Email updates

Keep up to date with the latest news and content from BMC Proceedings and BioMed Central.

This article is part of the supplement: International Conference on Prevention & Infection Control (ICPIC 2011)

Open Access Poster presentation

Modeling infectious diseases – urinary tract infection, a practical example

C Huszka1*, J DeRoo1, H Cools1, D Colaert1, C Lovis2 and H Hanberger3

  • * Corresponding author: C Huszka

Author Affiliations

1 Advanced Clinical Applications, AGFA Healthcare, Gent, Belgium

2 Division of Medical Information Sciences, University Hospitals of Geneva, Geneva, Switzerland

3 Faculty of Health Sciences, Linköping University, Linköping, Sweden

For all author emails, please log on.

BMC Proceedings 2011, 5(Suppl 6):P203  doi:10.1186/1753-6561-5-S6-P203

The electronic version of this article is the complete one and can be found online at: http://www.biomedcentral.com/1753-6561/5/S6/P203


Published:29 June 2011

© 2011 Huszka et al; licensee BioMed Central Ltd.

This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Introduction / objectives

The proper management of infectious conditions often requires optimized clinical decisions based on a complex decision making process. Computers are able to aid this process only if the medical knowledge used is machine readable, which requires scrupulous preparation and modeling. Here we describe and evaluate a dynamic model framework for Urinary Tract Infection (UTI).

Methods

We used Bayesian Belief Network (BBN), a probabilistic model that can represent conditional dependencies between variables. Its dynamic characteristics allow modeling complex relationships between medical processes and symptoms. For the medical information we turned to clinical practice guidelines for UTI and abstracted as well as modeled their content with standard ontological concepts. Alltogether we extracted several hundreds of rules describing diagnostic and therapeutic relationships. Finally we validated these rules against the original guidelines using an open-source reasoner (Euler) and a battery of test cases.

Results

We found that the results proposed by the reasoner coincided with that of the clinical guideline in 97% while allowing a much higher complexity with the possibility of freely adding and combining diagnostic and therapeutic parameters.

Conclusion

We conclude the BBN models in infectious conditions deliver not only accurate decisions but their application may also be warranted to effectively deal with large number of combinations of conditions when making complex decision. Similarly it also allows dynamic addition of non-medical (cost, other) parameters for a better optimized decision making process.

Disclosure of interest

None declared.