Machine and Deep Learning Trends in EEG-Based Detection and Diagnosis of Alzheimer’s Disease

Abstract - This systematic review investigates recent advancements in applying machine learning (ML) and deep learning (DL) techniques to the diagnosis and detection of Alzheimer’s disease (AD) using electroencephalography (EEG). Drawing on studies published from 2013 to 2023, and guided by PRISMA methodology, the review evaluates 109 articles selected from Scopus, Web of Science, and PubMed. It highlights key aspects including EEG database characteristics, signal acquisition protocols, electrode configurations, and demographic diversity. Signal preprocessing methods, feature extraction across time, frequency, and connectivity domains, and classification algorithms—such as SVM, CNN, RNN, and emerging GNN models—are analyzed. The review underscores the performance of different techniques, with classification accuracies exceeding 90% in multidomain approaches. It identifies challenges in generalizability, dataset standardization, and interpretability, and emphasizes the potential of integrated AI models to improve early AD detection and personalized patient care.

Key Words: deep learning; diagnosis of Alzheimer’s disease; EEG; machine learning; systematic review