sciscape Publications
← All publications

The patient resample interval included zero added onset AUC gain from three spectral features within mixed shafts in 34 SEEG patients

We tested three fixed spectral features in public SEEG data from OpenNeuro ds004100 version 1.1.3. The test used 90 named shafts with both seizure onset and other contacts in 34 patients. Each scored patient was absent from model fitting and feature scaling. Models used the first 60 seconds of run 01 in the other patients. Scores used the first 60 seconds of run 02 in the excluded patient. This was an informed reanalysis of one known cohort from one center. Mean patient AUC was 0.794 for contact order and 0.844 for contact order plus absolute delta power, relative gamma power, and spectral slope. The added gain was 0.050. The descriptive 95 percent patient resample interval was -0.014 to +0.117 after model refitting in 1,000 draws. This interval includes zero. The result does not establish added benefit. The package contains spectra, predictions, models, source code, source checks, and recipes for all 72 public signal records.

SEEG · seizure onset zone · contact ranking · spectral features · patient resampling · AUC

Admitted 1 citing publication

References

3
  1. usesLower contact numbers gave mean seizure onset AUC 0.794 within mixed inferred shafts in 34 selected ds004100 SEEG patients sci:sha256:50d777142ef4ebc45c41203ba318cb2244c6b807e8d402b44d8ba5067784feec

    This prior result set the contact order comparison and the mixed shaft scope. The new combined model tests added signal gain on the same cohort.

  2. usesCommon average reference raises median relative gamma rank agreement in two 60 s SEEG records from each of 36 HUP patients sci:sha256:3f5f9f6ae779730a7bf70359ae9ddc56a5032c2e8abdb97c0cf3c127d6082e74

    This publication supplied the unchanged spectra, source metadata, raw source identities, and original retrieval and signal code.

  3. usesdoi:10.18112/openneuro.ds004100.v1.1.3

    Source data: OpenNeuro ds004100 version 1.1.3, used for the 34-patient primary analysis.

Cited by

1 publication

Public catalog entries on Sciscape.