Seebold, Heiko and Lüdtke, Andreas and Möbus, Claus (2005) Bayesian Belief Network based Diagnostics in a Problem-oriented Learning Environment for Cardiology. In: Proceedings of Training, Education & Simulation International 2005 (THESI 2005). MEEC Maastricht, Maastricht.

Preview

- Published Version Volltext (94Kb)

Abstract

Bayesian Belief Networks (BBNs) have a long tradition in medical expert systems for the support of diagnostic reasoning and therapy planning. Usually the BBN containing the uncertain knowledge of the experts is hidden from the user. New to case-based diagnostic training systems is that the structure of the BBN is not hidden from the student, but supports the learning process by visualizing the correlations between different symptoms, sequelas, causes, etc. of the contemplated disease. The BBN is embedded in a problem-oriented context. Students are confronted with naturalistic diagnostic problems. They are asked to state diagnostic hypotheses and to test these hypotheses using the BBN. In this way strategic diagnostic skills are developed. In an evaluation study students confirmed the novelty and importance of the learning environment and that the complexity of the BBN representation was no problem to them.

Item Type:	Book Section
Uncontrolled Keywords:	Bayesian Belief Networks, medical expert system, diagnostic reasoning, therapy planning, case-based diagnostic training system, problem-oriented context, naturalistic diagnostic problems, diagnostic hypotheses, hypothesis-testing, startegic diagnostic skills
Subjects:	Generalities, computers, information > Computer science, internet Philosophy and psychology > Psychology
Divisions:	School of Computing Science, Business Administration, Economics and Law > Department of Computing Science
Date Deposited:	19 Nov 2014 09:23
Last Modified:	19 Nov 2014 09:23
URI:	https://oops.uni-oldenburg.de/id/eprint/1947
URN:	urn:nbn:de:gbv:715-oops-20287
DOI:
Nutzungslizenz:

Actions (login required)

View Item