A contact metric preserved regularized beta interaction scores across two bases of fixed SEEG path spaces in eight patients
We tested regularized beta interaction scores in public SEEG data from eight patients. The sample had 512 trials and 148 contact paths. Adjacent and Helmert contrasts retained the same space within each path. At ridge weight 0.1, a contact metric kept scores equal to numerical precision. The largest absolute basis difference was 6.33e-15 across 23,058 frequency scores. An output identity penalty replaced 19.0 percent of the strongest baseline edges on average. The contact metric replaced none. A proof states the required fixed node spaces and real invertible transforms. Trial halves had weak agreement for response-minus-baseline edge changes. The package includes code, inputs, source versions, raw fixtures, results, figures, and retrieval instructions. Fresh public retrieval matched all 1,134,720 complex Fourier coefficients exactly. This is a method benchmark on reused trials. It does not establish neural source recovery or clinical use.
SEEG · multivariate interaction measure · regularization · basis invariance · contact metric · signal processing
References
4- usesIn 36 patients, fixed 1–80 Hz SEEG contrasts showed larger covariance shape changes between files than within files sci:sha256:daf70bbdcc88f893fea25fa9870f5621a6fd8289ee4133a9ec263f83268cd815
We used its contact metric penalty as the coordinate model. We applied the rule to a ridge extension of trace MIM.
- usesBeta covariance determinant ratios agreed across two bases of the same SEEG contrast space in eight patients sci:sha256:1516f34ddd74af81909da7c0aa225fd986f7045e92cb54460d591827e76816e2
We used its Fourier inputs, contrast bases, fixed trial sample, metadata, raw fixtures, and retrieval code. We rebuilt all windows from pinned public EDF records.
- usesdoi:10.18112/openneuro.ds004473.v1.0.2
Source data and task metadata for the secondary analysis; the exact recorded dataset snapshot is version 1.0.2.
- doi:10.1016/j.neuroimage.2011.11.084
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesTrial self-product correction reduced mean calibrated beta interaction scores at keypress by 95 percent in eight SEEG patients
This work supplied the fixed SEEG Fourier inputs, contact paths, metadata, source locks, EDF fixtures, and retrieval code. We rebuilt the inputs from fresh public EDF ranges. We then used separate scale calibration and removed trial self-products. These new scores have a different scale target from the upstream scores.