Early Detection Of Alzheimer's Disease Using Cognitive Features A Voting-Based Ensemble Machine Learning Approach

Abstract

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that significantly impairs cognitive functions and quality of life, making early diagnosis critical for effective intervention and management. Traditional diagnostic methods often rely on clinical evaluations and neuroimaging techniques, which can be expensive, time-consuming, and not readily accessible in many settings. This study proposes a novel machine learning framework that utilizes cognitive features derived from neuropsychological assessments to facilitate the early detection of Alzheimer’s disease. By leveraging a voting-based ensemble learning approach, which combines the predictive strengths of multiple classifiers—including logistic regression, random forests, gradient boosting, and support vector machines—the model aims to improve classification accuracy, robustness, and generalizability across different patient populations. The dataset comprises various cognitive test scores representative of memory, attention, executive function, and language abilities, along with corresponding diagnostic labels categorized into normal control, mild cognitive impairment, and Alzheimer’s disease groups. Comprehensive preprocessing steps such as imputation of missing data, feature scaling, and selection are performed to enhance model performance. The ensemble model is trained and validated using stratified cross-validation techniques to mitigate overfitting, and its predictive capability is assessed through standard metrics such as accuracy, precision, recall, F1-score, and area under the ROC curve. Results demonstrate that the voting-based ensemble outperforms individual classifiers, highlighting the advantage of aggregating diverse decision boundaries to capture the complex patterns inherent in cognitive decline.