Medical-based Deep Curriculum Learning for Improved Fracture Classification

  • Amelia Jiménez-SánchezEmail author
  • Diana Mateus
  • Sonja Kirchhoff
  • Chlodwig Kirchhoff
  • Peter Biberthaler
  • Nassir Navab
  • Miguel A. González Ballester
  • Gemma Piella
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11769)


Current deep-learning based methods do not easily integrate to clinical protocols, neither take full advantage of medical knowledge. In this work, we propose and compare several strategies relying on curriculum learning, to support the classification of proximal femur fracture from X-ray images, a challenging problem as reflected by existing intra- and inter-expert disagreement. Our strategies are derived from knowledge such as medical decision trees and inconsistencies in the annotations of multiple experts, which allows us to assign a degree of difficulty to each training sample. We demonstrate that if we start learning “easy” examples and move towards “hard”, the model can reach a better performance, even with fewer data. The evaluation is performed on the classification of a clinical dataset of about 1000 X-ray images. Our results show that, compared to class-uniform and random strategies, the proposed medical knowledge-based curriculum, performs up to 15% better in terms of accuracy, achieving the performance of experienced trauma surgeons.


Curriculum learning Multi-label classification Bone fractures Computer-aided diagnosis Medical decision trees 



This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 713673 and by the Spanish Ministry of Economy [MDM-2015-0502]. A. Jiménez-Sánchez has received financial support through the “la Caixa” Foundation (ID Q5850017D), fellowship code: LCF/BQ/IN17/11620013. D. Mateus has received funding from Nantes Métropole and the European Regional Development, Pays de la Loire, under the Connect Talent scheme. Authors thank Nvidia for the donation of a GPU.

Supplementary material

490281_1_En_77_MOESM1_ESM.pdf (170 kb)
Supplementary material 1 (pdf 170 KB)


  1. 1.
    Elman, J.L.: Learning and development in neural networks: the importance of starting small. Cognition 48(1), 71–99 (1993)CrossRefGoogle Scholar
  2. 2.
    Bengio, Y., Louradour, J., Collobert, R., Weston, J.: Curriculum learning. In: Proceedings of the 26th Annual International Conference on Machine Learning ICML 2009, pp. 41–48. ACM, New York (2009)Google Scholar
  3. 3.
    Havaei, M., Guizard, N., Chapados, N., Bengio, Y.: HeMIS: hetero-modal image segmentation. In: Ourselin, S., Joskowicz, L., Sabuncu, M.R., Unal, G., Wells, W. (eds.) MICCAI 2016. LNCS, vol. 9901, pp. 469–477. Springer, Cham (2016). Scholar
  4. 4.
    Maicas, G., Bradley, A.P., Nascimento, J.C., Reid, I., Carneiro, G.: Training medical image analysis systems like radiologists. In: Frangi, A.F., Schnabel, J.A., Davatzikos, C., Alberola-López, C., Fichtinger, G. (eds.) MICCAI 2018. LNCS, vol. 11070, pp. 546–554. Springer, Cham (2018). Scholar
  5. 5.
    Tang, Y., Wang, X., Harrison, A.P., Lu, L., Xiao, J., Summers, R.M.: Attention-guided curriculum learning for weakly supervised classification and localization of thoracic diseases on chest radiographs. In: Shi, Y., Suk, H.-I., Liu, M. (eds.) MLMI 2018. LNCS, vol. 11046, pp. 249–258. Springer, Cham (2018). Scholar
  6. 6.
    Kellam, J.F., Meinberg, E.G., Agel, J., Karam, M.D., Roberts, C.S.: Introduction. J. Orthop. Trauma 32, S1–S10 (2018)CrossRefGoogle Scholar
  7. 7.
    Moran, C.G., Wenn, R.T., Sikand, M., Taylor, A.M.: Early mortality after hip fracture: is delay before surgery important? JBJS 87(3), 483–489 (2005)Google Scholar
  8. 8.
    van Embden, D., Rhemrev, S., Meylaerts, S., Roukema, G.: The comparison of two classifications for trochanteric femur fractures: the AO/ASIF classification and the Jensen classification. Injury 41(4), 377–381 (2010)CrossRefGoogle Scholar
  9. 9.
    Kazi, A., Albarqouni, S., Sanchez, A.J., Kirchhoff, S., Biberthaler, P., Navab, N., Mateus, D.: Automatic classification of proximal femur fractures based on attention models. In: Wang, Q., Shi, Y., Suk, H.-I., Suzuki, K. (eds.) MLMI 2017. LNCS, vol. 10541, pp. 70–78. Springer, Cham (2017). Scholar
  10. 10.
    Jesson, A., Guizard, N., Ghalehjegh, S.H., Goblot, D., Soudan, F., Chapados, N.: CASED: curriculum adaptive sampling for extreme data imbalance. In: Descoteaux, M., Maier-Hein, L., Franz, A., Jannin, P., Collins, D.L., Duchesne, S. (eds.) MICCAI 2017. LNCS, vol. 10435, pp. 639–646. Springer, Cham (2017). Scholar
  11. 11.
    Wang, W., Lu, Y., Wu, B., Chen, T., Chen, D.Z., Wu, J.: Deep active self-paced learning for accurate pulmonary nodule segmentation. In: Frangi, A.F., Schnabel, J.A., Davatzikos, C., Alberola-López, C., Fichtinger, G. (eds.) MICCAI 2018. LNCS, vol. 11071, pp. 723–731. Springer, Cham (2018). Scholar
  12. 12.
    Ren, Z., Dong, D., Li, H., Chen, C.: Self-paced prioritized curriculum learning with coverage penalty in deep reinforcement learning. IEEE Trans. Neural Netw. Learn. Syst. 29(6), 2216–2226 (2018). Scholar
  13. 13.
    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 (2016)Google Scholar
  14. 14.
    Wang, Y., Gan, W., Wu, W., Yan, J.: Dynamic curriculum learning for imbalanced data classification. CoRR abs/1901.06783 (2019)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.BCN MedTech, DTICPompeu Fabra UniversityBarcelonaSpain
  2. 2.Ecole Centrale de Nantes, LS2N, UMR CNRS 6004NantesFrance
  3. 3.Institute of Clinical RadiologyLMU MünchenMunichGermany
  4. 4.Department of Trauma Surgery, Klinikum rechts der IsarTechnische Universität MünchenMunichGermany
  5. 5.Computer Aided Medical ProceduresTechnische Universität MünchenMunichGermany
  6. 6.ICREABarcelonaSpain

Personalised recommendations