Multiclass Deep Active Learning for Detecting Red Blood Cell Subtypes in Brightfield Microscopy

  • Ario Sadafi
  • Niklas Koehler
  • Asya Makhro
  • Anna Bogdanova
  • Nassir Navab
  • Carsten MarrEmail author
  • Tingying PengEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11764)


The recent success of deep learning approaches relies partly on large amounts of well annotated training data. For natural images object annotation is easy and cheap. For biomedical images however, annotation crucially depends on the availability of a trained expert whose time is typically expensive and scarce. To ensure efficient annotation, only the most relevant objects should be presented to the expert. Currently, no approach exists that allows to select those for a multiclass detection problem. Here, we present an active learning framework that identifies the most relevant samples from a large set of not annotated data for further expert annotation. Applied to brightfield images of red blood cells with seven subtypes, we train a faster R-CNN for single cell identification and classification, calculate a novel confidence score using dropout variational inference and select relevant images for annotation based on (i) the confidence of the single cell detection and (ii) the rareness of the classes contained in the image. We show that our approach leads to a drastic increase of prediction accuracy with already few annotated images. Our original approach improves classification of red blood cell subtypes and speeds up the annotation. This important step in diagnosing blood diseases will profit from our framework as well as many other clinical challenges that suffer from the lack of annotated training data.


Active learning Multiclass annotation Single cell microscopy 



CM acknowledges support from the DFG funded SFB1243 (A09) and the “MicMode-I2T” project (FKZ#017x1710D) funded by the Federal Ministry of Education and Research of Germany (BMBF), DLR project management, in the “e:Med” framework program.


  1. 1.
    Andolfo, I., et al.: Multiple clinical forms of dehydrated hereditary stomatocytosis arise from mutations in piezo1. Blood 121(19), 3925–3935 (2013)CrossRefGoogle Scholar
  2. 2.
    Coudray, N., et al.: Classification and mutation prediction from non-small cell lung cancer histopathology images using deep learning. Nat. Med. 24(10), 1559 (2018)CrossRefGoogle Scholar
  3. 3.
    Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L.: ImageNet: a large-scale hierarchical image database. In: CVPR09 (2009)Google Scholar
  4. 4.
    Esteva, A., et al.: Dermatologist-level classification of skin cancer with deep neural networks. Nature 542(7639), 115 (2017)CrossRefGoogle Scholar
  5. 5.
    Gal, Y., Islam, R., Ghahramani, Z.: Deep bayesian active learning with image data. In: Proceedings of the 34th International Conference on Machine Learning, vol. 70, pp. 1183–1192. JMLR. org (2017)Google Scholar
  6. 6.
    Girshick, R.: Fast R-CNN. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1440–1448 (2015)Google Scholar
  7. 7.
    Girshick, R., Donahue, J., Darrell, T., Malik, J.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 580–587 (2014)Google Scholar
  8. 8.
    Hw, G.L., Wortis, M., Mukhopadhyay, R.: Stomatocyte-discocyte-echinocyte sequence of the human red blood cell: evidence for the bilayer-couple hypothesis from membrane mechanics. Proc. Natl. Acad. Sci. 99(26), 16766–16769 (2002)CrossRefGoogle Scholar
  9. 9.
    Ilhan, H.O., Amasyali, M.F.: Active learning as a way of increasing accuracy. Int. J. Comput. Theory Eng. 6(6), 460 (2014)CrossRefGoogle Scholar
  10. 10.
    Kendall, A., Gal, Y.: What uncertainties do we need in bayesian deep learning for computer vision? In: Advances in Neural Information Processing Systems, pp. 5574–5584 (2017)Google Scholar
  11. 11.
    Minetti, G., et al.: Red cell investigations: art and artefacts. Blood Rev. 27(2), 91–101 (2013)CrossRefGoogle Scholar
  12. 12.
    Ren, S., He, K., Girshick, R., Sun, J.: Faster R-CNN: towards real-time object detection with region proposal networks. In: Advances in Neural Information Processing Systems, pp. 91–99 (2015)Google Scholar
  13. 13.
    Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014)
  14. 14.
    Xu, M., Papageorgiou, D.P., Abidi, S.Z., Dao, M., Zhao, H., Karniadakis, G.E.: A deep convolutional neural network for classification of red blood cells in sickle cell anemia. PLoS Comput. Biol. 13(10), e1005746 (2017)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Ario Sadafi
    • 1
    • 2
    • 3
  • Niklas Koehler
    • 1
  • Asya Makhro
    • 4
  • Anna Bogdanova
    • 4
  • Nassir Navab
    • 2
    • 5
  • Carsten Marr
    • 1
    Email author
  • Tingying Peng
    • 1
    • 2
    Email author
  1. 1.Institute of Computational BiologyHelmholtz Zentrum München – German Research Center for Environmental HealthMunichGermany
  2. 2.Computer Aided Medical ProceduresTechnische Universität MünchenMunichGermany
  3. 3.Arivis AGMunichGermany
  4. 4.Red Blood Cell Research Group, Institute of Veterinary Physiology, Vetsuisse Faculty and the Zurich Center for Integrative Human PhysiologyUniversity of ZurichZurichSwitzerland
  5. 5.Computer Aided Medical ProceduresJohns Hopkins UniversityBaltimoreUSA

Personalised recommendations