Advertisement

Time-Dependent Survival Neural Network for Remaining Useful Life Prediction

  • Jianfei Zhang
  • Shengrui WangEmail author
  • Lifei Chen
  • Gongde Guo
  • Rongbo Chen
  • Alain Vanasse
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11439)

Abstract

Remaining useful life (RUL) prediction has been a topic of practical interest in many fields involving preventive intervention, including manufacturing, medicine and healthcare. While most of the conventional approaches suffer from censored failures arising and statistically circumscribed assumptions, few attempts have been made to predict RUL by developing a survival learning machine that explores the underlying relationship between time-varying prognostic variables and failure-free survival probability. This requires a purely data-driven prediction approach, devoid of any a survival model and all statistical assumptions. To this end, we propose a time-dependent survival neural network that additively estimates a latent failure risk and performs multiple binary classifications to generate prognostics of RUL-specific probability. We train the neural network by a new survival learning criterion that minimizes the censoring Kullback-Leibler divergence and guarantees monotonicity of the resulting probability. Experiments on four datasets demonstrate the great promise of our approach in real applications.

Keywords

RUL prediction Neural network Survival learning Failure risk Time-varying data 

Notes

Acknowledgments

This work was supported by the National Natural Science Foundation of China (NSFC) under Grant No. 61672157, the Natural Sciences and Engineering Research Council of Canada (NSERC) under Grant No. 396097-2010, the program PAFI of Centre de Recherche du CHUS.

References

  1. 1.
    Buckley, J., James, I.: Linear regression with censored data. Biometrika 66(3), 429–436 (1979)CrossRefGoogle Scholar
  2. 2.
    Chen, L., Wang, S.: Central clustering of categorical data with automated feature weighting. In: IJCAI, pp. 1260–1266 (2013)Google Scholar
  3. 3.
    Cox, D.R.: Regression models and life tables. J. R. Stat. Soc. Ser. B. Stat. Methodol. 34, 187–220 (1972)MathSciNetzbMATHGoogle Scholar
  4. 4.
    Faraggi, D., Simon, R.: A neural network model for survival data. Stat. Med. 14(1), 73–82 (1995)CrossRefGoogle Scholar
  5. 5.
    Fisher, L.D., Lin, D.Y.: Time-dependent covariates in the Cox proportional-hazards regression model. Annu. Rev. Public Health 20(1), 145–157 (1999)CrossRefGoogle Scholar
  6. 6.
    Ishwaran, H., Kogalur, U.B., Blackstone, E.H., Lauer, M.S.: Random survival forests. Ann. Appl. Stat. 2, 841–860 (2008)MathSciNetCrossRefGoogle Scholar
  7. 7.
    Katzman, J., Shaham, U., Bates, J., Cloninger, A., Jiang, T., Kluger, Y.: DeepSurv: personalized treatment recommender system using a Cox proportional hazards deep neural network. BMC Med. Res. Methodol. 18, 18–24 (2018)CrossRefGoogle Scholar
  8. 8.
    Kim, H., Golub, G.H., Park, H.: Imputation of missing values in DNA microarray gene expression data. In: CSB, pp. 572–573 (2004)Google Scholar
  9. 9.
    Li, H., Ge, Y., Zhu, H., Xiong, H., Zhao, H.: Prospecting the career development of talents: a survival analysis perspective. In: KDD, pp. 917–925 (2017)Google Scholar
  10. 10.
    Li, Y., Wang, J., Ye, J., Reddy, C.K.: A multi-task learning formulation for survival analysis. In: KDD, pp. 1715–1724 (2016)Google Scholar
  11. 11.
    Li, Y., Wang, L., Wang, J., Ye, J., Reddy, C.K.: Transfer learning for survival analysis via efficient L2, 1-norm regularized Cox regression. In: ICDM, pp. 231–240 (2017)Google Scholar
  12. 12.
    Li, Y., Xu, K.S., Reddy, C.K.: Regularized parametric regression for high-dimensional survival analysis. In: SDM, pp. 765–773 (2016)Google Scholar
  13. 13.
    Lin, H.C., Baracos, V., Greiner, R., Chun-Nam, J.Y.: Learning patient-specific cancer survival distributions as a sequence of dependent regressors. In: NIPS, pp. 1845–1853 (2011)Google Scholar
  14. 14.
    Michalewicz, Z., Schoenauer, M.: Evolutionary algorithms for constrained parameter optimization problems. Evol. Comput. 4(1), 1–32 (1996)CrossRefGoogle Scholar
  15. 15.
    Moghaddass, R., Rudin, C.: The latent state hazard model, with application to wind turbine reliability. Ann. Appl. Stat. 9(4), 1823–1863 (2014)MathSciNetCrossRefGoogle Scholar
  16. 16.
    Sinha, D., Ibrahim, J.G., Chen, M.: A Bayesian justification of Cox’s partial likelihood. Biometrika 90(3), 629–641 (2003)MathSciNetCrossRefGoogle Scholar
  17. 17.
    Street, W.N.: A neural network model for prognostic prediction. In: ICML, pp. 540–546 (1998)Google Scholar
  18. 18.
    Vinzamuri, B., Li, Y., Reddy, C.K.: Active learning based survival regression for censored data. In: CIKM, pp. 241–250 (2014)Google Scholar
  19. 19.
    Wang, L., Li, Y., Zhou, J., Zhu, D., Ye, J.: Multi-task survival analysis. In: ICDM, pp. 485–494 (2017)Google Scholar
  20. 20.
    Wang, P., Li, Y., Reddy, C.K.: Machine learning for survival analysis: a survey. ACM Comput. Surv. 51(6), 1–36 (2019)CrossRefGoogle Scholar
  21. 21.
    Wei, L.J.: The accelerated failure time model: a useful alternative to the Cox regression model in survival analysis. Stat. Med. 11(14–15), 1871–1879 (1992)CrossRefGoogle Scholar
  22. 22.
    Wilamowski, B.M., Yu, H.: Neural network learning without backpropagation. IEEE Trans. Neural Netw. 21(11), 1793–1803 (2010)CrossRefGoogle Scholar
  23. 23.
    Wu, Y., Yuan, M., Dong, S., Lin, L., Liu, Y.: Remaining useful life estimation of engineered systems using vanilla LSTM neural networks. Neural Comput. 275, 167–179 (2018)Google Scholar
  24. 24.
    Yu, S., et al.: Privacy-preserving Cox regression for survival analysis. In: KDD, pp. 1034–1042 (2008)Google Scholar
  25. 25.
    Zhang, J., Chen, L., Vanasse, A., Courteau, J., Wang, S.: Survival prediction by an integrated learning criterion on intermittently varying healthcare data. In: AAAI, pp. 72–78 (2016)Google Scholar
  26. 26.
    Zhang, J., Wang, S., Courteau, J., Chen, L., Bach, A., Vanasse, A.: Predicting COPD failure by modeling hazard in longitudinal clinical data. In: ICDM, pp. 639–648 (2016)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Jianfei Zhang
    • 1
    • 2
  • Shengrui Wang
    • 1
    • 2
    Email author
  • Lifei Chen
    • 1
  • Gongde Guo
    • 1
  • Rongbo Chen
    • 2
  • Alain Vanasse
    • 3
    • 4
  1. 1.College of Mathematics and InformaticsFujian Normal UniversityFuzhouChina
  2. 2.Département d’InformatiqueUniversité de SherbrookeSherbrookeCanada
  3. 3.Département de Médecine de Famille et de Médecine d’UrgenceUniversité de SherbrookeSherbrookeCanada
  4. 4.Centre de Recherche du Centre Hospitalier Universitaire de SherbrookeSherbrookeCanada

Personalised recommendations