Neuro-Fuzzy Optimization Technique for Enhancing Accuracy of Predication Hepatitis Disease

Abstract

Hepatitis remains one of the most widespread infectious diseases globally, posing significant challenges to public health systems and the WHO(WHO). Its inherent ambiguity, clinical variability, and data heterogeneity make accurate diagnosis a complex task. This study presents a hybrid optimization model that integrates Particle Swarm Optimization (PSO) with the Dragonfly Algorithm (DA) to improve the performance of an Adaptive Neuro-Fuzzy Inference System (ANFIS). The proposed hybrid algorithm, referred to as PSODA, is designed to optimize two critical components in the ANFIS structure: the number of membership functions in the initial layer and the model’s learning rate. By balancing the exploration behavior of DA and the exploitation capabilities of PSO, the PSODA framework enhances both the stability and accuracy of the diagnostic system. Experimental results suggest that the PSODA-ANFIS model achieves a classification accuracy of 90%, outperforming standard ANFIS as well as models optimized individually by PSO or Grey Wolf Optimizer (GWO). This hybrid model demonstrates promising potential for clinical decision support in the diagnosis of hepatitis and similar medical conditions characterized by uncertainty and complexity.