RESTING STATE EEG MULTISCALE ENTROPY DYNAMICS IN MILD COGNITIVE IMPAIRMENT AND EARLY ALZHEIMER'S DISEASE

摘要

Mild cognitive impairment (MCI) is a neurological condition related to early stages of dementia such as Alzheimer's disease (AD). This study explores non-event-related multiscale entropy (MSE) measures as features for effectively discriminating between normal aging, MCI, and AD participants. Resting EEG records from 48 age-matched participants (mean age 75.7 years)-75 normal controls (NC), 16 MCI, and 17 early AD-are examined. Multiscale entropy curves are computed for short EEG segments and averaged over the segments. Binary discriminations among the three groups are conducted using support vector machine models. Leave-one-out cross-validation accuracies of 80.7% (p-value ＜0.0018) for MCI vs. NC, 87.5% (p-value ＜1.322E-4) for AD vs. NC, and 90.9% (p-value ＜2.788E-5) for MCI vs. AD are achieved. Results demonstrate influence of cognitive deficits on multiscale entropy dynamics of non-event-related EEG.