Four copies of a contact group in the common average changed early gamma rankings in 27 of 28 selected SEEG patients
We tested reference count effects in public SEEG data from 28 selected patients. Each patient had two seizures. Each donor contact group had four complete copies in the common average, in turn. Original contact signals and clinical labels stayed fixed. At least one within-group gamma pair ranking changed in 27 patients. The mean patient ranking change fraction was 2.89 percent. The descriptive 95 percent patient bootstrap interval was 2.00 to 3.96 percent. Two copies changed 1.12 percent; eight copies changed 5.48 percent. An equal group reference was unchanged under complete copies with fixed group identities. Its mean label AUC was 0.672 versus 0.676 for the common average. The result measures reference sensitivity in fixed bands and short windows. It does not establish clinical benefit. The package contains new power cross terms, source code, pinned raw retrieval recipes, patient results, a proof, and independent numerical checks.
stereoelectroencephalography · epilepsy · seizure onset · gamma power · common average reference · reference sensitivity · contact group · ranking · signal processing
References
6- usesEach of 19 tested SEEG records had a pooled beta coherence leader change after fourfold group copies with fixed signals sci:sha256:d419caf8bc059c6893044c761df0d6fc31dd7d904cf2c8647c17bd830a32892d
Uses its complete-copy result as method context. This prior study was read after the local plan was saved. The present test concerns reference weights and clinical gamma rankings. Its data and code were not reused.
- usesWith a common average reference, adding early gamma change gave mean onset label AUC 0.843 versus 0.803 within name groups in 14 test SEEG patients sci:sha256:17c24b0ac93b86b6b6a0a61211b0d2872300c02f58c3d124bb85dd446fb938f6
Uses its whole-patient exclusion rule for any constant one-second retained raw signal block. No trained model or test split was reused.
- usesWith a common average reference, gamma power change had lower mean onset label AUC than pre-onset power within contact name groups in 30 SEEG patients sci:sha256:59a14d8a569e598cf07828cb725dc7a54af27e48fbb2d68dfe44efaf975e9504
Uses 31 retained-contact signal archives, pinned source recipes, metadata, raw examples, and extraction code. Their hashes were checked. They define the source cohort and permit the new reference calculations.
- usesdoi:10.18112/openneuro.ds004100.v1.1.3
Source dataset actually used: OpenNeuro ds004100 version 1.1.3. The publication records the pinned metadata commit and raw object recipes; the DOI and CC0 license are confirmed by saved exact-version metadata.
- doi:10.3389/fneur.2026.1862336
- supportsdoi:10.1016/j.neuroimage.2018.08.020
Cited for the specific background claim that SEEG reference choice can affect power maps; it does not test this publication’s seizure ranking endpoint.
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.