A Rule Based Expert System to Assess Coronary Artery Disease under Uncertainty

TL;DR

This paper presents a rule-based expert system that predicts five stages of coronary artery disease using belief rule systems and evidential reasoning, aiming to improve early detection and decision support.

Contribution

It introduces a novel rule-based expert system combining belief rule systems and evidential reasoning for CAD stage prediction, enhancing diagnostic accuracy.

Findings

Higher success rate compared to existing systems

Lower false omission rate than current CDSS

Effective in early CAD detection

Abstract

The coronary artery disease (CAD) involves narrowing and damaging the major blood vessels has become the most life threating disease in the world especially in south Asian reason. Although outstanding medical facilities are available in Singapore and India for CAD patients, early detection of CAD stages are necessary to minimize the patients' sufferings and expenses. It is really challenging for doctors to incorporate numerous factors for details analysis and CAD detections are expensive as it needs expensive medical facilities. Clinical Decision Support Systems (CDSS) may assist to analyze numerous factors for patients. In this paper, a Rule Based Expert System (RBES) is proposed which can predict five different stages of CAD. RBES contains five different Belief Rule Based (BRB) systems and the final output is produced by combining all BRBs using the Evidential Reasoning (ER). Success, Error, Failure, False Omission rates are calculated to measures the performance of the RBES. The Success Rate and False Omission Rate show better performance comparing to existing CDSS.