Leukemic B-Lymphoblast Cell Detection with Monte Carlo Dropout Ensemble Models

  • Hao-Yu YangEmail author
  • Lawrence H. Staib
Conference paper
Part of the Lecture Notes in Bioengineering book series (LNBE)


Automatic detection of Leukemic B-lymphoblast cells from normal cells is both crucial and difficult with subject variability being a major obstacle. While deep learning models are widely adopted in image recognition and demonstrate state-of-the-art performance, deep learning models in their vanilla form do not provide an estimate of prediction uncertainty. This is especially problematic when there is a high discrepancy between training and testing data such as in blood smear images where subject level variability and sample noise from staining and illumination are non-negligible. To address these issues, we propose a novel ensemble method by weighting each base model according to its respective predictive confidence obtained from Monte Carlo dropout. Our ensemble method can effectively prevent overfitting and increases classification performance compared to any single model result. Furthermore, we demonstrate the model’s ability to identify samples that do not fall in the training data distribution. We achieved 0.8929 prediction score according to the leaderboard on the SBILab challenge website.


Monte Carlo dropout Model uncertainty Ensemble model Convolutional neural network 


  1. 1.
    Duggal, R., Gupta, A., Gupta, R.: Segmentation of overlapping/touching white blood cell nuclei using artificial neural networks. CME Series on Hemato-Oncopathology (2016)Google Scholar
  2. 2.
    Duggal, R., Gupta, A., Gupta, R., Mallick, P.: SD-layer: stain deconvolutional layer for CNNs in medical microscopic imaging. In: Medical Image Computing and Computer-Assisted Intervention, MICCAI’17 (2017)CrossRefGoogle Scholar
  3. 3.
    Duggal, R., Gupta, A., Gupta, R., Wadhwa, M., Ahuja, C.: Overlapping cell nuclei segmentation in microscopic images using deep belief networks. In: Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP) (2016)Google Scholar
  4. 4.
    Gal, Y., Ghahramani, Z.: Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning (2015). arXiv:1506.02142
  5. 5.
    Gupta, A., Duggal, R., Gupta, R., Kumar, L., Thakkar, N., Satpathy, D.: GCTI-SN: geometry-inspired chemical and tissue invariant stain normalization of microscopic medical images. Under reviewGoogle Scholar
  6. 6.
    Gupta, R., Mallick, P., Duggal, R., Gupta, A., Sharma, O.: Stain color normalization and segmentation of plasma cells in microscopic images as a prelude to development of computer assisted automated disease diagnostic tool in multiple Myeloma. In: 16th International Myeloma Workshop (IMW), Mar 2017Google Scholar
  7. 7.
    He, K., Zhang, X., Ren, S., Sun, J.: Deep Residual Learning for Image Recognition (2015). arXiv:1512.03385
  8. 8.
    Hu, J., Shen, L., Albanie, S., Sun, G., Wu, E.: Squeeze-and-Excitation Networks (2017). arXiv:1709.01507
  9. 9.
    Huang, G., Liu, Z., van der Maaten, L., Weinberger, K.Q.: Densely Connected Convolutional Networks (2016). arXiv:1608.06993
  10. 10.
    Liu, C., Zoph, B., Shlens, J., Hua, W., Li, L., Fei-Fei, L., Yuille, A.L., Huang, J., Murphy, K.: Progressive neural architecture search. In: CoRR (2017).
  11. 11.
    Rasmussen, C.E., Williams, C.K.I.: Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning). The MIT Press (2005)Google Scholar
  12. 12.
    Szegedy, C., Ioffe, S., Vanhoucke, V.: Inception-v4, inception-ResNet and the impact of residual connections on learning. In: CoRR (2016).

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  1. 1.Yale UniversityNew HavenUSA
  2. 2.Cura Cloud CooperationSeattleUSA

Personalised recommendations