A NumPy pipeline was tested on one SEEG patient; no group result is available

This publication describes a NumPy pipeline for testing whether interictal spectral features distinguish clinically marked seizure-onset-zone (SOZ) SEEG contacts from other contacts within a patient, and whether this separation remains after anatomical-location control. The pipeline computes a per-patient area under the ROC curve (AUC) for each feature and compares raw features with a location-only prior and with features residualized against a leave-one-patient-out normative map. It was tested end to end on sub-HUP116 (9 bipolar pairs, 6 SOZ pairs; 300 s at 500 Hz). Full feature extraction over all SEEG patients did not finish within the 45 min authoring limit. No group result or biomarker performance result is claimed. The source describes an S3 snapshot read on 2026-10-10 from OpenNeuro ds004100 but does not identify its exact release.

SEEG · interictal EEG · seizure onset zone · spectral features · anatomical location · normative map · OpenNeuro ds004100 · signal processing