A probability-based expert system for diagnosing pacemaker-related complications

Author

Chiang, C.J. ; Bernstein, A.D. ; Parsonnet, V.

Author_Institution

Telectronics Pacings Syst., Englewood, CO, USA

fYear

1994

fDate

25-28 Sep 1994

Firstpage

89

Lastpage

92

Abstract

A computerized expert system has been designed to improve the speed and accuracy of pacemaker problem-solving during follow-up and programming procedures performed in the pacemaker clinic. The expert system consists of a series of “if-then” rules that describe the relationship between observed complications and their potential causes. Diagnostic uncertainty is treated by Bayesian statistics and each potential cause is stated with an associated probability according to the signs and symptoms identified by the user. The probabilities are calculated from statistics derived from a database of more than 1400 problems evaluated at Newark Beth Israel Medical Center. The system accommodates multiple problems, each with a multiplicity of symptoms, as well as uncertainty in observed complications. It incorporates a “user-friendly” interface for easy usage by clinical personnel to provide a faster and more accurate diagnosis of pacemaker-related problems

Keywords

Bayes methods; case-based reasoning; diagnostic reasoning; medical expert systems; medical information systems; pacemakers; Bayesian statistics; Newark Beth Israel Medical Center; clinical personnel; computerized expert system; database; diagnosis; diagnostic uncertainty; follow-up; if-then rules; multiple problems; observed complications; pacemaker clinic; pacemaker problem-solving; pacemaker-related complications; potential causes; probability; probability-based expert system; signs; statistics; symptoms; user-friendly interface; Bayesian methods; Databases; Diagnostic expert systems; Medical diagnostic imaging; Pacemakers; Personnel; Probability; Problem-solving; Statistics; Uncertainty;

fLanguage

English

Publisher

ieee

Conference_Titel

Computers in Cardiology 1994

Conference_Location

Bethesda, MD

Print_ISBN

0-8186-6570-X

Type

conf

DOI

10.1109/CIC.1994.470242

Filename

470242