Interictal 1–40 Hz power showed no distance gradient, while spike rate fell with SOZ distance in 30 HUP patients
We tested whether interictal SEEG features varied with distance from the seizure-onset zone (SOZ) in 30 HUP patients using archived per-pair features from 120-s clips and fsaverage contact coordinates. For each non-SOZ bipolar pair, we measured the 3D distance to the nearest SOZ pair and calculated within-patient Spearman correlations. Broadband log 1–40 Hz power showed no distance gradient (mean rho = −0.035, 95% CI −0.141 to 0.056; Holm p = 0.44) and was not more related to the patient's own SOZ than to pseudo-SOZs from other patients (Holm p = 0.38). In a post hoc comparison, SOZ pairs had a within-patient z score of +0.56 versus +0.03 for non-SOZ pairs within 10 mm (step = +0.50 z, p = 0.0001). Secondary tests found that spike rate decreased with distance (mean rho = −0.162, 22/30 patients negative, p = 0.0012), more for the patient's own SOZ than for pseudo-SOZs (p = 0.0009); the gradient was also present on shafts without SOZ contacts (post hoc p = 0.0097). The spike-rate result was outside the Holm-corrected primary family, although p = 0.0012 remained below 0.05 after Bonferroni correction over 10 secondary tests. Clinical labels may depend on the same interictal features, so the power step may be partly circular. Other limits include template rather than native coordinates, pair midpoints, no white/grey matter control, sparse near-SOZ sampling, one 120-s clip per patient, and upstream rather than recomputed features. Archived records identify the source as OpenNeuro ds004100 but do not establish the exact release analyzed.
stereo-EEG · epilepsy · seizure onset zone · interictal activity · broadband power · spike rate · spatial gradient · HUP · OpenNeuro ds004100
References
3- usesA HUP-trained interictal SEEG classifier ranked SOZ channels across four centers but did not outperform spike rate sci:sha256:d9a3c8bd125407cde826c17edcde69ee63194ae20c775341d02234ef31adf9f1
Per-pair interictal features and clinical SOZ labels of 32 HUP SEEG patients (inputs/hup_channel_features.tsv, 120-s clips of ds004100) are the input of all tests.
- usesAperiodic-knee timescales did not increase anteriorly in 36 epilepsy patients; post hoc, inferior pairs had longer timescales sci:sha256:45dee25007d7bcc9aa07b818671117a326eff39f45727f2afe8494a520373aaa
Archived ds004100 fsaverage electrode sidecars gave the contact coordinates used to calculate pair-to-SOZ distances and the pseudo-SOZ control.
- supportsIn 43 SEEG patients, sleep HFA did not decline with contact-step distance from the clinical SOZ sci:sha256:f957aa6334baf6ab772fd787b16e25ac4c615e789717abb504160d39524ba71a
That work found no shaft-step gradient of 80-300 Hz power around the SOZ in NREM sleep (ds007118/9/20). We found no 3D-distance gradient of 1-40 Hz power around the SOZ in interictal HUP data (ds004100), with a pseudo-SOZ control.
Cited by
2 publicationsPublic catalog entries on Sciscape.
- usesIn 31 HUP patients, distance correction did not change SOZ power AUC; a power–distance association was not detected
Source of the archived HUP feature table and fsaverage electrode files.
- usesA pre-registered ds003876 test did not confirm an along-shaft spike-rate gradient from the clinical SOZ
Its secondary spike-rate gradient (rho -0.16, HUP, 3D distance) gave the hypothesis that we tested. Our pre-registered test in ds003876 gave a similar direction and size (rho -0.19), but it did not meet the Holm threshold, and a contact-depth adjustment removed most of the effect.