Probabilistic Patient Monitoring with Multivariate, Multimodal Extreme Value Theory

  • Samuel Hugueny
  • David A. Clifton
  • Lionel Tarassenko
Part of the Communications in Computer and Information Science book series (CCIS, volume 127)


Conventional patient monitoring is performed by generating alarms when vital signs exceed pre-determined thresholds, but the false-alarm rate of such monitors in hospitals is so high that alarms are typically ignored. We propose a principled, probabilistic method for combining vital signs into a multivariate model of patient state, using extreme value theory (EVT) to generate robust alarms if a patient’s vital signs are deemed to have become sufficiently “extreme”. Our proposed formulation operates many orders of magnitude faster than existing methods, allowing on-line learning of models, leading ultimately to patient-specific monitoring.


Patient monitoring Extreme value theory Extreme value distributions Density estimation 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Hodgetts, T., Kenward, G., Vlackonikolis, I., Payne, S., Castle, N., Crouch, R., Ineson, N., Shaikh, L.: Incidence, location, and reasons for avoidable in-hospital cardiac arrest in a district general hospital. Resuscitation 54, 115–123 (2002)CrossRefGoogle Scholar
  2. 2.
    McQuillan, P., Pilkington, S., Allan, A., Taylor, B., Short, A., Morgan, G., Nielsen, M., Barrett, D., Smith, G.: Confidential inquiry into quality of care before admission to intensive care. British Medical Journal 316, 1853–1858 (1998)Google Scholar
  3. 3.
    Hann, A.: Multi-parameter monitoring for early warning of patient deterioration. PhD thesis, University of Oxford (2008)Google Scholar
  4. 4.
    Tsien, C., Fackler, J.: Poor prognosis for existing monitors in the intensive care unit. Critical Care Medicine 25(4), 614–619 (1997)CrossRefGoogle Scholar
  5. 5.
    Clifton, D., McGrogan, N., Tarassenko, L., King, S., Anuzis, P., King, D.: Bayesian extreme value statistics for novelty detection in gas-turbine engines. In: Proceedings of IEEE Aerospace, Montana, USA, pp. 1–11 (2008)Google Scholar
  6. 6.
    Clifton, D., Tarassenko, L., Sage, C., Sundaram, S.: Condition monitoring of manufacturing processes. In: Proceedings of Condition Monitoring 2008, Edinburgh, UK, pp. 273–279 (2008)Google Scholar
  7. 7.
    Tarassenko, L., Hann, A., Young, D.: Integrated monitoring and analysis for early warning of patient deterioration. British Journal of Anaesthesia 98(1), 149–152 (2006)Google Scholar
  8. 8.
    Parzen, E.: On estimation of a probability density function and mode. Annals of Mathematical Statistics 33, 1065–1076 (1962)MathSciNetzbMATHCrossRefGoogle Scholar
  9. 9.
    Clifton, D., Hugueny, S., Tarassenko, L.: Novelty detection with multivariate extreme value theory, part I: Numerical approach for multimodal estimation. In: Proceedings of IEEE Workshop on Machine Learning in Signal Processing (2009)Google Scholar
  10. 10.
    Hugueny, S., Clifton, D., Tarassenko, L.: Novelty detection with multivariate extreme value theory, part II: Analytical approach for unimodal estimation. In: Proceedings of IEEE Workshop on Machine Learning in Signal Processing (2009)Google Scholar
  11. 11.
    Embrechts, P., Klüppelberg, C., Mikosch, T.: Modelling Extremal Events for Insurance and Finance. Springer, Heidelberg (1997)zbMATHGoogle Scholar
  12. 12.
    Fisher, R., Tippett, L.: Limiting forms of the frequency distributions of the largest or smallest members of a sample. Proc. Camb. Philos. Soc. 24, 180–190 (1928)zbMATHCrossRefGoogle Scholar
  13. 13.
    Clifton, D., Hugueny, S., Tarassenko, L.: A comparison of approaches to multivariate extreme value theory for novelty detection. In: Proceedings of IEEE Workshop on Statistical Signal Processing (2009)Google Scholar
  14. 14.
    Hourihan, F., Bishop, G., Hillman, K., Daffurn, K.: The medical emergency team: a new strategy to identify and intervene in high-risk patients. Clinical Intensive Care 6(6), 269–272 (1995)CrossRefGoogle Scholar
  15. 15.
    Hillman, K., Chen, J., Cretikos, M., Bellomo, R., Brown, D., Doig, G., Finfer, S., Flabouris, A.: Introduction of the medical emergency team (MET) system: a cluster-randomised controlled trial. The Lancet 365(9477), 2091–2097 (2005)CrossRefGoogle Scholar
  16. 16.
    Mayrose, I., Friedman, N., Pupko, T.: A gamma mixture model better accounts for among site rate heterogeneity. Bioinformatics 21(2), 151–158 (2005)CrossRefGoogle Scholar
  17. 17.
    Svensen, M., Bishop, C.: Robust Bayesian mixture modelling. Neurocomputing 64, 235–252 (2005)CrossRefGoogle Scholar
  18. 18.
    Ebden, M., Stranjak, A., Dutta, P., Rogers, A.: A multi-agent simulation system for prediction and scheduling of aero engine overhaul. In: Proceedings of the 7th Conference on Autonomous Agents and Multiagent Systems (AAMAS), Estoril, Portugal (2008)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Samuel Hugueny
    • 1
  • David A. Clifton
    • 1
  • Lionel Tarassenko
    • 1
  1. 1.Institute of Biomedical Engineering, Department of Engineering ScienceUniversity of OxfordOxfordU.K.

Personalised recommendations