Machine Learning Approaches for Early Detection and Progression Prediction of Alzheimer’s Disease

Abstract

Alzheimer's disease (AD) is a degenerative neurological condition that declines cognitive ability; early detection is essential for successful treatment and care planning. The OASIS longitudinal dataset is used in this work to investigate machine learning algorithms for predicting early Alzheimer's disease and a collection of advanced machine learning models such as Random Forest, Gaussian Naive Bayes, and Support Vector Machine (SVM), SVM with kernel tricks, and Decision Tree. By focusing on structural MRI and demographic data, this research evaluates and compares these classifiers' predictive performance across several metrics, including precision, F1-score, and recall. In addition to overall classification accuracy, this work also provides an in-depth analysis of critical features like Mini-Mental State Examination (MMSE) scores, age, brain volume ratios (ASF, eTIV, and nWBV), and years of education for demented and non-demented categories. The results demonstrate that the Random Forest and SVM models show superior performance in terms of precision and recall. This research underscores the potential of using machine learning models as robust tools for early Alzheimer’s prediction and contributes valuable insights for clinical decision-making.