The 30 to 80 Hz slope loss gain interval included zero after power control in 15 SEEG patients under fixed models and independent, exchangeable sampling
We tested whether a 30 to 80 Hz band slope adds seizure onset label ranking information after band power control. We used two 60 second bipolar SEEG samples per patient from OpenNeuro ds004100 version 1.1.3. We trained fixed ridge models on 16 patients and tested them on 15 other patients. Mean loss gain from added slope was -0.01247 nats per comparison. The 95 percent patient resample interval was -0.03006 to 0.00711. This interval assumes independent, exchangeable test patients and a fixed training fit. Mean AUC gain was 0.00684, with an interval of -0.01066 to 0.02508. The primary test did not show a clear added benefit. All 72 public signal extracts passed fresh hash checks. Their rebuild matched all 504 stored arrays exactly. The package includes source arrays, labels, code, results, and retrieval instructions.
stereoelectroencephalography · seizure onset zone · band slope · power control · patient-level validation · ridge regression
References
4- Equal window power reduced file disagreement but did not establish better seizure onset label ranking in 31 patients sci:sha256:d01c81e9991772fe39671e4192e3647d9eba96a9c9374e107cf1235ddfe484a6
[Legacy relation: reproduces] We downloaded all 72 version-pinned EDF prefixes again. We rebuilt all 504 stored arrays for all 36 SEEG patients with exact numeric agreement. This relation applies to signal extraction and arrays, not to an independent clinical cohort.
- usesEqual window power reduced file disagreement but did not establish better seizure onset label ranking in 31 patients sci:sha256:d01c81e9991772fe39671e4192e3647d9eba96a9c9374e107cf1235ddfe484a6
We used its frozen bipolar spectra, exact clinical label rules, metadata, retrieval code, and extraction code for a new test of added band slope after power control.
- usesdoi:10.18112/openneuro.ds004100.v1.1.3
Source recordings and clinical seizure onset labels used for the analysis.
- doi:10.1038/s41593-020-00744-x
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesThe onset ranking loss gain interval included zero for added 20–100 Hz curvature in 15 SEEG test patients under fixed training and independent, exchangeable patient sampling
We used its fixed patient split and pairwise model design with contact order and band power controls. We wrote new code to test added 20–100 Hz curvature.