Advertisement

GP-Unet: Lesion Detection from Weak Labels with a 3D Regression Network

  • Florian DubostEmail author
  • Gerda Bortsova
  • Hieab Adams
  • Arfan Ikram
  • Wiro J. Niessen
  • Meike Vernooij
  • Marleen De Bruijne
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10435)

Abstract

We propose a novel convolutional neural network for lesion detection from weak labels. Only a single, global label per image - the lesion count - is needed for training. We train a regression network with a fully convolutional architecture combined with a global pooling layer to aggregate the 3D output into a scalar indicating the lesion count. When testing on unseen images, we first run the network to estimate the number of lesions. Then we remove the global pooling layer to compute localization maps of the size of the input image. We evaluate the proposed network on the detection of enlarged perivascular spaces in the basal ganglia in MRI. Our method achieves a sensitivity of \(62\%\) with on average 1.5 false positives per image. Compared with four other approaches based on intensity thresholding, saliency and class maps, our method has a \(20\%\) higher sensitivity.

Notes

Acknowledgments

This research was funded by The Netherlands Organisation for Health Research and Development (ZonMw) Project 104003005.

References

  1. 1.
    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). doi: 10.1007/978-3-319-46723-8_49 CrossRefGoogle Scholar
  2. 2.
    Chen, H., Dou, Q., Yu, L., Qin, J., Heng, P.A.: VoxResNet: deep voxelwise residual networks for volumetric brain segmentation. NeuroImage (2017)Google Scholar
  3. 3.
    Bortsova, G., van Tulder, G., Dubost, F., Peng, T., Navab, N., van der Lugt, A., Bos, D., de Bruijne, M.: Segmentation of intracranial arterial calcification with deeply supervised residual dropout networks. In: Descoteaux, M., et al. (eds.) MICCAI 2017, Part III. LNCS, vol. 10435, pp. 359–367. Springer, Cham (2017). doi: 10.1007/978-3-319-66179-7_41 Google Scholar
  4. 4.
    Long, J., Shelhamer, E. and Darrell, T.: Fully convolutional networks for semantic segmentation. In: CVPR (2015)Google Scholar
  5. 5.
    Zhou, B., Khosla, A., Lapedriza, A., Oliva, A. and Torralba, A.: Learning deep features for discriminative localization. In: CVPR (2016)Google Scholar
  6. 6.
    Hwang, S., Kim, H.-E.: Self-transfer learning for weakly supervised lesion localization. In: Ourselin, S., Joskowicz, L., Sabuncu, M.R., Unal, G., Wells, W. (eds.) MICCAI 2016. LNCS, vol. 9901, pp. 239–246. Springer, Cham (2016). doi: 10.1007/978-3-319-46723-8_28 CrossRefGoogle Scholar
  7. 7.
    Ronneberger, O., Fischer, P., Brox, T.: Convolutional networks for biomedical image segmentation. In: MICCAI, pp. 234–241 (2015)Google Scholar
  8. 8.
    Ramirez, J., Berezuk, C., McNeely, A.A., Scott, C.J., Gao, F., Black, S.E.: Visible Virchow-Robin spaces on magnetic resonance imaging of Alzheimer’s disease patients and normal elderly from the Sunnybrook Dementia Study. J. Alzheimers Dis. 43(2), 415–424 (2015)CrossRefGoogle Scholar
  9. 9.
    Park, S.H., Zong, X., Gao, Y., Lin, W., Shen, D.: Segmentation of perivascular spaces in 7T MR image using auto-context model with orientation-normalized features. NeuroImage 134, 223–235 (2016)CrossRefGoogle Scholar
  10. 10.
    Ballerini, L., Lovreglio, R., Hernandez, M., del C. Valdés Hernández, M., Maniega, S.M., Pellegrini, E., Wardlaw, J.M.: Application of the ordered logit model to optimising frangi filter parameters for segmentation of perivascular spaces. Procedia Comput. Sci. 90, 6167 (2016)CrossRefGoogle Scholar
  11. 11.
    Desikan, R.S., Sgonne, F., Fischl, B., Quinn, B.T., Dickerson, B.C., Blacker, D., Buckner, R.L., Dale, A.M., Maguire, R.P., Hyman, B.T., Albert, M.S.: An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. Neuroimage 31(3), 968–980 (2006)CrossRefGoogle Scholar
  12. 12.
    Lin, M., Chen, Q., Yan, S.: Network in network. In: ICLR (2014)Google Scholar
  13. 13.
    Hofman, A., Brusselle, G.G., Murad, S.D., van Duijn, C.M., Franco, O.H., Goedegebure, A., Ikram, M.A., Klaver, C.C., Nijsten, T.E., Peeters, R.P., Stricker, B.H.C.: The Rotterdam Study: 2016 objectives and design update. Eur. J. Epidemiol. 30(8), 661–708 (2015)CrossRefGoogle Scholar
  14. 14.
    Klein, S., Staring, M., Murphy, K., Viergever, M.A., Pluim, J.P.W.: Elastix: a toolbox for intensity based medical image registration. TMI 29(1), 196–205 (2010)Google Scholar
  15. 15.
    Zeiler, M.D.: ADADELTA: an adaptive learning rate method. arXiv preprint. arxiv:1212.5701
  16. 16.
    Simonyan, K., Vedaldi, A., Zisserman, A.: Deep inside convolutional networks: Visualising image classification models and saliency maps. In: CVPR (2014)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Florian Dubost
    • 1
    • 2
    • 3
    Email author
  • Gerda Bortsova
    • 1
    • 2
    • 3
  • Hieab Adams
    • 3
    • 4
  • Arfan Ikram
    • 3
    • 4
    • 5
  • Wiro J. Niessen
    • 1
    • 2
    • 3
    • 6
  • Meike Vernooij
    • 3
    • 4
  • Marleen De Bruijne
    • 1
    • 2
    • 3
    • 7
  1. 1.Biomedical Imaging Group RotterdamErasmus MCRotterdamThe Netherlands
  2. 2.Department of Medical InformaticsErasmus MCRotterdamThe Netherlands
  3. 3.Department of RadiologyErasmus MCRotterdamThe Netherlands
  4. 4.Department of EpidemiologyErasmus MCRotterdamThe Netherlands
  5. 5.Department of NeurologyErasmus MCRotterdamThe Netherlands
  6. 6.Imaging Physics, Faculty of Applied SciencesTU DelftDelftThe Netherlands
  7. 7.Department of Computer ScienceUniversity of CopenhagenCopenhagenDenmark

Personalised recommendations