A Multitask Learning Architecture for Simultaneous Segmentation of Bright and Red Lesions in Fundus Images

  • Clément PlayoutEmail author
  • Renaud Duval
  • Farida Cheriet
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11071)


Recent CNN architectures have established state-of-the-art results in a large range of medical imaging applications. We propose an extension to the U-Net architecture relying on multi-task learning: while keeping a single encoding module, multiple decoding modules are used for concurrent segmentation tasks. We propose improvements of the encoding module based on the latest CNN developments: residual connections at every scale, mixed pooling for spatial compression and large kernels for convolutions at the lowest scale. We also use dense connections within the different scales based on multi-size pooling regions. We use this new architecture to jointly detect and segment red and bright retinal lesions which are essential biomarkers of diabetic retinopathy. Each of the two categories is handled by a specialized decoding module. Segmentation outputs are refined with conditional random fields (CRF) as RNN and the network is trained end-to-end with an effective Kappa-based function loss. Preliminary results on a public dataset in the segmentation task on red (resp. bright) lesions shows a sensitivity of 66,9% (resp. 75,3%) and a specificity of 99,8% (resp. 99,9%).


  1. 1.
    Seoud, L., Hurtut, T., Chelbi, J., Cheriet, F., Langlois, J.M.P.: Red lesion detection using dynamic shape features for diabetic retinopathy screening. IEEE Trans. Med. Imaging 35(4), 1116–1126 (2016)CrossRefGoogle Scholar
  2. 2.
    van Grinsven, M.J.J.P., van Ginneken, B., Hoyng, C.B., Theelen, T., Sánchez, C.I.: Fast convolutional neural network training using selective data sampling: application to hemorrhage detection in color fundus images. IEEE Trans. Med. Imaging 35(5), 1273–1284 (2016)CrossRefGoogle Scholar
  3. 3.
    Vanithamani, R., Renee Christina, R.: Exudates in detection and classification of diabetic retinopathy. In: Abraham, A., Cherukuri, A.K., Madureira, A.M., Muda, A.K. (eds.) SoCPaR 2016. AISC, vol. 614, pp. 252–261. Springer, Cham (2018). Scholar
  4. 4.
    Moeskops, P., et al.: Deep learning for multi-task medical image segmentation in multiple modalities. In: Ourselin, S., Joskowicz, L., Sabuncu, M.R., Unal, G., Wells, W. (eds.) MICCAI 2016. LNCS, vol. 9901, pp. 478–486. Springer, Cham (2016). Scholar
  5. 5.
    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). Scholar
  6. 6.
    Caruana, R.: Multitask learning. Mach. Learn. 28(1), 41–75 (1997)MathSciNetCrossRefGoogle Scholar
  7. 7.
    Yu, D., Wang, H., Chen, P., Wei, Z.: Mixed pooling for convolutional neural networks. In: Miao, D., Pedrycz, W., Ślȩzak, D., Peters, G., Hu, Q., Wang, R. (eds.) RSKT 2014. LNCS (LNAI), vol. 8818, pp. 364–375. Springer, Cham (2014). Scholar
  8. 8.
    He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778, June 2016Google Scholar
  9. 9.
    Peng, C., Zhang, X., Yu, G., Luo, G., Sun, J.: Large kernel matters - improve semantic segmentation by global convolutional network. CoRR abs/1703.02719 (2017)Google Scholar
  10. 10.
    Krähenbühl, P., Koltun, V.: Efficient inference in fully connected CRFs with Gaussian edge potentials. CoRR abs/1210.5644 (2012)Google Scholar
  11. 11.
    Zheng, S., et al.: Conditional random fields as recurrent neural networks. In: 2015 IEEE International Conference on Computer Vision (ICCV), pp. 1529–1537, December 2015Google Scholar
  12. 12.
    Zeiler, M.D.: ADADELTA: an adaptive learning rate method. CoRR abs/1212.5701 (2012)Google Scholar
  13. 13.
    Kauppi, T., et al.: DIARETDB1 diabetic retinopathy database and evaluation protocol (01 2007)Google Scholar
  14. 14.
    Decenciàre, E., et al.: TeleOphta: machine learning and image processing methods for teleophthalmology. IRBM 34(2), 196–203 (2013). Special issue: ANR TECSAN: Technologies for Health and AutonomyCrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  • Clément Playout
    • 1
    Email author
  • Renaud Duval
    • 2
  • Farida Cheriet
    • 1
  1. 1.LIV4D, École Polytechnique de MontréalMontrealCanada
  2. 2.CUO-Hôpital Maisonneuve RosemontMontrealCanada

Personalised recommendations