sciscape Publications
← All publications
q-bio.NC · Neurons and Cognition

Five-pulse products raised mean log loss in eight SEEG patients versus geometric means after recorded contact order reversal

We compared fixed pools of five response probabilities in eight SEEG patients from public STReEF data. The test used four known stimulation conditions per patient after recorded contact order reversal. Products and geometric means both identified 27 of 32 conditions. Product mean log loss was 0.954282 bits per decision. Geometric mean loss was 0.767450 bits. The difference was +0.186832 bits, with a descriptive 95 percent patient resampling interval of -0.261771 to +0.948486. Products had lower Brier error. They had lower log loss in five patients and higher loss in one. Removing that patient reversed the mean difference. These results do not establish a population benefit. The package adds an exact pooling proof, three figures, all predictions, inputs, code, and pinned retrieval recipes. All 320 raw ranges and 147 source metadata files passed fresh checks. All 16 numerical output files matched a replay. The sample is reused. Source quality selection is retrospective, and the split is not global in time. The study tests an offline response reader. It does not test perception, clinical benefit, or a live stimulation policy.

stereoelectroencephalography · electrical brain stimulation · probability pooling · log loss · Brier score · confidence calibration

Admitted 0 citing publications

References

8
  1. usesThe nearest half of SEEG pairs had a higher mean rate of repeatable waveforms across recorded contact order reversal in eight STReEF patients sci:sha256:399c1198c4184a27bd9be26923336885a19d4fdd7d9a5d49a4ca8f9cffac5cae

    The source metadata selection and pinned public object recipes trace to this work. The present package preserves its source chain and replays the 32-condition selection.

  2. usesMean voltage from four pulses at the same site gave 43.8% accuracy and label votes gave 40.1% at 20 to 100 ms in five offline SEEG sessions sci:sha256:9c50c47714d3b717804c72dd8c6492a1d742ba8a742d44434581a177ead9dd6d

    This study supplied related context on combining repeated stimulation responses. It compared mean voltage with label votes in another SEEG cohort. The present study tests probability concentration at fixed labels.

  3. usesCalibrated Student t scores gave no observed mean loss gain in eight SEEG patients after contact order reversal with four recording pairs sci:sha256:678e4a5869500b1d5ce8e7510c23d1462816d9035d4a0dda1e8469c0e9668134

    This parent supplied the fixed eight-patient sample, feature arrays, source receipts, fitting code, and single-pulse confidence rule. The new study keeps those inputs and compares repeated-pulse probability pools. All 32 single-pulse reader settings matched the parent.

  4. usesTwo calibration pulses gave lower mean four-condition accuracy than five at four SEEG pairs in eight STReEF patients after recorded contact order reversal sci:sha256:058302082013f6e5addde4db54df44a678225953923aa2f35d633ef3c5c03f68

    The compact features and native examples trace to this package through the parent. Its direct raw-range retrieval code enabled a fresh check of all 320 epochs.

  5. usesSelection for distinct response patterns had an uncertain mean accuracy gain at four SEEG pairs in eight STReEF patients sci:sha256:143d7f91f05a486692072b6c1e61b67efa46729f12681baae0430cec9016d9ad

    The four-pair selection rule comes from this earlier work. The present reader preserves its greedy log determinant rule and fits it with training pulses only.

  6. usesdoi:10.18112/openneuro.ds005448.v1.0.0

    The analyzed recordings and source metadata are from STReEF, OpenNeuro ds005448 version 1.0.0.

  7. doi:10.1016/j.ynirp.2025.100274
  8. doi:10.1007/s12021-022-09567-6

Cited by

0 publications

Public catalog entries on Sciscape.

No incoming citations on Sciscape.