Event Detection on Patient Monitoring Data Using Causal Probabilistic Networks

Abstract:

Applications are described of the knowledge-representation technique Causal Probabilistic Networks for the detection of events related to the patient and the monitoring equipment, in the context of intensive care patient monitoring. The purpose of the event detection is to reduce the number of false alarms generated by the patient monitor, and to generate alarms related to patient or equipment events rather than to threshold violation. A PC-based system has been developed and tested on patient data. A system for real-time use in patient monitoring is under development.

Keywords

Event Detection - Patient Monitoring - Intensive Care - Causal Probabilistic Networks

• REFERENCES

• 1 Laursen P. Intelligent Monitoring. Report from ‘Erhvervsforskerprojekt EF259’ (Ph.D. in Industrial Research), S&W Medico Teknik A/S, Sept. 1991 (Unpublished)
• 2 Koski EMJ. et al. Frequency and reliability of alarms in the monitoring of cardiac postoperative patients. Int J Clin Monit Comput 1990; 02: 129-33.
• 3 Deller A, Konrad F, Kilian J, Schühle B. Alarms in an operative intensive care unit: response of the nursing staff. In: Hedley-White J. ed. Operating Room and Intensive Care Alarms and Information Transfer, ASTM STP 1152. Philadelphia PA: American Society for Testing and Materials; 1992: 19-26.
• 4 Andreassen S, Jensen FV, Olesen KG. Medical Expert Systems Based On Causal Probabilistic Networks. IES Report R91-6. Aalborg University. Denmark: Institute of Electronic Systems; Jan. 1991
• 5 Jensen FV, Olesen KG, Andersen SK. An Algebra of Bayesian Belief Universes for Knowledge Based Systems. IES Report R88-25. Aalborg University, Denmark: Institute of Electronic Systems; July 1988
• 6 Jensen FV, Lauritzen SL, Olesen KG. Bayesian Updating in Recursive Graphical Models by Local Computation. IES Report R89-15. Aalborg University, Denmark: Institute of Electronic Systems; June 1989
• 7 Harrison PJ, Stevens CF. A Bayesian approach to short-term forecasting. Op Res Quart 1971; 04: 341-62.