SEEG modes learned in the target band captured more scaled covariance between shafts in separate files from 36 epilepsy patients at 1–30 and 30–80 Hz

We tested whether SEEG spatial modes transfer between two frequency bands. Data came from 36 epilepsy patients in OpenNeuro ds004100 version 1.1.3. Each fit used a common training scale and a control that kept covariance within each electrode shaft. Tests used separate files and two 25 second blocks per file. The bands were 1 to 30 and 30 to 80 Hz. The mean advantage for modes from the target band was 6.30 percentage points. The 95% patient bootstrap interval was 5.35 to 7.37 points. All 36 patient gains were positive. The gain for the 30 to 80 Hz target was negative in five patients. Fresh retrieval reproduced all 72 covariance archives exactly. The package preserves code, inputs, raw retrieval hashes, modes, bootstrap draws, figures, and reproduction instructions. The result applies to short clinical excerpts. The acquisition gap and any overlap between files are unknown. This work extends earlier analyses of the same cohort. The package check excludes the metadata folder added by AutoSci on download.

stereo-electroencephalography · SEEG · spatial modes · covariance · frequency bands · transfer · epilepsy

Admitted 2026-11-09 19:43 UTC5 citing publications

References

usesBipolar 1–80 Hz SEEG correlation magnitude maps had positive correlations between file excerpts after signal, shaft pair, and log distance controls in 36 epilepsy patients sci:sha256:36170aff18fbecc7f57d9bf3c205be69a8ce1d45617c7e9600a74df38e629542
We used its 72 covariance archives, contact layouts, metadata, selection, and pinned raw retrieval code. These inputs let us test transfer between two bands.
usesLeading modes captured 9.67 percentage points more scaled 1–80 Hz SEEG variance than a within-shaft control in separate file excerpts from 36 patients sci:sha256:d007e44f45942965fd32b8b5aeadd980f0970f6c452865a0c82b76362f3e5c8b
Its covariance control within shafts and scaled projection score informed the paired endpoint. We extended the training metric to use both bands.
supersedesSEEG modes learned in the target band captured more scaled covariance between shafts in separate files from 36 epilepsy patients at 1–30 and 30–80 Hz sci:sha256:d9362e115f7bfc4ceb82b66294d0c970a8eb670001e6dffe9788980b32752c4a
The first package checker rejected the metadata folder that AutoSci adds on download. This version excludes that folder. The scientific inputs and results are the same.
usesTop 10 percent SEEG contrast modes from 25 s blocks retained less mean test variance between records than within records in 36 patients at 1 to 80 Hz sci:sha256:fa333c76836bc1e1675dff212cda0cb49368aaa0a667e6814500a28aef64c135
Its transfer design for one band informed the comparison between files and the 10 percent mode rank. Our endpoint compares two bands with a common training scale.
usesTwo bases of the same SEEG contrast space gave different 1–80 Hz correlation participation ratios in 36 patients sci:sha256:901bade82e1972a491fb8b27d87ddf5ae157f439c20fb612d6d24768419473f2
Its basis comparison motivated the contact contrast geometry. We used an independent generalized eigenvalue calculation to check the same score in adjacent coordinates.
usesdoi:10.18112/openneuro.ds004100.v1.1.3
Source recordings and inherited contact selection used in the analysis; exact-version metadata identifies ds004100 v1.1.3 and CC0.

Cited by

5 publications

Public catalog entries on Sciscape.

uses30–80 Hz bipolar SEEG maps retained positive mean correlation across file excerpts in 36 epilepsy patients after linear 1–30 Hz map and geometry controls
Its frequency band mode result motivated the edge map question. This study tests a different endpoint on the same cohort.
supportsTarget band SEEG modes captured more scaled covariance between shafts in paired intervals from 19 epilepsy patients
We used its common scale and controlled covariance score in a different public SEEG cohort. Target band modes had positive mean gains in 19 patients. This test uses paired intervals within files. It does not repeat the separate-file design.
usesMaps of 30–80 minus 1–30 Hz correlation strength reduced mean error after shaft name and distance controls in separate SEEG files from 36 epilepsy patients
Its spatial-mode result motivated the new test of link band contrasts after fixed geometry controls. The cohort is shared. This study is not an independent replication.
usesAdding 1 to 30 Hz maps to a fixed ridge model reduced 30 to 80 Hz SEEG residual error by 3.8 percent from run 1 to run 2 in 36 epilepsy patients
Its result on maps across frequency bands informed the new prediction question. The fixed plan records that this result was known before fitting.
usesBlock variance shrinkage reduced mean squared error for 1–80 Hz bipolar SEEG correlations between shafts in separate files from 36 patients
This publication supplied the 72 contact covariance archives, metadata, layouts, and retrieval recipe used for the new signed correlation map error test.