Dynamic Entropic Mapping of Alzheimer's Neuropathology: Eyes-Open and Eyes-Closed EEG Biomarkers through Wavelet Entropy Analysis.

Abstract

Background: Alzheimer's Disease (AD) is characterized by amyloid-β plaques and tau tangles, while current diagnostic tools are often invasive and costly. Electroencephalography (EEG) offers a non-invasive alternative, with alpha rhythm abnormalities as key features. Impaired alpha reactivity during the transition from Eyes-Closed (EC) to Eyes-Open (EO) reflects early thalamocortical dysfunction.

Objective: This study employs EO/EC wavelet entropy analysis to assess neurodynamic features across frequency bands, aiming to explore entropy-based EEG markers for early AD detection.

Methods: This cross-sectional study enrolled 60 participants (30 AD and 30 controls). EEG was recorded during 60-second EC and Eyes-Open (EO) states using a 20-channel system. Continuous Wavelet Transform (CWT) and multiscale entropy were used to assess complexity differences across bands and regions between groups.

Results: AD patients showed reduced alpha entropy differences between EC and EO states. Multiscale entropy difference (ΔEN α ) analysis revealed weaker modulation in β-α and θ-δ bands compared with HC. Occipital ΔEN α correlated significantly with MMSE, with the strongest association in the alpha band (r = 0.9000, P < 0.001).

Discussion: This study applies dual-state wavelet entropy analysis to reveal impaired neural reactivity in AD during eyes-open and eyes-closed transitions. The ΔEN α metric distinguishes AD from controls and correlates with cognitive decline, reflecting reduced neural flexibility and disrupted frequency-specific network dynamics.

Conclusion: This study shows that dual-state wavelet entropy analysis, especially ΔEN α, is a noninvasive tool for detecting neurodynamic abnormalities in AD. It reflects loss of state-dependent responsiveness and neural complexity, with potential for early screening and objective evaluation of treatment.