In 18 SEEG sleep records, the STE HFO distance contrast between minutes 10 to 20 and 60 to 70 had a 95 percent patient bootstrap interval that included zero

This study compared STE HFO channel maps in 18 public SEEG sleep records from OpenNeuro ds005398, version 1.1.3. The first map covered minutes 0 to 10. The next map covered minutes 10 to 20. The later map covered minutes 60 to 70. Each patient had the same selected event count in all three maps. The code calculated the exact expected squared Hellinger distance for independent selection without replacement. The mean later distance minus next distance was +0.0767. Its 95 percent patient bootstrap interval was -0.0125 to +0.1700. The two-sided sign-flip p value was 0.1293. Thirteen patients had a positive contrast and five had a negative contrast. The evidence did not establish a common mean increase or no effect. Patient inference requires the stated sampling and null sign assumptions. This is a secondary analysis of candidate detector marks in the same cohort as prior AutoSci work. It does not identify a neural cause or clinical benefit. All 94 source objects passed new public checks. Independent counts, probability calculations, and inference checks passed. Two new-process repeats reproduced all 24 scientific analysis files byte for byte, including a repeat from fresh public inputs. The package preserves source tables, provenance, code, results, figures, and instructions. This version corrects the file check for metadata added during AutoSci download.

high-frequency oscillations · stereoelectroencephalography · SEEG · epilepsy · channel maps · squared Hellinger distance · event sampling · reproducibility