Advertisement

Novelty Detection for Identifying Deterioration in Emergency Department Patients

  • David A. Clifton
  • David Wong
  • Susannah Fleming
  • Sarah J. Wilson
  • Rob Way
  • Richard Pullinger
  • Lionel Tarassenko
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6936)

Abstract

This paper presents the preliminary results of an observational study into the use of novelty detection techniques for detecting physiological deterioration in vital-sign data acquired from Emergency Department (ED) patients. Such patients are typically in an acute condition with a significant chance of deteriorating during their stay in hospital. Existing methods for monitoring ED patients involve manual “early warning score” (EWS) systems based on heuristics in which clinicians calculate a score based on the patient vital signs. We investigate automated novelty detection methods to perform “intelligent” monitoring of the patient between manual observations, to provide early warning of patient deterioration. Analysis of the performance of classification systems for on-line novelty detection is not straightforward. We discuss the obstacles that must be considered when determining the efficacy of on-line classification systems, and propose metrics for evaluating such systems.

Keywords

Novelty Detection Support Vector Machines 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Safer Care for Acutely Ill Patients: Learning from Serious Accidents. Technical Report. National Patient Safety Association (2007)Google Scholar
  2. 2.
    Recognition of and Response to Acute Illness in Adults in Hospital. Technical Report. National Institute for Clinical Excellence (2007)Google Scholar
  3. 3.
    Tarassenko, L., Clifton., D.A., Pinsky, M.R., Hravnak, M.T., Woods, J.R., Watkinson, P.J.: Centile-Based Early Warning Scores Derived from Statistical Distributions of Vital Signs. Resuscitation (2011), doi:10.1016/j.resuscitation.2011.03.006Google Scholar
  4. 4.
    Williams, C.K.I., Quinn, J., McIntosh, N.: Factorial Switched Kalman Filters for Condition Monitoring in Neonatal Intensive Care. In: Advances in Neural Information Processing Systems, vol. 18, pp. 1513–1520. MIT Press, Cambridge (2006)Google Scholar
  5. 5.
    Tarassenko, L., Hann, A., Young, D.: Integrated Monitoring and Analysis for Early Warning of Patient Deterioration. Brit. J. Anaesthesia 98(1), 149–152 (2007)Google Scholar
  6. 6.
    Hann, A.: Multi-parameter Monitoring for Early Warning of Patient Deterioration. Ph.D. Thesis. University of Oxford (2008)Google Scholar
  7. 7.
    Schölkopf, B., Platt, J.C., Shawe-Taylor, J., Smola, A.J., Williamson, R.C.: Estimating the Support of a High-Dimensional Distribution. Neural Computation 13, 1443–1471 (2001)CrossRefMATHGoogle Scholar
  8. 8.
    Tax, D.M.J., Duin, R.P.W.: Data Domain Description using Support Vectors. In: Proc. ESANN, pp. 215–256 (1999)Google Scholar
  9. 9.
    Bishop, C.M.: Pattern Recognition and Machine Learning. Springer, Heidelberg (2006)MATHGoogle Scholar
  10. 10.
    Prytherch, D.R., Smith, G.B., Schmidt, P.E., Featherstone, P.I.: ViEWS - Towards a National Early Warning Score for Detecting Adult In-Patient Deterioration. Resuscitation 81, 932–937 (2010)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • David A. Clifton
    • 1
  • David Wong
    • 1
  • Susannah Fleming
    • 2
  • Sarah J. Wilson
    • 3
    • 4
  • Rob Way
    • 4
  • Richard Pullinger
    • 4
  • Lionel Tarassenko
    • 1
  1. 1.Institute of Biomedical EngineeringUniversity of OxfordOxfordUK
  2. 2.Department of Primary Health CareUniversity of OxfordOxfordUK
  3. 3.Heatherwood and Wexham Park HospitalsNHS Foundation TrustWexhamUK
  4. 4.Oxford Radcliffe HospitalsNHS TrustOxfordUK

Personalised recommendations