Removing six seconds with high beta power reduced mean seizure-onset label AUC for 80 to 150 Hz power concentration by 0.043 compared with shifted removal in 34 SEEG patients with two 60 s clips each

This study tested whether high-gamma power concentration depends on seconds with high beta power. It used public OpenNeuro ds004100 version 1.1.3 data from 34 SEEG patients. Each patient had two 60 s interictal clips. There were 3,325 bipolar pairs and 251 seizure-onset pairs. For each pair, the test removed its six highest beta-power seconds. It compared the remaining 80 to 150 Hz concentration with removal after 59 circular shifts of the same masks. Mean seizure-onset label AUC was 0.661 after aligned removal and 0.704 after shifted removal. The paired patient mean difference was -0.04346. Its 95 percent patient bootstrap interval was -0.06958 to -0.01972. A two-sided patient sign-flip test gave p = 0.001005. This test assumes independent patients and null sign symmetry. The within-name-group difference was smaller, and its interval included zero. The result shows time coincidence between label separation and high beta power. It does not identify oscillations, spikes, or artifacts. All 72 raw prefixes and 221 metadata files were checked against pinned public sources. The fresh rebuild matched all 748 archived power-array fields. An offline rerun reproduced all statistical files and score arrays. The package preserves the local plan, source records, powers, selectors, all shifted scores, code, figures, independent checks, and reproduction commands.

SEEG · epilepsy · high-gamma power · beta power · power concentration · AUC · circular shift · intracranial EEG