SEEG variance after mean removal at 20 to 100 ms exceeded a baseline of equal length in eight epilepsy sessions

We tested residual voltage variance at 20 to 100 ms after pulses in public OpenNeuro ds004696 version 1.0.1. We used eight epilepsy SEEG sessions, 86 stimulation sites, and 886 site/readout pairs. The calculation removed each pulse temporal mean and a separate mean waveform for each five-pulse group. With an equal-length baseline, the geometric variance ratio was 1.609, with a descriptive 95% interval of 1.355 to 1.935. With a longer baseline, the ratio was 0.512. Each patient mean changed from below one to above one. Fixed subsets supported the average direction. A stationary-noise proof showed how temporal mean removal can create a duration effect. The archive preserves voltage products, selected voltage epochs, source metadata, results, figures, tables, and reproduction records. GPL analysis and extraction scripts are excluded from the SCI payload. The result concerns processed clinical voltage. It does not establish a neural cause or online interface performance.

SEEG · residual variance · electrical stimulation · baseline · signal measurement · epilepsy