A Multimodal Hybrid Ensemble Approach Integrating ML and Explainable AI for Alzheimer's Prediction

A severe brain disorder named Alzheimer's disease (AD), which worsens the memory and operational impairments. It is crucial to detect the disease at an early stage, which enhances the patient's health and therapeutic planning. In this paper, we proposed an ensemble approach integrating diverse learning classifiers such as Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Naïve Bayes (NB) to optimize the prognostic efficiency and dependability. This proposed architecture comprises data preprocessing, feature extraction, and a feature selection process based on a voting method for effective classification performance. The outcome shows that the proposed ensemble model achieves the highest accuracy and F1 score of 99.56%, outperforming traditional existing models stated in the study. This model offers a promising solution for achieving optimized prognostic accuracy, reliability, and practicality for predicting AD.