Gamma power spread had a 0.005 dB median onset boundary contrast under supplied reference in 34 SEEG patients with two 60-second records each

This study tested gamma power spread at adjacent contacts with opposite clinical seizure-onset labels on the same named shaft. The public source was OpenNeuro ds004100 version 1.1.3. Each patient supplied the first 60 seconds of two interictal records. The feature was the interquartile range of two-second gamma powers in dB. Gamma power used 30 to 80 Hz and excluded 55 to 65 Hz. The fixed primary analysis used the supplied reference and equal patient weights. In 34 patients with 126 named boundaries, the median onset-minus-other contrast was 0.0053 dB. There were 18 positive and 16 negative patient contrasts. The exact two-sided sign test gave p = 0.8642. The median interval was -0.1404 to 0.1005 dB, with 97.569 percent nominal coverage under the stated sampling assumptions. This test did not establish a common directional association. It does not establish feature equality or clinical equivalence. Contact names do not prove actual spacing or tissue type. The package preserves window powers for 4,260 contacts in 36 patients. It includes source metadata, pinned raw retrieval recipes, code, patient results, figures, and independent checks. A second complete extraction and analysis reproduced saved numerical outputs.

stereoelectroencephalography · gamma power · power spread · seizure onset · interquartile range · intracranial EEG