EEG Feature Extraction for Neurodegenerative Disease Diagnosis: A Review

Electroencephalography (EEG) has emerged as a promising tool for the diagnosis and monitoring of neurodegenerative disorders such as Alzheimer's disease (AD), frontotemporal dementia (FTD), and Parkinson's disease (PD).Characterized by progressive neuronal loss and overlapping clinical symptoms, these conditions require early detection and reliable biomarkers for improved patient outcomes.EEG offers noninvasiveness, costeffectiveness, and high temporal resolution, enabling the capture of subtle alterations in brain dynamics that precede clinical manifestations.However, raw EEG signals are nonlinear, noisy, and heterogeneous, necessitating effective feature extraction techniques to achieve disease specificity.This study provides a narrative review of EEG-based feature extraction methods across spatial, temporal, frequency, and time-frequency domains.Representative studies were selected based on their relevance, methodological significance, and impact in neurodegenerative disease research.We further examine their integration with machine learning and deep learning classifiers, highlighting applications in distinguishing AD from FTD, detecting early cognitive decline, and identifying electrophysiological signatures of PD.Despite these advances, challenges remain regarding reproducibility, standardization, and the balance between accuracy and interpretability.Future directions include the development of standardized analysis pipelines, large-scale multicenter datasets, and explainable AI approaches, as well as the adoption of wearable EEG systems for real-time screening.Taken together, EEG-based feature extraction holds great potential as a clinically viable biomarker strategy for the early diagnosis and monitoring of neurodegenerative diseases.