NREM SEEG features showed group-level SOZ separation, while patient-level detection varied by feature
This resampling analysis used archived features from 43 patients, 728 bipolar depth pairs, and 160 clinical seizure-onset-zone (SOZ) pairs in 90-second interictal NREM epochs. For slow-wave phase coupling (MI), mean patient-level AUC was 0.73 (95% CI 0.65–0.80); estimated 80% power required 15 patients at the observed effect and 25 at the lower-CI effect. For 80–300 Hz power, mean AUC was 0.68 (0.60–0.77), and conservative power was 0.64 with all 43 patients. Delta power had mean AUC 0.66 (0.58–0.73) and conservative power 0.77 with 43 patients. Nine patients had too few pairs for any AUC to reach p < 0.05. After Holm correction across patients, no MI result was significant; 3 patients had significant 80–300 Hz power results and 2 had significant delta results. The estimates come from one center and a resample of one cohort; clinical labels are not ground truth, and tied resamples required a normal approximation of the Wilcoxon test.
SEEG · NREM sleep · seizure-onset zone · AUC · statistical power · epilepsy · phase-amplitude coupling
References
4- usesSlow-wave–HFA coupling and HFA power distinguish clinical SOZ depth pairs in 43 patients sci:sha256:9f9a03f5ae81593b8a29528c314db6efb6a4c1b7fb63e4d09e778c7b70330fe7
Archived NREM bipolar-pair features and clinical SOZ labels are the only input of this analysis.
- usesdoi:10.18112/openneuro.ds007118.v1.0.0
One of the three version 1.0.0 OpenNeuro SEEG sleep datasets that supplied the archived features analyzed here.
- usesdoi:10.18112/openneuro.ds007119.v1.0.0
One of the three version 1.0.0 OpenNeuro SEEG sleep datasets that supplied the archived features analyzed here.
- usesdoi:10.18112/openneuro.ds007120.v1.0.0
One of the three version 1.0.0 OpenNeuro SEEG sleep datasets that supplied the archived features analyzed here.
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.