A Review on EEG-Based Techniques for Early Dementia Classification Using Machine Learning Approaches

Abstract

Dementia, including Mild Cognitive Impairment (MCI) and Alzheimer's Disease (AD), poses significant challenges to healthcare systems worldwide. Early and accurate diagnosis is crucial for effective intervention and management. This study investigates the use of Electroencephalogram (EEG) signals to classify different stages of dementia, focusing on MCI and AD. EEG data were collected from participants diagnosed with MCI, AD, and healthy controls. The signals were preprocessed to remove artifacts and enhance signal quality. Key features, such as spectral power and coherence, were extracted to capture the neural dynamics associated with cognitive decline. Machine learning models, including Support Vector Machines (SVM) and Random Forests, were employed to classify the conditions based on the extracted features. The performance of these models was evaluated using metrics such as accuracy, sensitivity, and specificity. Our results demonstrate that EEG-based classification can achieve high accuracy in differentiating between MCI, AD, and healthy controls. The findings highlight the potential of EEG as a non- invasive and cost-effective tool for early dementia diagnosis. This research contributes to the growing body of literature on neurophysiological biomarkers for dementia and offers insights into the development of clinical applications for EEG-based diagnostics. Future work will focus on validating these findings with larger datasets and exploring the integration of EEG with other biomarkers to enhance diagnostic accuracy.