Comparison of different inference algorithms for medical decision making

Kose G., Sever H., BAL M., Üstündağ A.

INTERNATIONAL JOURNAL OF COMPUTATIONAL INTELLIGENCE SYSTEMS, vol.7, pp.29-44, 2014 (SCI-Expanded, Scopus)

  • Publication Type: Article / Article
  • Volume: 7
  • Publication Date: 2014
  • Doi Number: 10.1080/18756891.2014.853929
  • Journal Name: INTERNATIONAL JOURNAL OF COMPUTATIONAL INTELLIGENCE SYSTEMS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.29-44
  • Keywords: Medical Decision Support Systems, Bayesian Networks, Rule-Based Systems, ALARM Network, ARTIFICIAL-INTELLIGENCE, BAYESIAN NETWORKS
  • Yıldız Technical University Affiliated: Yes

Abstract

A medical diagnosis system (DRCAD), which consists of two sub-modules Bayesian and rule-based inference models, is presented in this study. Three types of tests are conducted to assess the performances of the models producing synthetic data based on the ALARM network. The results indicate that the linear combination of the aforementioned models leads to a 5% and a 30% improvement in medical diagnosis when compared to the Rule Based Method and the Bayesian Network Based Method, respectively.