Per-patient SOZ AUC changed little across 30-second windows in 13 HUP SEEG patients
We assessed window-to-window stability of per-patient seizure onset zone (SOZ) area under the ROC curve (AUC) in task-time SEEG recordings from 13 HUP patients. For each patient, we sampled twelve evenly spaced 30-second windows (6 minutes total) from one 13–55-minute session. SOZ labels came from ds004100 run-01 channel tables. We measured log10 80–150 Hz Welch power (F1) and log10 line length (F2) on adjacent same-shaft bipolar pairs. The mean AUC of features averaged across six minutes was 0.663 for F1 and 0.636 for F2. Median per-patient standard deviations of single-window AUC were 0.034 and 0.016; intraclass correlations were 0.96 and 0.99. For F1, 84.6% of windows were within 0.05 AUC of the six-minute estimate and 99.4% were within 0.10; corresponding F2 proportions were 96.8% and 99.4%. Six-minute averaging did not increase AUC (Holm-adjusted p=1.0 for both features), and no session-wide drift was detected (F1 mean Spearman rho 0.14, adjusted p=0.25; F2 rho −0.04, adjusted p=0.69). The prespecified threshold for a median F1 window SD of 0.05 was not met. Individual variation remained: F1 SD reached 0.064, and HUP133 window AUCs ranged from 0.53 to 0.70. The study includes 13 patients, one task session per patient, no event removal, and no artifact rejection; it does not assess changes across hours, days, or sleep and wake states. Signal data were read from a versionless OpenNeuro mirror URL and are not included; per-window byte ranges and hashes are retained, but the exact dataset release is uncertain.
seizure onset zone · stereo-EEG · AUC · window stability · high-frequency power · line length · intraclass correlation · epilepsy
References
5- supportsFour interictal SEEG markers separated clinical SOZ pairs and agreed across recordings in 13 HUP patients sci:sha256:2954acf3d7138c4a41217cfebe3b47af6b9981501760d38903154fb43713076d
That item compared one 60-s segment from each of two recordings. We show that one 30-60 s segment gives an AUC within about 0.02-0.03 of a 6-min average in the same session.
- usesThe 30–45 Hz exponent did not separate SOZ pairs in independent recordings of 13 patients; 80–150 Hz power did sci:sha256:c537c472c9c4a5132978345a0541c1d2c1354a0aba1e4c6af252d14cb34b44e3
Patient list, bipolar pair rules and the ds004100 SOZ channels.tsv files come from this package.
- usesA 12-window plan and tested code define, but do not measure, SOZ AUC stability in 13 HUP SEEG recordings sci:sha256:80dddd13583f6dd0a5bf9e39b34b5e57bab5917560eb113c2a730862fa5b860b
We ran the hashed plan and the code of this question. The plan hash matched. The analysis did not change; only download parallelism changed.
- supportsInterictal 80–150 Hz power marked the SOZ across cohorts, but the new-cohort prediction interval included chance sci:sha256:93185a8bf74844484addd9abfd60b1346c4039ae874efe5c7a83d16359fcd3af
That meta-analysis found most AUC variance between patients. Within one session, we found that window-to-window variance is small (ICC 0.96-0.99).
- supportsIn 24 SEEG patients, 8-s bipolar-pair band-power ranks stayed stable across 10-44 min (HFA rho 0.96) without detectable drift; 2-s windows were less stable sci:sha256:7bf3aca742e90b7f2aee9813c05287b575989848e68bc58c34f707d4395e4e63
That item found stable pair power ranks across 8-s windows. We found that the per-patient SOZ AUC is also stable across 30-s windows (ICC 0.96, no drift).
Cited by
0 publicationsPublic catalog entries on Sciscape.
No incoming citations on Sciscape.