An Improved Multi-task Learning Approach with Applications in Medical Diagnosis

  • Jinbo Bi
  • Tao Xiong
  • Shipeng Yu
  • Murat Dundar
  • R. Bharat Rao
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5211)

Abstract

We propose a family of multi-task learning algorithms for collaborative computer aided diagnosis which aims to diagnose multiple clinically-related abnormal structures from medical images. Our formulations eliminate features irrelevant to all tasks, and identify discriminative features for each of the tasks. A probabilistic model is derived to justify the proposed learning formulations. By equivalence proof, some existing regularization-based methods can also be interpreted by our probabilistic model as imposing a Wishart hyperprior. Convergence analysis highlights the conditions under which the formulations achieve convexity and global convergence. Two real-world medical problems: lung cancer prognosis and heart wall motion analysis, are used to validate the proposed algorithms.

Keywords

Loss Function Ground Glass Opacity Multitask Learning Training Sample Size Convex Loss Function 
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.

References

  1. 1.
    Roehrig, J.: The promise of CAD in digital mamography. European Journal of Radiology 31, 35–39 (1999)CrossRefGoogle Scholar
  2. 2.
    Armato-III, S.G., Giger, M.L., MacMahon, H.: Automated detection of lung nodules in CT scans: preliminary results. Medical Physics 28(8), 1552–1561 (2001)CrossRefGoogle Scholar
  3. 3.
    Suzuki, K., Kusumoto, M., Watanabe, S., Tsuchiya, R., Asamura, H.: Radiologic classfication of small adenocarcinoma of the lung: Radiologic-pathologic correlation and its prognostic impact. The Annals of Thoracic Surgery CME Program 81, 413–420 (2006)CrossRefGoogle Scholar
  4. 4.
    Caruana, R.: Multitask learning. Machine Learning 28(1), 41–75 (1997)CrossRefGoogle Scholar
  5. 5.
    Yu, K., Tresp, V., Schwaighofer, A.: Learning Gaussian processes from multiple tasks. In: ICML 2005 (2005)Google Scholar
  6. 6.
    Ando, R.K., Zhang, T.: A framework for learning predictive structures from multiple tasks and unlabeled data. Journal of Machine Learning Research 6, 1855–1887 (2005)MathSciNetGoogle Scholar
  7. 7.
    Heskes, T.: Empirical Bayes for learning to learn. In: Proc. 17th International Conf. on Machine Learning, pp. 367–374. Morgan Kaufmann, San Francisco (2000)Google Scholar
  8. 8.
    Jebara, T.: Multi-task feature and kernel selection for SVMs. In: Proceedings of the 21st International Conference on Machine learning (2004)Google Scholar
  9. 9.
    Obozinski, G., Taskar, B., Jordan, M.I.: Multi-task feature selection. Technical report, UC Berkeley (2006)Google Scholar
  10. 10.
    Argyriou, A., Evgeniou, T., Pontil, M.: Multi-task feature learning. In: Schölkopf, B., Platt, J., Hoffman, T. (eds.) Advances in Neural Information Processing Systems, vol. 19, pp. 41–48. MIT Press, Cambridge (2007)Google Scholar
  11. 11.
    Srebro, N., Rennie, J.D.M., Jaakola, T.S.: Maximum-margin matrix factorization. In: NIPS 2005 (2005)Google Scholar
  12. 12.
    Xiong, T., Bi, J., Rao, R.B., Cherkassky, V.: Probabilistic joint feature selection for multi-task learning. In: SIAM International Conference on Data Mining (2006)Google Scholar
  13. 13.
    Bezdek, J.C., Hathaway, R.J.: Convergence of alternating optimization. Neural, Parallel Sci. Comput. 11, 351–368 (2003)MATHMathSciNetGoogle Scholar
  14. 14.
    Bezdek, J.C., Hathaway, R.J.: Some notes on alternating optimization. In: Pal, N.R., Sugeno, M. (eds.) AFSS 2002. LNCS (LNAI), vol. 2275, pp. 288–300. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  15. 15.
    Evegniou, T., Pontil, M.: Regularized multi–task learning. In: Proc. of 17–th SIGKDD Conf. on Knowledge Discovery and Data Mining (2004)Google Scholar
  16. 16.
    Rao, R.B., Bi, J., Fung, G., Salganicoff, M., Obuchowski, N., Naidich, D.: LungCAD: a clinically approved, machine learning system for lung cancer detection. In: ACM International Conference on Knowledge Discovery and Data Mining (2007)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Jinbo Bi
    • 1
  • Tao Xiong
    • 2
  • Shipeng Yu
    • 1
  • Murat Dundar
    • 1
  • R. Bharat Rao
    • 1
  1. 1.CAD and Knowledge Solutions, Siemens Medical SolutionsMalvernUSA
  2. 2.Risk Management, Applied Research, eBay Inc.San JoseUSA

Personalised recommendations