PROBABILISTIC PATIENT MONITORING USING EXTREME VALUE THEORY - A Multivariate, Multimodal Methodology for Detecting Patient Deterioration

Samuel Hugueny, David A. Clifton, Lionel Tarassenko

Abstract

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.

References

  1. Clifton, D., Hugueny, S., and Tarassenko, L. (2009a). A comparison of approaches to multivariate extreme value theory for novelty detection. Proceedings of IEEE Workshop on Statistical Signal Processing.
  2. Clifton, D., Hugueny, S., and Tarassenko, L. (2009b). Novelty detection with multivariate extreme value theory, part I: Numerical approach for multimodal estimation. Proceedings of IEEE Workshop on Machine Learning in Signal Processing.
  3. Clifton, D., McGrogan, N., Tarassenko, L., King, S., Anuzis, P., and King, D. (2008a). Bayesian extreme value statistics for novelty detection in gas-turbine engines. In Proceedings of IEEE Aerospace, Montana, USA, pages 1-11.
  4. Clifton, D., Tarassenko, L., Sage, C., and Sundaram, S. (2008b). Condition monitoring of manufacturing processes. In Proceedings of Condition Monitoring 2008, Edinburgh, UK, pages 273-279.
  5. Ebden, M., Stranjak, A., Dutta, P., and Rogers, A. (2008). 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.
  6. Embrechts, P., Kl├╝ppelberg, C., and Mikosch, T. (1997). Modelling Extremal Events for Insurance and Finance. Springer.
  7. Fisher, R. and Tippett, L. (1928). Limiting forms of the frequency distributions of the largest or smallest members of a sample. Proc. Camb. Philos. Soc., 24:180- 190.
  8. Hann, A. (2008). Multi-parameter monitoring for early warning of patient deterioration. PhD thesis, University of Oxford.
  9. Hodgetts, T., Kenward, G., Vlackonikolis, I., Payne, S., Castle, N., Crouch, R., Ineson, N., and Shaikh, L. (2002). Incidence, location, and reasons for avoidable in-hospital cardiac arrest in a district general hospital. Resuscitation, 54:115-123.
  10. Hugueny, S., Clifton, D., and Tarassenko, L. (2009). Novelty detection with multivariate extreme value theory, part II: Analytical approach for unimodal estimation. Proceedings of IEEE Workshop on Machine Learning in Signal Processing.
  11. Mayrose, I., Friedman, N., and Pupko, T. (2005). A gamma mixture model better accounts for among site rate heterogeneity. Bioinformatics, 21(2):151-158.
  12. McQuillan, P., Pilkington, S., Allan, A., Taylor, B., Short, A., Morgan, G., Nielsen, M., Barrett, D., and Smith, G. (1998). Confidential inquiry into quality of care before admission to intensive care. British Medical Journal, 316:1853-1858.
  13. Parzen, E. (1962). On estimation of a probability density function and mode. Annals of Mathematical Statistics, 33:1065-1076.
  14. Svensen, M. and Bishop, C. (2005). Robust Bayesian mixture modelling. Neurocomputing, 64:235-252.
  15. Tarassenko, L., Hann, A., and Young, D. (2006). Integrated monitoring and analysis for early warning of patient deterioration. British Journal of Anaesthesia, 98(1):149-152.
  16. Tsien, C. and Fackler, J. (1997). Poor prognosis for existing monitors in the intensive care unit. Critical Care Medicine, 25(4):614-619.
Download


Paper Citation


in Harvard Style

Hugueny S., A. Clifton D. and Tarassenko L. (2010). PROBABILISTIC PATIENT MONITORING USING EXTREME VALUE THEORY - A Multivariate, Multimodal Methodology for Detecting Patient Deterioration . In Proceedings of the Third International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2010) ISBN 978-989-674-018-4, pages 5-12. DOI: 10.5220/0002690200050012


in Bibtex Style

@conference{biosignals10,
author={Samuel Hugueny and David A. Clifton and Lionel Tarassenko},
title={PROBABILISTIC PATIENT MONITORING USING EXTREME VALUE THEORY - A Multivariate, Multimodal Methodology for Detecting Patient Deterioration},
booktitle={Proceedings of the Third International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2010)},
year={2010},
pages={5-12},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002690200050012},
isbn={978-989-674-018-4},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Third International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2010)
TI - PROBABILISTIC PATIENT MONITORING USING EXTREME VALUE THEORY - A Multivariate, Multimodal Methodology for Detecting Patient Deterioration
SN - 978-989-674-018-4
AU - Hugueny S.
AU - A. Clifton D.
AU - Tarassenko L.
PY - 2010
SP - 5
EP - 12
DO - 10.5220/0002690200050012