Learning to Segment Skin Lesions from Noisy Annotations

  • Zahra MirikharajiEmail author
  • Yiqi Yan
  • Ghassan Hamarneh
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11795)


Deep convolutional neural networks have driven substantial advancements in the automatic understanding of images. Requiring a large collection of images and their associated annotations is one of the main bottlenecks limiting the adoption of deep networks. In the task of medical image segmentation, requiring pixel-level semantic annotations performed by human experts exacerbate this difficulty. This paper proposes a new framework to train a fully convolutional segmentation network from a large set of cheap unreliable annotations and a small set of expert-level clean annotations. We propose a spatially adaptive reweighting approach to treat clean and noisy pixel-level annotations commensurately in the loss function. We deploy a meta-learning approach to assign higher importance to pixels whose loss gradient direction is closer to those of clean data. Our experiments on training the network using segmentation ground truth corrupted with different levels of annotation noise show how spatial reweighting improves the robustness of deep networks to noisy annotations.



We thank NVIDIA Corporation for the donation of Titan X GPUs used in this research and Compute Canada for computational resources.


  1. 1.
    Codella, et al.: Skin lesion analysis toward melanoma detection: a challenge at the 2017 ISBI. arXiv:1710.05006 (2017)
  2. 2.
    Goldberger, J., Ben-Reuven, E.: Training deep neural-networks using a noise adaptation layer. In: ICLR (2017)Google Scholar
  3. 3.
    Ibtehaz, N., Rahman, M. S.: Multiresunet: rethinking the U-Net architecture for multimodal biomedical image segmentation. arXiv preprint arXiv:1902.04049 (2019)
  4. 4.
    Jiang, et al.: Mentornet: regularizing very deep neural networks on corrupted labels. arXiv preprint arXiv:1712.05055 (2017)
  5. 5.
    Long, et al.: Fully convolutional networks for semantic segmentation. In: IEEE CVPR, pp. 3431–3440 (2015)Google Scholar
  6. 6.
    Mirikharaji, Z., Hamarneh, G.: Star shape prior in fully convolutional networks for skin lesion segmentation. In: Frangi, A.F., Schnabel, J.A., Davatzikos, C., Alberola-López, C., Fichtinger, G. (eds.) MICCAI 2018. LNCS, vol. 11073, pp. 737–745. Springer, Cham (2018). Scholar
  7. 7.
    Oktay, et al.: Attention U-Net: learning where to look for the pancreas. In: MIDL (2018)Google Scholar
  8. 8.
    Ren, et al.: Learning to reweight examples for robust deep learning. In: ICML (2018)Google Scholar
  9. 9.
    Ronneberger, et al.: 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).
  10. 10.
    Siegel, et al.: Cancer statistics. CA. Cancer J. Clin. 67(1), 7–30 (2017)Google Scholar
  11. 11.
    Sukhbaatar, et al.: Learning from noisy labels with deep neural networks. arXiv preprint arXiv:1406.2080 (2014)
  12. 12.
    Vahdat, A.: Toward robustness against label noise in training deep discriminative neural networks. In: NIPS, pp. 5596–5605 (2017)Google Scholar
  13. 13.
    Veit, et al.: Learning from noisy large-scale datasets with minimal supervision. In: IEEE CVPR, pp. 839–847 (2017)Google Scholar
  14. 14.
    Venkatesh, et al.: A deep residual architecture for skin lesion segmentation. In: OR 2.0 Context-Aware Operating Theaters, Computer Assisted Robotic Endoscopy, Clinical Image-Based Procedures, and Skin Image Analysis, pp. 277–284 (2018)Google Scholar
  15. 15.
    Wang, et al.: Iterative learning with open-set noisy labels. In: IEEE CVPR, pp. 8688–8696 (2018)Google Scholar
  16. 16.
    Xiao, et al.: Learning from massive noisy labeled data for image classification. In: IEEE CVPR, pp. 2691–2699 (2015)Google Scholar
  17. 17.
    Yuan, et al.: Automatic skin lesion segmentation using deep fully convolutional networks with jaccard distance. IEEE TMI 36(9), 1876–1886 (2017)Google Scholar
  18. 18.
    Zhang, et al.: Understanding deep learning requires rethinking generalization. In: ICLR (2017)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.School of Computing ScienceSimon Fraser UniversityBurnabyCanada

Personalised recommendations