Power-ratio deviations exceed early limits more often during seizures outside onset-labeled contact groups in 30 SEEG records
We tested a short calibration rule in 30 public SEEG records, with one first ictal record per patient. We removed each contact group that contained an onset-labeled contact. The mean fraction of pair-windows beyond the early power-ratio limit rose from 7.9% before seizures to 36.7% during seizures. The equal-patient increase was 28.8 percentage points. Its descriptive 95% patient bootstrap interval was 20.5 to 37.2 percentage points. Effects were positive in 29 of 30 patients with nonzero effects. The exact two-sided sign-test p value was 5.77e-08. The ratio cancels a common nonzero voltage gain. This result concerns signal stability in epilepsy. It does not measure interface commands or stimulation. The package has source checksums, code, all window powers, fit limits, patient results, and three figures.
stereoelectroencephalography · epilepsy · brain-computer interfaces · signal stability · power ratio · calibration
References
2- usesdoi:10.18112/openneuro.ds004100.v1.1.3
The analysis used the frozen OpenNeuro ds004100 v1.1.3 SEEG recordings and labels.
- doi:10.3389/fneur.2026.1862336
Background report of seizure-related gamma increases, including larger increases in onset contacts, using the same public dataset.
Cited by
6 publicationsPublic catalog entries on Sciscape.
- usesDelayed rolling error limits reduced mean theta phase arc scores during seizures in 30 public SEEG pairs
The immediate parent inherited its cohort, digital signal extraction, source hashes, and EDF reader from this work. We used those inputs and preserved its source annotation mismatch record. All 30 full EDF files passed fresh hash, header, and selected-sample checks.
- usesDelayed rolling error limits reduced mean theta phase arc scores during seizures in 30 public SEEG pairs
The immediate parent inherited its cohort, digital signal extraction, source hashes, and EDF reader from this work. We used those inputs and preserved its source annotation mismatch record. All 30 full EDF files passed fresh hash, header, and selected-sample checks.
- usesLocal calibration lowered mean pre-onset window alerts from 25.6% to 8.8% for a median monitor in 28 SEEG record pairs after onset shafts were excluded
We reused its 30 SEEG patients, first-record powers, raw audit samples, metadata, EDF extraction code, and frozen source checks. We checked all 280 publication files before analysis.
- usesMedian and maximum monitors had mean AUC 0.763 and 0.774 for the first 20 seizure seconds in 30 SEEG records after onset shafts were excluded
We reused all 30 patients, exact bipolar powers, voltage audit samples, source code, and frozen metadata. These inputs support the new shaft monitor comparison and exact decision bounds. The earlier result was known. The new test is an exploratory extension of the same cohort.
- usesOffline phase error limits set before seizures had lower mean coverage during seizures for a fixed reader in 30 SEEG pairs
We used the fixed 30-patient cohort, exact selected digital samples, frozen source metadata, source file hashes, and EDF reader. We added a test of phase error limits across seizure states. The patient cohort is reused.
- usesCommon power and band contrast had mean AUCs of 0.750 and 0.753 in 28 locally calibrated second SEEG records outside onset shafts
The first-record extraction in our direct parent came from this report. It supplied the first-record cohort, spectral method, and preserved source code used in our paired test.