Deep Learning from Label Proportions for Emphysema Quantification

  • Gerda BortsovaEmail author
  • Florian Dubost
  • Silas Ørting
  • Ioannis Katramados
  • Laurens Hogeweg
  • Laura Thomsen
  • Mathilde Wille
  • Marleen de Bruijne
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11071)


We propose an end-to-end deep learning method that learns to estimate emphysema extent from proportions of the diseased tissue. These proportions were visually estimated by experts using a standard grading system, in which grades correspond to intervals (label example: 1–5% of diseased tissue). The proposed architecture encodes the knowledge that the labels represent a volumetric proportion. A custom loss is designed to learn with intervals. Thus, during training, our network learns to segment the diseased tissue such that its proportions fit the ground truth intervals. Our architecture and loss combined improve the performance substantially (8% ICC) compared to a more conventional regression network. We outperform traditional lung densitometry and two recently published methods for emphysema quantification by a large margin (at least 7% AUC and 15% ICC), and achieve near-human-level performance. Moreover, our method generates emphysema segmentations that predict the spatial distribution of emphysema at human level.


Emphysema quantification Weak labels Multiple instance learning Learning from label proportions 



This research is financed by the Netherlands Organization for Scientific Research (NWO) and COSMONiO.


  1. 1.
    Bos, D., et al.: Intracranial carotid artery atherosclerosis and the risk of stroke in whites: the Rotterdam study. JAMA Neurol. 71(4), 405–411 (2014)CrossRefGoogle Scholar
  2. 2.
    De Jong, P.A., Tiddens, H.A.: Cystic fibrosis-specific computed tomography scoring. Proc. Am. Thorac. Soc. 4(4), 338–342 (2007)CrossRefGoogle Scholar
  3. 3.
    Dery, L.M., Nachman, B., Rubbo, F., Schwartzman, A.: Weakly supervised classification in high energy physics. JHEP 2017(5), 145 (2017)CrossRefGoogle Scholar
  4. 4.
    Dubost, F., et al.: GP-Unet: lesion detection from weak labels with a 3D regression network. In: Descoteaux, M., Maier-Hein, L., Franz, A., Jannin, P., Collins, D.L., Duchesne, S. (eds.) MICCAI 2017. LNCS, vol. 10435, pp. 214–221. Springer, Cham (2017). Scholar
  5. 5.
    He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR 2016, pp. 770–778. IEEE (2016)Google Scholar
  6. 6.
    Li, F., Taylor, G.: Alter-CNN: an approach to learning from label proportions with application to ice-water classification. In: NIPSW (2015)Google Scholar
  7. 7.
    Ørting, S.N., Petersen, J., Thomsen, L.H., Wille, M.M.W., De Bruijne, M.: Detecting emphysema with multiple instance learning. In: ISBI (2018)Google Scholar
  8. 8.
    Ørting, S.N., Petersen, J., Wille, M.M.W., Thomsen, L.H., De Bruijne, M.: Quantifying emphysema extent from weakly labeled CT scans of the lungs using label proportions learning. In: Proceedings of Sixth International Workshop on Pulmonary Image Analysis (2016)Google Scholar
  9. 9.
    Patrini, G., Nock, R., Rivera, P., Caetano, T.: (Almost) No Label No Cry. NIPS 2014 (c), 1–9 (2014)Google Scholar
  10. 10.
    Wang, X., Yan, Y., Tang, P., Bai, X., Liu, W.: Revisiting multiple instance neural networks. Pattern Recognit. 74, 15–24 (2018)CrossRefGoogle Scholar
  11. 11.
    Wille, M.M.W., et al.: Visual assessment of early emphysema and interstitial abnormalities on CT is useful in lung cancer risk analysis. Eur. Radiol. 26(2), 487–494 (2016)CrossRefGoogle Scholar
  12. 12.
    Wille, M.M.W., Thomsen, L.H., Dirksen, A., Petersen, J., Pedersen, J.H., Shaker, S.B.: Emphysema progression is visually detectable in low-dose CT in continuous but not in former smokers. Eur. Radiol. 24(11), 2692–2699 (2014)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  • Gerda Bortsova
    • 1
    Email author
  • Florian Dubost
    • 1
  • Silas Ørting
    • 2
  • Ioannis Katramados
    • 3
  • Laurens Hogeweg
    • 3
  • Laura Thomsen
    • 4
  • Mathilde Wille
    • 5
  • Marleen de Bruijne
    • 1
    • 2
  1. 1.Biomedical Imaging Group RotterdamErasmus MCRotterdamThe Netherlands
  2. 2.Department of Computer ScienceUniversity of CopenhagenCopenhagenDenmark
  3. 3.COSMONiOGroningenThe Netherlands
  4. 4.Department of Respiratory MedicineHvidovre HospitalHvidovreDenmark
  5. 5.Department of Diagnostic ImagingBispebjerg HospitalCopenhagenDenmark

Personalised recommendations