Classification of risk in dengue fever and dengue haemorrhagic fever using rule based expert system

This paper describes the development of a rule based expert system to classify risk in dengue infections using bioelectrical impedance analysis (BIA). This system is able to classify three types of risk groups in dengue infections; higher risk, lower risk and no risk (healthy) groups. The classification process was done according to gender, reactance value of the BIA and 'day of fever' on a daily basis diagnosis. BIA measurements have been conducted on 209 serological confirmed dengue patients during their hospitalisation in University Kebangsaan Malaysia Hospital (HUKM), and 223 healthy subjects in Malaysia. The rules were developed based on the 95% confidence interval ranges of reactance value for each risk group obtained from the univariate Analysis of variance (ANOVA) analysis. The system was designed and programmed via Visual Basic 6.0. Graphical user interface were employed, in order to enhance user interaction with the system. The system is able to classify the risk in dengue patients non-invasively with total classification accuracy of 66.7%.