MindCare: Enabling Accessible Healthcare Through a Web-Based Predictive Tool for Alzheimer’s Disease Diagnosis

Abstract

Alzheimer’s disease (AD) is a progressive neurodegenerative disease and one of the leading causes of cognitive decline in the older population. Conventional diagnostic tools often rely on clinical symptoms that appear in later stages, highlighting the need for an accurate and accessible tool for early prediction. This study introduces MindCare, a web-based application for AD prediction using clinical and neuropsychological data. Multiple machine learning algorithms were implemented to generate predictive models, evaluate their performance, and highlight important features that contribute to classification. This application provides real-time predictions for new patient instances, comparative visualization of classifier performance, and correlation analysis of features with the outcome. Furthermore, an administrator module supports model training, continuous updating, and performance monitoring to ensure reliability over time. Decision Tree (DT) produced the most effective results, achieving an accuracy of 98% with an average F1 score of 95.7%. The findings demonstrate that integrating machine learning into a decision-support system can provide accurate and interpretable predictions of AD. MindCare offers a practical, user-friendly tool that combines predictive analytics with clinical insight, highlighting its potential to predict early and contribute to improved care strategies for individuals at risk of Alzheimer’s disease.