حجت قیمت گر

Abstract—__Discriminating eyes-open (EO) from eyes-closed (EC) in resting-state EEG is a standard benchmark for evaluating compact biomarkers. This study quantifies normalized permutation entropy (PE) in band-limited alpha (8–13 Hz) and gamma (30–45 Hz) activity using EEGMMIDB baseline runs (R01/R02; 64 channels, 160 Hz; 60 s per condition). Signals were band-pass filtered (1–50 Hz), notched at 50 Hz, average-referenced, segmented into 2-s epochs, and screened with a two-stage automated quality-control procedure (amplitude and peak-to-peak thresholds). Subjects were included if both EO and EC retained at least 30 clean epochs, yielding N = 79.A 2×2 repeated-measures ANOVA with within-subject factors band (alpha, gamma) and condition (EO, EC) revealed significant main effects of both band and condition, as well as a significant band × condition interaction (all p < 10.6). At the group level, PE showed a pronounced band effect: gamma PE was near ceiling, whereas alpha PE was substantially lower, with EO exhibiting a small but consistent increase relative to EC in both bands. In a balanced 60-s subset (30 clean epochs per condition; N = 20), second-by-second PE trajectories remained stable over time, suggesting that PE primarily reflects band-dependent dynamics rather than slow temporal drift. Overall, these findings support band-specific PE as a compact feature for characterizing resting EO/EC states under automated quality control .