Advertisement

Learning Density Independent Texture Features

  • Michiel Kallenberg
  • Mads Nielsen
  • Katharina Holland
  • Nico Karssemeijer
  • Christian Igel
  • Martin Lillholm
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9699)

Abstract

Breast cancer risk assessment is becoming increasingly important in clinical practice. It has been suggested that features that characterize mammographic texture are more predictive for breast cancer than breast density. Yet, strong correlation between both types of features is an issue in many studies. In this work we investigate a method to generate texture features and/or scores that are independent of breast density. The method is especially useful in settings where features are learned from the data itself. We evaluate our method on a case control set comprising 394 cancers, and 1182 healthy controls. We show that the learned density independent texture features are significantly associated with breast cancer risk. As such it may aid in exploring breast characteristics that are predictive of breast cancer irrespective of breast density. Furthermore it offers opportunities to enhance personalized breast cancer screening beyond breast density.

Keywords

Texture Breast density Deep learning Breast cancer risk assessment 

References

  1. 1.
    Bengio, Y., Courville, A., Vincent, P.: Representation learning: a review and new perspectives. IEEE Trans. Pattern Anal. Mach. Intell. 35(8), 1798–1828 (2013)CrossRefGoogle Scholar
  2. 2.
    Häberle, L., Wagner, F., Fasching, P.A., Jud, S.M., Heusinger, K., Loehberg, C.R., Hein, A., Bayer, C.M., Hack, C.C., Lux, M.P., et al.: Characterizing mammographic images by using generic texture features. Breast Cancer Res. 14(2), R59 (2012)CrossRefGoogle Scholar
  3. 3.
    Hall, M.A.: Correlation-based feature selection for machine learning. Ph.D. thesis, The University of Waikato (1999)Google Scholar
  4. 4.
    Hansen, B.B., Klopfer, S.O.: Optimal full matching and related designs via network flows. J. Comput. Graph. Stat. 15, 609–627 (2012)MathSciNetCrossRefGoogle Scholar
  5. 5.
    Kallenberg, M., Lilholm, M., Diao, P., Petersen, K., Holland, K., Karssemeijer, N., Igel, C., Nielsen, M.: Assessing breast cancer masking risk with automated texture analysis in full field digital mammography. In: Annual Meeting of the Radiological Society of North America (2015)Google Scholar
  6. 6.
    Kallenberg, M., Petersen, K., Lilholm, M., Jørgensen, D., Diao, P., Holland, K., Karssemeijer, N., Igel, C., Nielsen, M.: Automated texture scoring for assessing breast cancer masking risk in full field digital mammography. In: ECR (2015)Google Scholar
  7. 7.
    Manduca, A., Carston, M., Heine, J., Scott, C., Pankratz, V., Brandt, K., Sellers, T., Vachon, C., Cerhan, J.: Texture features from mammographic images and risk of breast cancer. Cancer Epidemiol. Biomarkers Prev. 18, 837–845 (2009). http://dx.doi.org/10.1158/1055-9965.EPI-08-0631 CrossRefGoogle Scholar
  8. 8.
    Masci, J., Meier, U., Cireşan, D., Schmidhuber, J.: Stacked convolutional auto-encoders for hierarchical feature extraction. In: Honkela, T. (ed.) ICANN 2011, Part I. LNCS, vol. 6791, pp. 52–59. Springer, Heidelberg (2011)CrossRefGoogle Scholar
  9. 9.
    Nielsen, M., Vachon, C.M., Scott, C.G., Chernoff, K., Karemore, G., Karssemeijer, N., Lillholm, M., Karsdal, M.A.: Mammographic texture resemblance generalizes as an independent risk factor for breast cancer. Breast Cancer Res. 16, R37 (2014). http://dx.doi.org/10.1186/bcr3641 CrossRefGoogle Scholar
  10. 10.
    Petersen, K., Chernoff, K., Nielsen, M., Ng, A.: Breast density scoring with multiscale denoising autoencoders. In: Proceedings Sparsity Techniques in Medical Imaging 2012, in conjunction with MICCAI 2012 (2012)Google Scholar
  11. 11.
    Ranzato, M., Poultney, C.S., Chopra, S., LeCun, Y.: Efficient learning of sparse representations with an energy-based model. In: Advances in Neural Information Processing Systems, pp. 1137–1144 (2007)Google Scholar
  12. 12.
    Schmidhuber, J.: Deep learning in neural networks: an overview. Neural Netw. 61, 85–117 (2015)CrossRefGoogle Scholar
  13. 13.
    Vincent, P., Larochelle, H., Bengio, Y., Manzagol, P.A.: Extracting and composing robust features with denoising autoencoders. In: International Conference on Machine Learning, pp. 1096–1103 (2008)Google Scholar
  14. 14.
    Zheng, Y., Keller, B.M., Ray, S., Wang, Y., Conant, E.F., Gee, J.C., Kontos, D.: Parenchymal texture analysis in digital mammography: a fully automated pipeline for breast cancer risk assessment. Med. Phys. 42(7), 4149–4160 (2015)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Michiel Kallenberg
    • 1
    • 2
  • Mads Nielsen
    • 1
    • 2
  • Katharina Holland
    • 3
  • Nico Karssemeijer
    • 3
  • Christian Igel
    • 1
  • Martin Lillholm
    • 1
    • 2
  1. 1.University of CopenhagenCopenhagen OEDenmark
  2. 2.Biomediq A/SCopenhagen OEDenmark
  3. 3.Radboud University Medical CentreNijmegenNetherlands

Personalised recommendations