The mean decrease in the ratio of gamma power to 1–30 Hz power was larger at SOZ pairs within shafts in 29 ds004100 SEEG patients at 10–30 seconds after seizure end

We tested whether gamma decrease after marked seizure end exceeds low frequency decrease more at seizure onset zone (SOZ) pairs than at other pairs. The source was public SEEG dataset OpenNeuro ds004100 version 1.1.3. The fixed cohort had 32 patients and two records per patient. The main test used 29 patients with both labels on the same shaft. The score was the decrease in the gamma-to-1–30 Hz power ratio at 10–30 seconds after end. The mean within-shaft SOZ-minus-other contrast was 2.50 (1.27 to 3.78) dB, with a 95 percent patient bootstrap interval. The estimated two-sided sign-flip p-value was 0.000565 under independent patients and symmetric null patient effects. The ratio score had mean within-shaft AUC 0.642, compared with 0.790 for gamma decrease. The paired AUC difference was -0.148 (-0.233 to -0.062). This secondary interval is descriptive. The score cancels uniform waveform amplitude scale within each two-second window. It still depends on both frequency bands. The package preserves new band power arrays, public source pins, a raw example, code, all fixed checks, and repeat instructions. This is a secondary analysis of a reused cohort. It does not establish a neural cause, surgical benefit, or new-patient performance.

stereoelectroencephalography · epilepsy · postictal activity · gamma power · spectral power ratio · seizure onset zone · secondary analysis