SUMMARY
A new call triage system, a key part of an emergency support system with a stochastic network model, is examined. Call triage is an operation that allows efficient choice of service grade and dispatching of a suitable rescue team service based on information obtained from a phone call. Call triage is now being tested in a few cities and is showing some benefit. However, there is the issue that if under-triage, in which the condition of the patient is underestimated, is eliminated, then the efficiency is degraded in the form of over-triage. In this report, in order to overcome the issue, a Bayesian network scheme is evaluated for a call triage system. The results of experiments with a real call triage data set show that the Bayesian network improves accuracy. © 2013 Wiley Periodicals, Inc. Electron Comm Jpn, 97(1): 62–69, 2014; Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/ecj.11439
[1]
P. Caralis,et al.
Teaching residents to communicate: the use of a telephone triage system in an academic ambulatory clinic.
,
2010,
Patient education and counseling.
[2]
David Dunt,et al.
Testing the safety of after-hours telephone triage: Patient simulations with validated scenarios
,
2010
.
[3]
K. Chow,et al.
Telephone Triage in Peritoneal Dialysis Population
,
2008
.
[4]
G. Karabatsos,et al.
Universal telenursing triage in Australia and New Zealand - a new primary health service.
,
2008,
Australian family physician.
[5]
Gregory F. Cooper,et al.
A Bayesian method for the induction of probabilistic networks from data
,
1992,
Machine Learning.