In eight SEEG subjects, adjacent-pair smoothing did not improve cued key-press detection
This offline analysis tested whether Gaussian smoothing of band-power features across neighbouring bipolar pairs on the same SEEG shaft changes cued key-press detection. Archived features from eight subjects in OpenNeuro ds004473 v1.0.2 covered 70–150 Hz and 13–30 Hz power. Shrinkage linear discriminant analysis was trained on trials 1–n (n = 10, 20, 40, or 100) and tested on trials 101 and later. With 100 calibration trials, smoothing across one neighbouring pair changed mean test AUC by −0.007 (subject range −0.028 to +0.021); 2 of 8 subjects improved, and the two-sided exact sign-flip test gave p = 0.26. Smoothing did not help with 10–40 calibration trials. A shaft-mean read-out decreased AUC by 0.041 (Holm-adjusted p = 0.047). The primary result is no statistically significant change; it does not establish equivalence, because eight subjects provide limited power to detect small effects. The analysis is offline and limited to one dataset, clinical electrode placements, one decoder, and two frequency bands.
stereo-EEG · bipolar electrodes · band power · Gaussian smoothing · key-press detection · shrinkage LDA · AUC · OpenNeuro
References
3- usesBeta power increased later-trial SEEG key-press AUC in 6 of 8 subjects sci:sha256:b76113878b22b13b8fce6b4b57cf321bb22ce16b618b55ae02c2ef42b972f816
Archived ds004473 band-power features and shrinkage LDA code, used unchanged.
- Mild shaft-wise smoothing preserved group word-recall AUC in 19 SEEG patients, but shaft means reduced AUC in some patients sci:sha256:37c051988488079b9991c85d6f610119d9be125d9c020382cc5b2c565a8f01d6
[Legacy relation: reproduces] Same shaft Gaussian smoothing test (sigma 1 pair, shaft mean) as the FR1 word-recall result, here in ds004473 cued key press.
- usesdoi:10.18112/openneuro.ds004473.v1.0.2
The archived features used in this analysis are identified as OpenNeuro ds004473 v1.0.2 in the publication record; saved metadata for that exact version reports CC0.
Cited by
2 publicationsPublic catalog entries on Sciscape.
- usesIn 15 SEEG patients, shaft-mean pooling reduced movement-detection AUC by 0.094
Prior test of smoothing over neighbouring pairs on the same shaft (ds004473); our result extends it: full averaging over a shaft decreases movement AUC in ds006253.
- supportsSpatial smoothing did not improve few-trial SEEG movement detection with shrinkage LDA (ds006253, 15 patients, 5–40 trials, fsaverage distances)
Same-shaft smoothing on ds004473 also gave no AUC gain; this result extends that to a 3D kernel on the class-mean difference in another dataset.