GridNet with Automatic Shape Prior Registration for Automatic MRI Cardiac Segmentation

  • Clément Zotti
  • Zhiming Luo
  • Olivier Humbert
  • Alain Lalande
  • Pierre-Marc Jodoin
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10663)


In this paper, we propose a fully automatic MRI cardiac segmentation method based on a novel deep convolutional neural network (CNN) designed for the 2017 ACDC MICCAI challenge. The novelty of our network comes with its embedded shape prior and its loss function tailored to the cardiac anatomy. Our model includes a cardiac center-of-mass regression module which allows for an automatic shape prior registration. Also, since our method processes raw MR images without any manual preprocessing and/or image cropping, our CNN learns both high-level features (useful to distinguish the heart from other organs with a similar shape) and low-level features (useful to get accurate segmentation results). Those features are learned with a multi-resolution conv-deconv “grid” architecture which can be seen as an extension of the U-Net.

Experimental results reveal that our method can segment the left and right ventricles as well as the myocardium from a 3D MRI cardiac volume in 0.4 s with an average Dice coefficient of 0.90 and an average Hausdorff distance of \(10.4\) mm.


Convolutional neural networks MRI Heart Segmentation 


  1. 1.
    Epstein, F.H.: MRI of left ventricular function. J Nucl. Cardiol 14(5), 729–744 (2007)CrossRefGoogle Scholar
  2. 2.
    Vick, G.W.: The gold standard for noninvasive imaging in coronary heart disease: magnetic resonance imaging. Curr. opin. cardiol. 24(6), 567–579 (2009)CrossRefGoogle Scholar
  3. 3.
    Peng, P., et al.: A review of heart chamber segmentation for structural and functional analysis using cardiac magnetic resonance imaging. MAGMA 29(2), 155–195 (2016)MathSciNetCrossRefGoogle Scholar
  4. 4.
    Petitjean, C., et al.: Right ventricle segmentation from cardiac MRI: a collation study. Med. Image Anal. 19(1), 187–202 (2015)CrossRefGoogle Scholar
  5. 5.
    Auger, D.A., et al.: Semi-automated left ventricular segmentation based on a guide point model approach for 3D cine DENSE cardiovascular magnetic resonance. J. Cardiovasc. Magn. Reson. 16(1), 8 (2014)CrossRefGoogle Scholar
  6. 6.
    Grosgeorge, D., Petitjean, C., Dacher, J.-N., Ruan, S.: Graph cut segmentation with a statistical shape model in cardiac MRI. CVIU 117(9), 1027–1035 (2013)Google Scholar
  7. 7.
    Petitjean, C., Dacher, J.: A review of segmentation methods in short axis cardiac MR images. Med. Image Anal. 15(2), 169–184 (2011)CrossRefGoogle Scholar
  8. 8.
    Wang, L., Pei, M., Codella, N.C.F., et al.: Left ventricle: fully automated segmentation based on spatiotemporal continuity and myocardium information in cine cardiac magnetic resonance imaging (LV-FAST). BioMed Res. Int. 2015, 9 (2015). Article ID 367583Google Scholar
  9. 9.
    Liu, Y., Captur, G., et al.: Distance regularized two level sets for segmentation of left and right ventricles from cine-MRI. Magn. Reson. Img. 34(5), 699–706 (2016)CrossRefGoogle Scholar
  10. 10.
    Tran, P.V.: A fully convolutional neural network for cardiac segmentation in short-axis MRI. arXiv preprint arXiv:1604.00494 (2016)
  11. 11.
    Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. In: Proceedings of CVPR, pp. 3431–3440 (2015)Google Scholar
  12. 12.
    Noh, H., Hong, S., Han, S.: Learning deconvolution network for semantic segmentation. In: Proceedings of ICCV (2015)Google Scholar
  13. 13.
    Ronneberger, O., Fischer, P., Brox, T.: U-net: Convolutional networks for biomedical image segmentation. In: Proceedings of MICCAI, pp. 234–241 (2015)Google Scholar
  14. 14.
    Tan, L.K., et al.: Cardiac left ventricle segmentation using convolutional neural network regression. In: Proceedings of IECBES, pp. 490–493. IEEE (2016)Google Scholar
  15. 15.
    Ngo, T.A., Lu, Z., Carneiro, G.: Combining deep learning and level set for the automated segmentation of the left ventricle of the heart from cardiac cine magnetic resonance. Med. Image Anal. 35(1), 159–171 (2017)CrossRefGoogle Scholar
  16. 16.
    Kastler, B.: Cardiovascular anatomy and atlas of MR normal anatomy. MRI of Cardiovascular Malformations, pp. 17–39. Springer, Heidelberg (2011). CrossRefGoogle Scholar
  17. 17.
    ACDC-MICCAI challenge.
  18. 18.
    Tavakoli, V., Amini, A.A.: A survey of shaped-based registration and segmentation techniques for cardiac images. CVIU 117(9), 966–989 (2013)Google Scholar
  19. 19.
    Luo, Z., Mishra, A., Achkar, A., Eichel, J., Li, S.-Z., Jodoin, P.-M.: Non-local deep features for salient object detection. In: proceeding of CVPR (2017)Google Scholar
  20. 20.
    Srivastava, N., Hinton, G., et al.: Dropout: a simple way to prevent neural networks from overfitting. J. of Mach. Learn. Res. 15, 1929–1958 (2014)MathSciNetMATHGoogle Scholar
  21. 21.
    Kingma, D., Ba, J.: Adam: a method for stochastic optimization. In: Proceedings of ICLR (2015)Google Scholar
  22. 22.
    Zou, K.H., et al.: Statistical validation of image segmentation quality based on a spatial overlap index 1: scientific reports. Acad. rad. 11(2), 178–189 (2004)CrossRefGoogle Scholar
  23. 23.
    Huttenlocher, D., Klanderman, G., Rucklidge, W.J.: Comparing images using the Hausdorff distance. IEEE Trans PAMI 15(9), 850–863 (1993)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Computer Science DepartmentUniversité de SherbrookeSherbrookeCanada
  2. 2.Le2i, Université de Bourgogne Franche-ComtéDijonFrance
  3. 3.Department of Nuclear Medicine, Centre Antoine LacassagneNiceFrance
  4. 4.Cognitive Science DepartmentXiamen UniversityXiamenChina

Personalised recommendations