“Adaptive Particle Swarm Grey Wolf Optimization (APSGWO) in Early Alzheimer’s Disease Identification ”

Abstract

Alzheimer’s disease is a progressive neurodegenerative disorder that significantly influencing routine activities and overall health. The brain changes make the diagnosis challenging for the crucial function of early identification of Alzheimer’s disease (AD). The integration of population diversity based adaptive switching in a Hybrid Particle Swarm–Grey Wolf Optimization (PSO–GWO) framework for feature selection is presented in this work. This novelty of the methodology dynamically emphasizes the exploration and exploitation strategy during optimization. The highly discriminative feature subset selection is selected by the adaptive mechanism and it eliminates redundancy. For accurate AD detection, an XGBoost classifier is trained using the improved features. The proposed framework enhances feature selection capability, accelerates convergence and greater classification accuracy than conventional techniques. The significance of population variety in metaheuristic optimization is highlighted and supports early detection in practical applications. For the early diagnosis and clinical decision-making, the proposed work offers a computationally effective framework for automated AD prediction.