Purging False Negatives in Cancer Diagnosis Using Incremental Active Learning

  • Catarina Silva
  • Bernardete Ribeiro
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6936)


Cancer is becoming a human plague, and decision-support tools to help physicians better diagnosing are a fulsome research field. False negatives can be a huge problem for cancer diagnosticians, since while a false positive can result in time and money lost, a false negative can result in the lost of human lives, which puts an overwhelming burden on diagnosis.

In this framework, we propose a two-fold approach to purge false negatives in cancer diagnosis without compromising precision performance. First, we use an incremental background knowledge method and then, an active learning strategy completes the procedure. The defined incremental active learning SVM method was tested in the Wisconsin-Madison breast cancer diagnosis problem showing the effectiveness of such techniques in supporting cancer diagnosis.


Support Vector Machine Cancer Diagnosis False Negative Unlabeled Data Empirical Risk 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Merz, C.J., Murphy, P.M.: UCI repository of machine learning data bases, Irvine, CA (1998),
  2. 2.
    Wrensch, M., Georgianna Farren, T.C., Flavia Belli, J.B., Clarke, C., Erdmann, C.A., Lee, M., Moghadassi, M., Peskin-Mentzer, R., Quesenberry, C.P., Souders-Mason, V., Spence, L., Suzuki, M., Gould, M.: Risk factors for breast cancer in a population with high incidence rates. Breast Cancer Res. 5, 88–102 (2003)CrossRefGoogle Scholar
  3. 3.
    Mangasarian, O., Street, W., Wolberg, W.: Breast cancer diagnosis and prognosis via linear programming. Operations Research 43(4), 570–577 (1995)MathSciNetCrossRefzbMATHGoogle Scholar
  4. 4.
    Fogel, D.B., Wasson, E.C., Boughon, E.M., Porto, V.W., Angeline, P.J.: Linear andneural models for classifying breast masses. IEEE Transactions on Medical Imaging 17(3), 485–488 (1998)CrossRefGoogle Scholar
  5. 5.
    Xing, K., Chen, D., Henson, D., Sheng, L.: A clustering-based approach to predict outcome in cancer patients. In: ICMLA, pp. 541–546 (2007)Google Scholar
  6. 6.
    Oprea, A., Strungaru, R., Ungureanu, G.: A Self Organizing Map approach to breast cancer detection. In: International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2008, pp. 3032–3035 (2008)Google Scholar
  7. 7.
    Hong, J., Cho, S.: Incremental Support Vector Machine for Unlabeled Data Classification. In: ICONIP, pp. 1403–1407 (2002)Google Scholar
  8. 8.
    Liu, B., Dai, Y., Li, X., Lee, W., Yu, P.: Building Text Classifiers Using Positive and Unlabeled Examples. In: ICDM, pp. 179–188 (2003)Google Scholar
  9. 9.
    Seeger, M.: Learning with Labeled and Unlabeled Data, Technical Report, Institute for Adaptive and Neural Computation. University of Edinburgh (2001)Google Scholar
  10. 10.
    Silva, C., Ribeiro, B., Lopes, N.: Improving Recall Values in Breast Cancer Diagnosis with Incremental Background Knowledge. In: WCCI 2010 (2010)Google Scholar
  11. 11.
    Silva, C., Ribeiro, B.: On Text-based Mining with Active Learning and Background Knowledge using SVM. Journal of Soft Computing - A Fusion of Foundations, Methodologies and Applications 11(6), 519–530 (2007)Google Scholar
  12. 12.
    Silva, C., Ribeiro, B.: Improving text classification performance with incremental background knowledge. In: Alippi, C., Polycarpou, M., Panayiotou, C., Ellinas, G. (eds.) ICANN 2009. LNCS, vol. 5768, pp. 923–931. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  13. 13.
    Vapnik, V.: The Nature of Statistical Learning Theory, 2nd edn. Springer, Heidelberg (1999)zbMATHGoogle Scholar
  14. 14.
    Zelikovitz, S., Hirsh, H.: Using LSI for text classification in the presence of background text. In: Tenth International Conference on Information Knowledge Management, pp. 113–118 (2001)Google Scholar
  15. 15.
    Joachims, T.: Transductive Inference for Text Classification using Support Vector Machines. In: International Conference on Machine Learning, pp. 200–209 (1999)Google Scholar
  16. 16.
    Silva, C., Ribeiro, B.: Labeled and Unlabeled Data in Text Categorization. In: IEEE International Joint Conference on Neural Networks (2004)Google Scholar
  17. 17.
    Alex, G., Monaco, J., Doyle, S., Basavanhally, A., Reddy, A., Seiler, M., Ganesan, S., Bhanot, G., Madabhushi, A.: Towards improved cancer diagnosis and prognosis using analysis of gene expression data and computer aided imaging. Experimental Biology and Medicine 234(8), 860–879 (2009)CrossRefGoogle Scholar
  18. 18.
    Ribeiro, B.: Learning Adaptive Kernels for Model Diagnosis. Frontiers in Artificial Intelligence and Applications, vol. 104, pp. 563–571. IOS Press, Amsterdam (2003)Google Scholar
  19. 19.
    Akay, M.: SVM combined with feature selection for breast cancer diagnosis. Expert Systems with Applications, Part 2 36(2), 3240–3247 (2009)CrossRefGoogle Scholar
  20. 20.
    Cruz, J., Wishart, D.: Applications of machine learning in cancer prediction and prognosis. Cancer Informatics 2, 59–77 (2006)Google Scholar
  21. 21.
    Schohn, G., Cohn, D.: Less is more: Active Learning with Support Vector Machines. In: International Conference on Machine Learning, pp. 839–846 (2000)Google Scholar
  22. 22.
    Schölkopf, B., Burges, C., Smola, A.: Advances in Kernel Methods - Introduction to Support vector Learning, pp. 1–15. MIT Press, Cambridge (1999)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Catarina Silva
    • 1
    • 2
  • Bernardete Ribeiro
    • 2
  1. 1.School of Technology and ManagementPolytechnic Institute of LeiriaPortugal
  2. 2.Department of Informatics EngineeringCenter for Informatics and Systems of the University of Coimbra (CISUC)Portugal

Personalised recommendations