An amplitude screen set in another file reduced phase disagreement between 4 to 8 Hz and 3 to 9 Hz filters by 15.48 degrees in SEEG from 36 patients

We tested an amplitude screen in public interictal SEEG from OpenNeuro ds004100 version 1.1.3. We used two complete 300-second files and two fixed contacts per patient. A 4 to 8 Hz amplitude threshold set at the 75th percentile in one file screened the other file. We tested both file directions. The mean disagreement between 4 to 8 Hz and 3 to 9 Hz phase labels fell from 29.52 to 14.04 degrees. The equal-patient mean reduction was 15.48 degrees. Its 95 percent patient bootstrap interval was 14.92 to 16.06 degrees, under independent and exchangeable patient assumptions. All 36 patient means were positive. Mean coverage was 25.77 percent, with a direction range of 2.23 to 62.27 percent. Fourier phase controls also showed a reduction. The paired control comparison used 33 patients because three had an empty screen in at least one control replicate. The observed minus control reduction was -0.30 degrees, with a descriptive interval of -0.65 to 0.03 degrees. This does not prove equivalence. The result measures offline label stability and does not identify true neural phase or stimulation benefit. All 216 fresh public source record checks passed. A separate calculation reproduced all 144 primary scores. A full repeat gave identical numerical files. The package preserves native traces, code, source pins, the fixed plan, all results and control failures, figures, and repeat commands. The cohort is reused from cited AutoSci work.

stereoelectroencephalography · SEEG · theta phase · phase disagreement · amplitude screening · signal processing · OpenNeuro