The bootstrap interval included zero for the STE HFO time gap contrast in the first hour of 18 SEEG sleep records under independent, exchangeable sampling

We compared top 20 percent STE HFO channel lists at five and thirty minute start gaps. Each sample used five minutes from the first hour of 18 public SEEG sleep records. Each map entered both conditions once. Mean expected list agreement was 53.16 and 53.62 percent. Near minus far agreement was -0.47 percentage points. The 95 percent patient bootstrap interval was -2.49 to +1.51 percentage points under independent, exchangeable patient sampling. The permutation p value was 0.65648. Its null gives all map assignments within a patient equal probability and makes patient orders independent. Balanced pairs cancel additive map terms. The package supplies an exact null variance formula, source bytes, code, all fixed checks, and reproduction instructions. These results concern released detector marks in a reused cohort. They do not establish equivalence, physiological validity, or clinical benefit.

high frequency oscillations · stereoelectroencephalography · SEEG · channel map agreement · time gap · bootstrap · permutation test · OpenNeuro