Using Intelligent Systems in Predictions of the Bacterial Causative Agent of an Infection

  • Authors:
  • Diana R. Cundell;Randy S. Silibovsky;Robyn Sanders;Les M. Sztandera

  • Affiliations:
  • -;-;-;-

  • Venue:
  • ISMIS '00 Proceedings of the 12th International Symposium on Foundations of Intelligent Systems
  • Year:
  • 2000

Quantified Score

Hi-index 0.02

Visualization

Abstract

In this study, we designed a fuzzy logic system to examine the influence of the demographic variables of age, blood type, gender and race on bacterial infection rates using a medical database assembled over 17 months from patients presenting to Albert Einstein Medical Center. The intelligent system was created using 155 patients, randomly selected from the database, and consisted of four input categories of demographic variables and four output categories of bacterial infection ("streptococci", "staphylococci", "Escherichia coli" and "non-E. coli gram negative rods"). The remaining 32 patients were used to assess the program's ability to correctly determine bacterial infection when provided only with demographic data. Our intelligent system correctly assigned the bacterial output group in 27 of these 32 patients, giving an overall correlation of 84.4%. These studies suggest that demographic variables are major factors influencing bacterial infection. Such a system may, therefore, hold promise as a diagnostic tool.