Advertisement

IVD-Net: Intervertebral Disc Localization and Segmentation in MRI with a Multi-modal UNet

  • Jose DolzEmail author
  • Christian Desrosiers
  • Ismail Ben Ayed
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11397)

Abstract

Accurate localization and segmentation of intervertebral disc (IVD) is crucial for the assessment of spine disease diagnosis. Despite the technological advances in medical imaging, IVD localization and segmentation are still manually performed, which is time-consuming and prone to errors. If, in addition, multi-modal imaging is considered, the burden imposed on disease assessments increases substantially. In this paper, we propose an architecture for IVD localization and segmentation in multi-modal magnetic resonance images (MRI), which extends the well-known UNet. Compared to single images, multi-modal data brings complementary information, contributing to better data representation and discriminative power. Our contributions are three-fold. First, how to effectively integrate and fully leverage multi-modal data remains almost unexplored. In this work, each MRI modality is processed in a different path to better exploit their unique information. Second, inspired by HyperDenseNet [11], the network is densely-connected both within each path and across different paths, granting the model the freedom to learn where and how the different modalities should be processed and combined. Third, we improved standard U-Net modules by extending inception modules [22] with two convolutional blocks with dilated convolutions of different scale, which helps handling multi-scale context. We report experiments over the data set of the public MICCAI 2018 Challenge on Automatic Intervertebral Disc Localization and Segmentation, with 13 multi-modal MRI images used for training and 3 for validation. We trained IVD-Net on an NVidia TITAN XP GPU with 16 GBs RAM, using ADAM as optimizer and a learning rate of 1\(\,\times \) 10\(^{-5}\) during 200 epochs. Training took about 5 h, and segmentation of a whole volume about 2–3 s, on average. Several baselines, with different multi-modal fusion strategies, were used to demonstrate the effectiveness of the proposed architecture.

Notes

Acknowledgments

This work is supported by the National Science and Engineering Research Council of Canada (NSERC), discovery grant program, and by the ETS Research Chair on Artificial Intelligence in Medical Imaging.

References

  1. 1.
    An, H.S., et al.: Introduction: disc degeneration: summary. Spine 29(23), 2677–2678 (2004)CrossRefGoogle Scholar
  2. 2.
    Ben Ayed, I., Punithakumar, K., Garvin, G., Romano, W., Li, S.: Graph cuts with invariant object-interaction priors: application to intervertebral disc segmentation. In: Székely, G., Hahn, H.K. (eds.) IPMI 2011. LNCS, vol. 6801, pp. 221–232. Springer, Heidelberg (2011).  https://doi.org/10.1007/978-3-642-22092-0_19 CrossRefGoogle Scholar
  3. 3.
    Aygün, M., Şahin, Y.H., Ünal, G.: Multi modal convolutional neural networks forbrain tumor segmentation. arXiv preprint arXiv:1809.06191 (2018)
  4. 4.
    Chen, C., et al.: Localization and segmentation of 3D intervertebral discs in MR images by data driven estimation. IEEE Trans. Med. Imaging 34(8), 1719–1729 (2015)CrossRefGoogle Scholar
  5. 5.
    Chen, H., Dou, Q., Wang, X., Qin, J., Cheng, J.C.Y., Heng, P.-A.: 3D fully convolutional networks for intervertebral disc localization and segmentation. In: Zheng, G., Liao, H., Jannin, P., Cattin, P., Lee, S.-L. (eds.) MIAR 2016. LNCS, vol. 9805, pp. 375–382. Springer, Cham (2016).  https://doi.org/10.1007/978-3-319-43775-0_34 CrossRefGoogle Scholar
  6. 6.
    Chen, L., Wu, Y., DSouza, A.M., Abidin, A.Z., Wismüller, A., Xu, C.: MRI tumor segmentation with densely connected 3D CNN. In: Medical Imaging 2018: Image Processing. International Society for Optics and Photonics (2018)Google Scholar
  7. 7.
    Chen, Y., Wang, H., Long, Y.: Regularization of convolutional neural networks using shufflenode. In: 2017 IEEE International Conference on Multimedia and Expo (ICME), pp. 355–360. IEEE (2017)Google Scholar
  8. 8.
    Dolz, J., Ben Ayed, I., Desrosiers, C.: Dense multi-path U-Net for ischemic stroke lesion segmentation in multiple image modalities. arXiv preprint arXiv:1810.07003 (2018)
  9. 9.
    Dolz, J., Ben Ayed, I., Yuan, J., Desrosiers, C.: Isointense infant brain segmentation with a hyper-dense connected convolutional neural network. In: 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018, pp. 616–620. IEEE (2018)Google Scholar
  10. 10.
    Dolz, J., Desrosiers, C., Wang, L., Yuan, J., Shen, D., Ben Ayed, I.: Deep CNN ensembles and suggestive annotations for infant brain MRI segmentation. arXiv preprint arXiv:1712.05319, 2017
  11. 11.
    Dolz, J., Gopinath, K., Yuan, J., Lombaert, H., Desrosiers, C., Ben Ayed, I.: HyperDense-Net: a hyper-densely connected CNN for multi-modal image segmentation. In: IEEE Transactions on Medical Imaging (2018, in press)Google Scholar
  12. 12.
    Hamanishi, C., Matukura, N., Fujita, M., Tomihara, M., Tanaka, S.: Cross-sectional area of the stenotic lumbar dural tube measured from the transverse views of magnetic resonance imaging. J. Spinal Disord. 7(5), 388–393 (1994)CrossRefGoogle Scholar
  13. 13.
    Huang, G., Liu, Z., Van Der Maaten, L., Weinberger, K.Q.: Densely connected convolutional networks. In: CVPR, vol. 1, p. 3 (2017)Google Scholar
  14. 14.
    Ji, X., Zheng, G., Belavy, D., Ni, D.: Automated intervertebral disc segmentation using deep convolutional neural networks. In: Yao, J., Vrtovec, T., Zheng, G., Frangi, A., Glocker, B., Li, S. (eds.) CSI 2016. LNCS, vol. 10182, pp. 38–48. Springer, Cham (2016).  https://doi.org/10.1007/978-3-319-55050-3_4 CrossRefGoogle Scholar
  15. 15.
    Kamnitsas, K., et al.: Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation. Med. Image Anal. 36, 61–78 (2017)CrossRefGoogle Scholar
  16. 16.
    Kim, S., Bae, W., Masuda, K., Chung, C., Hwang, D.: Fine-grain segmentation of the intervertebral discs from MR spine images using deep convolutional neural networks: BSU-Net. Appl. Sci. 8(9), 1656 (2018)CrossRefGoogle Scholar
  17. 17.
    Li, X., et al.: 3D multi-scale FCN with random modality voxel dropout learning for intervertebral disc localization and segmentation from multi-modality MR images. Med. Image Anal. 45, 41–54 (2018)CrossRefGoogle Scholar
  18. 18.
    Moeskops, P., Viergever, M.A., Mendrik, A.M., de Vries, L.S., Benders, M.J., Išgum, I.: Automatic segmentation of MR brain images with a convolutional neural network. IEEE Trans. Med. Imaging 35(5), 1252–1261 (2016)CrossRefGoogle Scholar
  19. 19.
    Nie, D., Wang, L., Gao, Y., Sken, D.: Fully convolutional networks for multi-modality isointense infant brain image segmentation. In: 13th International Symposium on Biomedical Imaging (ISBI), pp. 1342–1345. IEEE (2016)Google Scholar
  20. 20.
    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_28 CrossRefGoogle Scholar
  21. 21.
    Srivastava, N., Salakhutdinov, R.: Multimodal learning with deep boltzmann machines. J. Mach. Learn. Res. 15, 2949–2980 (2014)MathSciNetzbMATHGoogle Scholar
  22. 22.
    Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., Wojna, Z.: Rethinking the inception architecture for computer vision. In: CVPR, pp. 2818–2826 (2016)Google Scholar
  23. 23.
    Valverde, S., et al.: Improving automated multiple sclerosis lesion segmentation with a cascaded 3D convolutional neural network approach. NeuroImage 155, 159–168 (2017)CrossRefGoogle Scholar
  24. 24.
    Wieser, S., et al.: Cost of low back pain in switzerland in 2005. Eur. J. Health Econ. 12(5), 455–467 (2011)CrossRefGoogle Scholar
  25. 25.
    Yu, F., Koltun, V.: Multi-scale context aggregation by dilated convolutions. arXiv preprint arXiv:1511.07122 (2015)
  26. 26.
    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_33 CrossRefGoogle Scholar
  27. 27.
    Zeng, G., Zheng, G.: DSMS-FCN: a deeply supervised multi-scale fully convolutional network for automatic segmentation of intervertebral disc in 3D MR images. In: Glocker, B., Yao, J., Vrtovec, T., Frangi, A., Zheng, G. (eds.) MSKI 2017. LNCS, vol. 10734, pp. 148–159. Springer, Cham (2018).  https://doi.org/10.1007/978-3-319-74113-0_13 CrossRefGoogle Scholar
  28. 28.
    Zhang, T., Qi, G.-J., Xiao, B., Wang, J.: Interleaved group convolutions. In: CVPR, pp. 4373–4382 (2017)Google Scholar
  29. 29.
    Zhang, W., et al.: Deep convolutional neural networks for multi-modality isointense infant brain image segmentation. NeuroImage 108, 214–224 (2015)CrossRefGoogle Scholar
  30. 30.
    Zhang, X., Zhou, X., Lin, M., Sun, J.: Shufflenet: an extremely efficient convolutional neural network for mobile devices. arXiv preprint arXiv:1707.01083 (2017)
  31. 31.
    Zheng, G., et al.: Evaluation and comparison of 3D intervertebral disc localization and segmentation methods for 3D T2 MR data: a grand challenge. Med. Image Anal. 35, 327–344 (2017)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Jose Dolz
    • 1
    Email author
  • Christian Desrosiers
    • 1
  • Ismail Ben Ayed
    • 1
  1. 1.ETS MontrealMontrealCanada

Personalised recommendations