Flower Pollination Algorithm in Optimization of Interval Type-2 Fuzzy for Telemedical Problem

Abstract

Determining the best parameters of the membership functions (MFs) in developing interval type-2 fuzzy logic controller (IT2FLC) is always difficult and complex because of the presence of high level of uncertainty and imprecision. Flower Pollination Algorithm (FPA) is a nature-inspired global optimization technique that imitates the pollination behavior of flowering plants, which avoids trapping in local optimum solution. This paper describes the optimization of IT2FLC using FPA for telemedical problem and applies to predict shock level in cardiac patients. The optimized IT2FLC is used to handle the uncertainties inherent in the membership functions (MFs) of the type-2 using cardiac patients’ data for remote vital signs monitoring. The parameters of the MFs of the IT2FLC had not been previously optimized as this study considers applying Flower Pollination Algorithm (FPA) to optimize interval Footprint of Uncertainty (FOU) MFs of IT2FLS. We use the FPA to find the best MFs parameters of the IT2FL and the choice of FPA is because of the linear nature of the algorithm which gives attractive performances when compared to state-of-the-art algorithms. The empirical comparison is made on the designed system using cardiac patients’ health datasets obtained both from real-time and clinical scenarios. Simulation results indicate that FPA-based interval type-2 fuzzy algorithm outperforms those results of a single IT2FL algorithm.