Med. Biol. 57(6), 1527 (2012)CrossRefGoogle Scholar
  2. 2.
    Toth, R., et al.: Accurate prostate volume estimation using multifeature active shape models on T2-weighted MRI. Acad. Radiol. 18(6), 745–754 (2011)CrossRefGoogle Scholar
  3. 3.
    Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015).  https://doi.org/10.1007/978-3-319-24574-4_28CrossRefGoogle Scholar
  4. 4.
    Zhu, Q., Du, B., Turkbey, B., Choyke, P., Yan, P.: Exploiting interslice correlation for MRI prostate image segmentation, from recursive neural networks aspect. Complexity 10 (2018)Google Scholar
  5. 5.
    Li, X., Chen, H., Qi, X., Dou, Q., Fu, C., Heng, P.: H-DenseUNet: hybrid densely connected UNet for liver and liver tumor segmentation from CT volumes. IEEE Trans. Med. Imaging 37(12), 2663–2674 (2018)CrossRefGoogle Scholar
  6. 6.
    Yu, L., et al.: Automatic 3D cardiovascular MR segmentation with densely-connected volumetric convnets. In: Descoteaux, M., Maier-Hein, L., Franz, A., Jannin, P., Collins, D.L., Duchesne, S. (eds.) MICCAI 2017. LNCS, vol. 10434, pp. 287–295. Springer, Cham (2017).  https://doi.org/10.1007/978-3-319-66185-8_33CrossRefGoogle Scholar
  7. 7.
    Chen, H., Dou, Q., Yu, L., Heng, P.-A.: VoxResNet: deep voxelwise residual networks for volumetric brain segmentation. arXiv preprint arXiv:1608.05895 (2016)
  8. 8.
    Klein, S., van der Heide, U., Lipps, I., Vulpen, M., Staring, M., Pluim, J.: Automatic segmentation of the prostate in 3-D MR images by atlas matching using localized mutual information. Med. Phys. 35(4), 1407–1417 (2008)CrossRefGoogle Scholar
  9. 9.
    Martin, S., Daanen, V., Troccaz, J.: Automated segmentation of the prostate 3-D MR images using a probabilistic atlas and a spatially constrained deformable model. Med. Phys. 37(4), 1579–1590 (2010)CrossRefGoogle Scholar
  10. 10.
    Ou, Y., Doshi, J., Erus, G., Davatzikos, C.: Multi-atlas segmentation of the prostate: a zooming process with robust registration and atlas selection. In: MICCAI Grand Challenge: Prostate MR Image Segmentation (2012)Google Scholar
  11. 11.
    Yan, P., Cao, Y., Yuan, Y., Turkbey, B., Choyke, P.L.: Label image constrained multi-atlas selection. IEEE Trans. Cybernet. 45(6), 1158–1168 (2015)CrossRefGoogle Scholar
  12. 12.
    Pasquier, D., Lacornerie, T., Vermandel, M., Rousseau, J., Lartigau, E., Betrouni, N.: Automatic segmentation of pelvic structures from magnetic resonance images for prostate cancer radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 68(2), 592–600 (2007)CrossRefGoogle Scholar
  13. 13.
    Makni, N., Puech, P., Lopes, R., Dewalle, A.: Combining a deformable model and a probabilistic framework for an automatic 3-D segmentation of prostate on MRI. Int. J. Comput. Assisted. Radiol. Surg. 4(2), 181–188 (2009)CrossRefGoogle Scholar
  14. 14.
    Toth, R., Madabhushi, A.: Multifeature landmark-free active appearance models: application to prostate MRI segmentation. IEEE Trans. Med. Imag 31(8), 1638–1650 (2012)CrossRefGoogle Scholar
  15. 15.
    Moschidis E., Graham, J.: Automatic differential segmentation of the prostate in 3-D MRI using random forest classification and graph cuts optimization. In: Proceedings of IEEE International Symposium on Biomedical Imaging, pp. 1727–1730 (2012)Google Scholar
  16. 16.
    Zouqi M., Samarabandu, J.: Prostate segmentation from 2-D ultrasound images using graph cuts and domain knowledge. In: Proceedings of Computer and Robot Vision Conference, pp. 359–362 (2008)Google Scholar
  17. 17.
    Tian, Z., Liu, L., Zhang, Z., Fei, B.: Superpixelbased segmentation for 3D prostate mr images. IEEE Trans. Med. Imaging 35(3), 791–801 (2016)CrossRefGoogle Scholar
  18. 18.
    Guo, Y., Gao, Y., Shen, D.: Deformable MR prostate segmentation via deep feature learning and sparse patch matching. IEEE Trans. Med. Imag. 35(4), 1077–1089 (2016)CrossRefGoogle Scholar
  19. 19.
    Jia, H., Xia, Y., Song, Y., Cai, W., Fulham, M., Feng, D.D.: Atlas registration and en- semble deep convolutional neural network-based prostate segmentation using magnetic resonance imaging. Neurocomputing 275, 1358–1369 (2017)CrossRefGoogle Scholar
  20. 20.
    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
  21. 21.
    Yu, L., Yang, X., Chen, H., Qin, J., Heng, P.-A.: Volumetric ConvNets with mixed residual connections for automated prostate segmentation from 3D MR images. In: AAAI, pp. 66–72 (2017)Google Scholar
  22. 22.
    Yan, K., Wang, X., Kim, J., et al.: A propagation-DNN: deep combination learning of multi-level features for MR prostate segmentation. Comput. Methods Programs Biomed. 170, 11–21 (2019)CrossRefGoogle Scholar
  23. 23.
    Li, S., et al.: Cascade dense-Unet for prostate segmentation in MR images. In: ICIC (2019, accepted)Google Scholar
  24. 24.
    Huang, G., Liu, Z., Weinberger, K.Q., van der Maaten, L.: Densely connected convolutional networks. CoRR, abs/1608.06993 (2016)Google Scholar
  25. 25.
    Litjens, G., et al.: Evaluation of prostate segmentation algorithms for MRI: the PROMISE12 challenge. Med. Image Anal. 18(2), 359–373 (2014)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Shanghai Key Laboratory of Intelligent Information Processing, School of Computer ScienceFudan UniversityShanghaiChina

Personalised recommendations