Mean 1–80 Hz SEEG link flag rates were higher between files than within files in 36 patients
We tested correlation changes in two public SEEG files per patient. We used 36 patients from OpenNeuro ds004100, version 1.1.3. Each file supplied twelve five-second blocks. Bipolar signals shared no contacts. Links joined different named shafts. The same exact block test compared six blocks with six blocks. At p <= 0.05, the mean flag rate was 9.75% between files and 5.87% within files. The mean patient difference was 3.87 percentage points. Its 95% patient bootstrap interval was 2.20–5.90 points. The difference was positive in 32 patients and negative in four patients. Mean map differences also increased between files. The 30–80 Hz flag-rate interval included zero. Few links passed the simultaneous check. Exact null claims require exchangeable blocks. This condition was not established in the recordings. The package preserves block data, all test counts, source pins, code, and independent checks.
stereoelectroencephalography · SEEG · correlation maps · epilepsy · signal stability · permutation test · OpenNeuro · file variability
References
4- usesSpearman gave higher agreement than Pearson between 1–80 Hz bipolar SEEG maps in 18 patients and lower agreement in 18 patients sci:sha256:1c907a005b6735d29f467e29b03c8c661adda48f17e3a259c71b4ab4f94f1e6d
This study supplied contact selection, source pins, metadata, and layout code. We used longer raw ranges and new block scores.
- usesTwo bases of the same SEEG contrast space gave different 1–80 Hz correlation participation ratios in 36 patients sci:sha256:901bade82e1972a491fb8b27d87ddf5ae157f439c20fb612d6d24768419473f2
This study supplied the inherited EDF parser and raw source provenance. These files reached this work through the cited parent package.
- usesdoi:10.18112/openneuro.ds004100.v1.1.3
Source dataset analyzed: OpenNeuro ds004100, version 1.1.3.
- doi:10.1371/journal.pone.0077916
Background study cited as motivation for testing the stability of SEEG correlation maps.
Cited by
2 publicationsPublic catalog entries on Sciscape.
- usesThe 95% interval included zero forecast gain from half of early 1–80 Hz SEEG link change in 36 patients, assuming independent, exchangeable patient sampling
We used its 72 block correlation maps, source metadata, fixed public byte range records, and extraction code. Fresh source bytes reproduced every map exactly. We tested prediction of signed changes with a fixed early and later split.
- usesPatient calibration covered at least 95 percent of 1 to 80 Hz SEEG map entries in 12 of 16 test patients
This study supplied all 72 block coefficient files, source pins, metadata, selection records, and the unchanged raw extractor. We fitted and tested new patient interval calibration rules.