Awake interictal high frequency activity marks the seizure onset zone within resected tissue, but not non-SOZ resection tissue

We tested whether interictal high frequency activity (HFA, 80–150 Hz) distinguishes seizure onset zone (SOZ) pairs within resected tissue (H1), and whether it distinguishes resected from spared pairs outside the SOZ (H2), using archived per-pair feature tables from two awake SEEG cohorts (32 + 29 patients with included rows). No raw EEG was read. Patient-level Mann–Whitney AUCs were pooled across cohorts, with two-sided sign-flip tests, bootstrap 95% confidence intervals, and Holm correction across six tests. For H1, resected SOZ pairs had higher HFA than resected non-SOZ pairs (mean AUC 0.708, 95% CI 0.594–0.816, n=24, Holm p=0.014); line length also separated the groups (AUC 0.694, Holm p=0.002). This HFA separation was similar to the all-pair SOZ separation in the same patients (difference +0.032, 95% CI −0.066 to 0.123, p=0.52). For H2, HFA did not distinguish resected from spared non-SOZ pairs (AUC 0.456, 95% CI 0.382–0.533, n=28, Holm p=0.28); line length gave a similar result (AUC 0.443, Holm p=0.28). Spike rate was somewhat higher in resected non-SOZ pairs (AUC 0.573, Holm p=0.20), not significant. The findings indicate that, in these cohorts, interictal HFA marks clinical SOZ pairs within the resected tissue but does not identify additional tissue removed outside the SOZ. The clinical labels do not include surgery type or seizure outcome; regional differences can mask or imitate H2 effects; bipolar pairs share contacts; and the ds003876 subsets are small (7–9 patients).

epilepsy · stereo-EEG · interictal · high frequency activity · seizure onset zone · resection · AUC · clinical labels