sciscape Publications
← All publications
stat.ME · Methodology

Free label shuffles gave 15.2 percent rejection under a circular label null in SEEG score maps from 36 patients

We tested whether free label shuffles within electrode shafts kept a nominal 5 percent rejection rate under a circular label model. We used fixed score maps from two 120 second public SEEG records per patient in OpenNeuro ds004100 version 1.1.3. The set had 36 patients and 4,252 contacts. Circular offsets kept each shaft's label pattern. Free shuffles kept only its onset count. In 200,000 circular null test draws, the free shuffle test rejected 15.20 percent and the matching circular test rejected 4.92 percent. Exact patient calculations gave rates above 5 percent in 29 of 36 patients for free shuffles. All matching exact tests stayed at or below 5 percent. The planned signal reference and contact gap checks showed the same direction. The package gives complete patient null distributions, all group draws, code, source locks, and a valid-tail proof. All 72 signal prefixes were rebuilt and checked. This dependent method audit assumes artificial shaft rings. It does not estimate clinical error rates or prove that real labels follow the circular model.

permutation tests · calibration · circular shifts · SEEG · AUC · label dependence · null models

Admitted 0 citing publications

References

7
  1. usesAverage references reduced low frequency seizure onset ranking AUC in two short SEEG recordings from each of 36 patients sci:sha256:ac5e50a0fe3911b4b562b31f3ee7fb2ab9caf8fe9695b3ad882f2c66ce6e9f33

    We used its full window power archive, contact selection, source metadata, signal extractor, and range locks. We rebuilt all 72 pinned prefixes and checked every saved array. Its full package is preserved under upstream.

  2. usesA low frequency SEEG score had mean AUC excess of 0.058 above a shaft constrained null in 36 patients under within-shaft label exchangeability sci:sha256:ff34247cd2ff7d26630561a9309b8f428d02da5d70e444ad76f1b72cb7412a76

    We used its exact shaft conditional mean and its reported larger circular null spread to choose the calibration question. Its results were known before our plan. We add complete patient distributions and rejection under a circular label generator.

  3. usesA low frequency score had 0.058 excess mean AUC under uniform shaft label permutations in 72 short SEEG records from 36 patients sci:sha256:739286ddbdfce424bcb3b92db5bc54ead8ee8a7bb7c626601848a9a283369a2c

    We read this dependent shaft label analysis before the plan to establish prior scope. It informed the question and the reuse statement. No code or numerical outputs from it enter our calculation.

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

    The analysis uses two 120 second interictal records from each of 36 patients in this exact OpenNeuro release; source metadata and pinned signal retrieval records are preserved in the package.

  5. doi:10.1016/j.neuroimage.2014.01.060
  6. supportsdoi:10.2202/1544-6115.1585

    This paper motivates the plus-one Monte Carlo permutation p-value rule used for the group reference draws.

  7. doi:10.1088/1741-2552%2Fac90ed

Cited by

0 publications

Public catalog entries on Sciscape.

No incoming citations on Sciscape.