Median scale correction reduced mean count error for a 10 percent 80–150 Hz power cutoff, with 118–122 Hz removed, in two 60-second SEEG clips from each of 34 patients

We tested whether median scale correction made a fixed SEEG power cutoff select a more stable number of pairs between recordings. The data were two 60-second interictal clips from each of 34 patients in OpenNeuro ds004100 version 1.1.3. The score was mean 80–150 Hz power, with 118–122 Hz removed. Each source selected its highest-power 10 percent of pairs, rounded up. The correction used the target median from all eligible pairs. Across 3,325 pairs, mean count error fell from 2.42 to 1.84 percentage points. The mean change was −0.58 percentage points. Its 95 percent patient bootstrap interval was −1.07 to −0.11. The exact sign-flip p value was 0.0256 under independent-patient and sign-symmetry assumptions. Fifteen patients improved, eight worsened, and eleven had no change. SOZ recall stability did not show a clear gain. The target scale used the complete clip. The study reused a published cohort and does not show clinical benefit. All 72 pinned public source prefixes were retrieved. All 68 power files matched exactly. Code, inputs, selection tables, checks, figures, and repeat instructions are preserved.

stereoelectroencephalography · SEEG · power cutoff · median scale correction · cutoff transfer · high-frequency power · epilepsy · measurement stability