Interval Type-2 Fuzzy Framework for Healthcare Monitoring and Prediction

Abstract

The popularity of interval type-2 fuzzy logic systems (IT2 FLSs) in the last decade cannot be overemphasized as they have shown superior and more accurate presentation in many applications. In this paper, we investigate healthcare monitoring and prediction using an interval type-2 fuzzy logic system (IT2FLS) based on Mamdani fuzzy inference. Also, the study investigates healthcare monitoring and prediction problems using conventional type-1 fuzzy logic system (T1FLS) for comparison purposes. The empirical comparison was carried out on the developed work using cardiovascular disease patients’ health datasets. The study observed that interval type-2 fuzzy logic could cope with more information and could handle more uncertainties in health data compared to it counterpart. The Root Mean Squared Errors (RMSEs) evaluation results of 0.018 and 0.0006 were observed for type-1 fuzzy logic and interval type-2 fuzzy logic respectively in cardiac shock level prediction experiment, which showed the superiority of IT2FLS paradigm over T1FLS.KeywordsFuzzy logicFuzzy inferenceType-2 fuzzy setsFuzzy controllerDefuzzificationCardiac patient