Mean eight-channel SEEG selection turnover was 5.26 percent after 96 to 104 Hz removal in two-second baselines from 21 patients

We tested if a 100 Hz mask changed a SEEG power selection rule more than an 84 Hz control. We used all 35 released records from 21 patients in OpenNeuro ds006519, version 1.0.0. The rule selected eight bipolar channels with the largest 70 to 110 Hz power. Turnover is the percentage of channel choices that changed after removal. Mean early turnover was 5.26 percent for removal of 96 to 104 Hz. It was 6.72 percent for removal of 80 to 88 Hz. The paired patient difference was -1.46 percentage points. The descriptive 95 percent patient bootstrap interval was -5.06 to 2.26 points. The exact sign reference gave p = 0.4873. It assumes independent patient contrasts with symmetric signs under the null. The primary comparison did not show greater turnover for the 100 Hz mask. This does not establish equivalence. Full power choices retained a mean of 67.84 percent in the next two seconds. We did not measure decoding accuracy, neural source identity, or stimulation effects. Two complete raw retrievals, direct Fourier checks, independent selections, and numerical replay passed. The package preserves spectra, audit voltage, code, source hashes, figures, and repeat instructions. A saved amendment handles one four-second source baseline.

stereoelectroencephalography · SEEG · channel selection · power spectrum · frequency masking · 100 Hz · repeatability