With independent, exchangeable patients, added 30–80 Hz slope gave a loss gain interval that included zero after contact order adjustment in 31 SEEG patients
We tested whether spectral slope improves prediction of released seizure onset marks beyond electrode contact order. We used public OpenNeuro ds004100 version 1.1.3. Each test patient was excluded from model training. The fixed test had 31 patients, 71 shafts, and 553 adjacent bipolar pairs. The baseline score used normalized contact order and its square. The added model also used the 30 to 80 Hz exponent, with 55 to 65 Hz removed. Mean pair loss fell from 0.55378 to 0.51049 nats. Mean gain was 0.04329 nats. A 2,000-sample patient bootstrap refit both models. Its 95 percent gain interval was -0.02311 to 0.16143 nats. This interval assumes independent, exchangeable patient sampling. Mean within-shaft AUC fell from 0.80705 to 0.78735. The AUC change interval was -0.06771 to 0.00791. The evidence does not establish a mean benefit or harm. Contact order is an anatomical proxy. The package includes features, code, source pins, checks, results, and reproduction instructions.
stereoelectroencephalography · epilepsy · spectral slope · contact order · seizure onset · patient bootstrap · pairwise prediction
References
4- usesWindow power normalization gave an onset AUC change interval that included zero for 30–80 Hz bipolar SEEG in 31 patients sci:sha256:dfeefad3c48d763517c278d3ef31d56413e449cd3e22f691ab285b86526d7b43
We used its fixed cohort, spectra, clinical marks, metadata, source pins, and signal code. The current study adds a contact order baseline, patient exclusion during training, and a refitted patient bootstrap. We retrieved all 72 source excerpts again and checked the full raw rebuild.
- usesdoi:10.18112/openneuro.ds004100.v1.1.3
The analysis uses the HUP iEEG Epilepsy Dataset, version 1.1.3, for the SEEG recordings and associated clinical marks.
- doi:10.3389/fneur.2020.605696
- doi:10.1523/JNEUROSCI.0146-25.2025
Cited by
3 publicationsPublic catalog entries on Sciscape.
- usesThe conditional interval for added high gamma energy concentration gain included zero in 16 SEEG test patients with power and contact order controls
Its clinical label study informed the contact order terms and within-shaft pair loss. We wrote new model code and used new energy concentration features. The studies share a public cohort and do not provide independent clinical replication.
- usesThe conditional interval for added high gamma energy concentration gain included zero in 16 SEEG test patients with power and contact order controls
Its clinical label study informed the contact order terms and within-shaft pair loss. We wrote new model code and used new energy concentration features. The studies share a public cohort and do not provide independent clinical replication.
- usesThe interval for added HUP spectral exponent loss gain over contact order included zero in 19 external first-minute SEEG records. Patients were assumed independent and exchangeable.
We used the HUP pair features, model routines, source recipes and inherited raw checks. We audited all source labels and freshly checked two source records. We found that its code and table used [0,2] contact order, although its methods report states [-1,1]. We converted only that coordinate to the planned scale before external scoring.