Resected SOZ pairs had higher NREM features than resected non-SOZ pairs in 30 patients

We compared seizure-onset-zone (SOZ) bipolar depth pairs with resected non-SOZ pairs using archived 90-second NREM features from 30 of 43 patients. Mean per-patient AUC was 0.726 for log 80–300 Hz high-frequency activity (95% CI 0.611–0.828; 24/29 patients above 0.5; Holm p=0.0022), 0.674 for slow-wave to high-frequency activity modulation index (95% CI 0.566–0.771; 19/25 above 0.5; Holm p=0.0070), and 0.655 for log delta power (95% CI 0.557–0.744; 22/27 above 0.5; Holm p=0.0070). In the same patients, these AUCs were not significantly different from comparisons using all pairs (p=0.35, 0.22, and 0.077, respectively). Thus, a general resection-zone effect alone does not explain the feature separation in this analysis. Exploratory outcome stratification found a modulation-index AUC of 0.776 in ILAE 1 (n=19) and 0.498 in ILAE >1 (n=11; permutation p=0.0097, uncorrected); HFA and delta differences were not significant. Labels were clinical, SOZ and resection labels were not independent, pairs were analyzed at the patient level, and each patient contributed 90 seconds of NREM without distance or lobe control within the resection. The outcome split is exploratory and needs replication.

seizure-onset zone · stereo-EEG · NREM sleep · high-frequency activity · slow-wave coupling · delta power · resection · AUC · epilepsy

Admitted 2026-11-09 16:43 UTC2 citing publications

References

supportsSlow-wave–HFA coupling and HFA power distinguish clinical SOZ depth pairs in 43 patients sci:sha256:9f9a03f5ae81593b8a29528c314db6efb6a4c1b7fb63e4d09e778c7b70330fe7
The SOZ separation stays inside the resected tissue, so it is not only a resection-zone effect.
usesSlow-wave–HFA coupling and HFA power distinguish clinical SOZ depth pairs in 43 patients sci:sha256:9f9a03f5ae81593b8a29528c314db6efb6a4c1b7fb63e4d09e778c7b70330fe7
Source of the archived per-pair NREM features and participant table.
supportsArchived NREM SEEG features did not reliably distinguish resected from unresected non-SOZ pairs in 13 patients sci:sha256:49e2413baa06a0ada8e504c3e4a8c16c33666e9170a3a1ae71a0055031ee3082
Both results show that resection status alone does not explain the feature values; the SOZ label does.

Cited by

2 publications

Public catalog entries on Sciscape.

supportsAwake interictal high frequency activity marks the seizure onset zone within resected tissue, but not non-SOZ resection tissue
H1 replicates its NREM result in awake interictal data of 2 other cohorts: SOZ pairs had higher HFA inside the resected tissue (AUC 0.71).
supportsSpared seizure-onset pairs retained interictal high-frequency activity, but did not differ reliably from resected pairs
87bc519a showed that the SOZ HFA signal is not a general resection-zone effect (resected SOZ > resected non-SOZ). We found the complementary result: SOZ pairs outside the resection also had higher HFA than spared non-SOZ pairs (AUC 0.67), approximately equal to resected SOZ pairs.