APPLICATION OF MACHINE LEARNING ALGORITHMS IN THE EARLY DETECTION OF VOLUMETRIC BIOMARKERS BY STRUCTURAL MAGNETIC RESONANCE IMAGING IN THE EARLY STAGES OF ALZHEIMER'S DISEASE

Abstract

This article aimed to evaluate the application of Machine Learning (ML) algorithms in the early detection of volumetric biomarkers by structural Magnetic Resonance Imaging (sMRI) in the early stages of Alzheimer's Disease (AD). This is a descriptive and exploratory literature review, based on the rigorous selection of 24 scientific articles from the PubMed/MEDLINE, Scopus, and Web of Science databases, focusing on the diagnostic accuracy of predictive models in differentiating between cognitively normal individuals and patients with Mild Cognitive Impairment (MCI). The results demonstrated the predictive superiority of supervised algorithms, notably Support Vector Machines (SVM) and Random Forest (RF). It was evident that the simultaneous quantification of multiple biomarkers, specifically hippocampal atrophy associated with entorhinal cortex thinning, provided the highest rates of accuracy and sensitivity in prodromal classification, significantly outperforming isolated anatomical analyses. It is concluded that radiomic automation via ML mitigates traditional diagnostic subjectivity and optimizes early risk stratification. However, the need for external validation in heterogeneous multicenter cohorts and the development of Explainable Artificial Intelligence for safe clinical translation is highlighted.