Bipolar SEEG gave lower mean onset AUC than common average for a spectral ratio in 60-second samples from 12 HUP patients

This study tested whether a power ratio ranked labelled seizure-onset pairs differently with two SEEG pipelines. It used the first 60 seconds of two interictal runs from 12 HUP patients in OpenNeuro ds004100, version 1.1.3. The score was log10 of 30-80 Hz power, excluding 58-62 Hz, divided by 1-20 Hz power. The primary mean AUC was 0.448 with adjacent bipolar signals and 0.530 with common average signals. The mean paired change was -0.082. The 95 percent percentile bootstrap interval was [-0.158, -0.0005]. The planned two-sided sign-flip test gave p = 0.0691. The repeat run gave a change of -0.101 and p = 0.1024. Gamma power alone gave an opposite pipeline difference in a planned secondary check. The original 36-patient plan was reduced to the first 12 patients after network throughput fell. The scope change preceded the group comparison; one pilot result had been seen. The result is descriptive and does not establish a clinical benefit or a population-wide reference choice. The package includes source metadata, source hashes, one full reader-check file, all derived features, code, figures, and reproduction instructions. It also preserves the full-cohort retrieval recipe.

stereoelectroencephalography · SEEG · intracranial EEG · reference choice · bipolar montage · common average reference · area under the curve · AUC · spectral power · seizure onset · OpenNeuro ds004100