Explainable AI-Driven Handwriting Analysis for Early Alzheimer’s Disease Detection Using Machine Learning and Ensemble Models

Alzheimer’s disease (AD) affects millions of people worldwide and is the primary cause of dementia. It is also one of the most common and deadly neurological conditions. Effective intervention requires an early diagnosis, particularly in underdeveloped nations like Bangladesh, where diagnostic resources and healthcare infrastructure are frequently limited. This research aims to use cutting-edge artificial intelligence (AI) and machine learning (ML) models to enhance early detection of AD. Random Forest achieved the highest accuracy of 91.00% among different ML and ensemble models that were trained on a comprehensive dataset that included neurological and cognitive factors. Explainable AI approaches like LIME and ELI5, which provide transparency by highlighting the significant features driving model predictions, were utilized to illustrate the decision-making process of the model further. These methods help clinicians comprehend and believe in AI-driven insights by simplifying sophisticated AI models. The research highlights the viability of AI-driven methods for early AD detection, particularly in resource-constrained settings, and illustrates how explainable AI can be integrated into healthcare settings. The findings support the usage of sophisticated AI models for the improvement of diagnosis accuracy and patient outcomes. To manage Alzheimer’s disease in clinical settings, this research offers a viable avenue for using AI-based predictive techniques.