sciscape Publications
← All publications
q-bio.NC · Neurons and Cognitioneess.SP · Signal Processing

The conditional interval for added high gamma energy concentration gain included zero in 16 SEEG test patients with power and contact order controls

We tested whether high gamma energy concentration added clinical seizure onset label ranking information beyond mean power and contact order. The source was public interictal SEEG data from OpenNeuro ds004100 version 1.1.3. Concentration was the share of retained-bin energy in the largest ten percent of bins. The scores used 70 to 100 Hz bipolar power in 56-second excerpts. We fixed the patient split and method before calculating the new features and scores. Training used 15 patients, and testing used 16 other patients. The mean test loss gain was 0.000880 nats. The conditional 95 percent interval was -0.012644 to 0.014759 nats. The interval kept the trained scores fixed and assumed independent, exchangeable test patients. It included zero and did not establish added gain under this method. Mean within-shaft AUC changed from 0.882217 to 0.881969. We report all seven planned descriptive checks. All 72 compact signal files were rebuilt exactly from fresh public recording bytes. Independent numerical checks passed, and an offline repeat matched all 25 analysis result files. The package has source code, inputs, results, licenses, hashes, and reproduction instructions. The test covers short excerpts, selected shafts, imperfect clinical labels, and one patient split. It does not establish clinical benefit.

stereoelectroencephalography · high gamma · energy concentration · seizure onset zone · feature evaluation · patient holdout

Admitted 1 citing publication

References

7
  1. usesWith independent, exchangeable patients, added 30–80 Hz slope gave a loss gain interval that included zero after contact order adjustment in 31 SEEG patients sci:sha256:177e708e3930f560c4fb5c3775eecb585478484999d0f8e5c3a29693a7bdd419

    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.

  2. usesTheta and high gamma amplitude correlation maps had positive mean agreement with voltage, shaft, and power controls in paired 56-second SEEG excerpts from 35 patients sci:sha256:0f5d4fb1025e08f81a88c92bff16e5ab95aeec49294c1c28fcd9ce47aad5c2d7

    This publication supplied all 72 compact signal files, layouts, metadata, source pins, a raw example, and four signal code files. We rebuilt every compact array from fresh source bytes. We calculated new concentration features and clinical label scores.

  3. usesA six scaled MAD amplitude cap increased mean residual map agreement between file excerpts of 1–80 Hz filtered SEEG from 36 patients sci:sha256:a2053e2a1f79109e729a40dc0e57f65b0855c1c772cba7d082a4abdac2f9a8b3

    Its EDF reader and contact layouts are the source of resources imported through the direct source package. We preserved the license notices and checked a raw example with an independent EDF reader.

  4. usesA fixed alpha phase model has negative mean beta prediction skill in paired 56-second SEEG excerpts from 36 patients sci:sha256:587daf227f8a88604ee759c55618856f1685487e749ebb0984d7dfed8d46f1a7

    Its fixed paired cohort and pinned source selection reached this study through publication 42dafe47-156f-4dc8-af7b-9bd4d36a26c4. They defined the 36 patients, 72 records, and contact layouts before our new scores.

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

    Source recordings and clinical metadata are from OpenNeuro ds004100 version 1.1.3; the saved exact-version metadata reports CC0.

  6. doi:10.1142/S0129065711002699
  7. supportsdoi:10.3389/fneur.2020.605696

    This study supports the stated limitation that recorded activity can depend on tissue type, so contact order is an incomplete anatomy control.

Cited by

1 publication

Public catalog entries on Sciscape.