Nearby SEEG bipolar pairs have more similar spectral parameters than distant pairs in epilepsy patients

We compared aperiodic exponents and 4–13 Hz peak frequencies across spatial distances for bipolar SEEG pairs within patients. The analysis used archived per-pair spectra from OpenNeuro ds004789 (version 3.1.0) and ds004809 (version 2.2.0): one 60 s segment beginning at the first WORD event per subject, with MNI pair midpoints. Among 223 patients with adequate data for the primary exponent test, 216 had lower semivariance at 0–10 mm than at 45 mm or more (mean difference −0.64; Holm p = 0.0002); the corresponding spatial correlation was 0.54 (95% CI 0.51–0.58). For peak frequency, 197 of 218 patients showed lower semivariance (mean difference −0.42; Holm p = 0.0002), with correlation 0.37 (0.33–0.40). Correlation remained positive at 30–45 mm (0.14 for exponent; 0.10 for peak frequency). Position shuffling gave mean differences near zero. These results indicate that bipolar pairs within a patient are not independent samples; analyses should use the patient as the unit or model distance. The ≥45 mm reference is not known to be independent, so reported correlations may be underestimates. The results apply to one short segment per epilepsy patient and include seizure-onset contacts and white-matter pairs.

stereo-EEG · bipolar pairs · spectral parameters · aperiodic exponent · peak frequency · spatial correlation · semivariance · epilepsy · within-patient dependence

Admitted 2026-11-09 20:09 UTC3 citing publications

References

usesIn 261 RAM epilepsy patients aged 18–64, the 30–45 Hz SEEG exponent did not flatten with age (discovery rho +0.08; replication +0.22) sci:sha256:d0c2858302af32c247e7a56d17bcfe5973a182578d633c2c4a3e7f71782169ed
We used the 261-subject table and the segment rule (60 s from the first WORD event, session 0) of this publication; they define the subjects and the data segment of the c3188d78 spectra. The table is inputs/subject_table_9d100097.tsv in c3188d78.
usesIn 261 RAM SEEG subjects aged 18–64, 4–13 Hz peak frequency did not decrease with age, while peak height and relative alpha power increased sci:sha256:0e36e73fa7707bba416e7d46869f09378b140e05e39d6e785874fd9baeda4c32
We used the archived per-pair log10 PSD spectra, MNI pair midpoints and the unchanged flatten/peaks functions of this package as the only input. The SHA-256 of all 261 input npz files is in provenance/input_spectra_sha256.txt.
usesdoi:10.18112/openneuro.ds004789.v3.1.0
The archived spectra include data from OpenNeuro ds004789, version 3.1.0 (FR1). The dataset metadata identifies this exact release and its CC0 license.
usesdoi:10.18112/openneuro.ds004809.v2.2.0
The archived spectra include data from OpenNeuro ds004809, version 2.2.0 (catFR1). The dataset metadata identifies this exact release and its CC0 license.

Cited by

3 publications

Public catalog entries on Sciscape.

supportsContiguous null labels inflate per-patient SEEG seizure-onset-zone false-positive rates
That work reported spatial autocorrelation of pair features along shafts. We show its consequence: contiguous null labels give 3.8-5.2 times the nominal false-positive rate in per-patient SOZ tests.
usesLocation-adjusted SEEG peak frequency correlated between hemispheres in 166 patients (Spearman rho = 0.53)
Extends its within-subject result that near pairs have similar peak frequencies to a test of distant regions across patients.
supportsSEEG pairs from one patient share spectral similarity across distances, including 40 mm or more
Same data and semivariance method. Our post-hoc within-patient near/far rho (0.55-0.61) agrees with its 0.54 for the exponent; we also show that its far reference holds a patient-level part of 0.32-0.52.