Contact choice changes more top 5 percent ordinary partial correlation links after fixed CAR in 36 SEEG patients at 1–80 Hz
We tested whether the choice of one removed contact changes conditional SEEG links after fixed common average reference (CAR). We used two public interictal files from each of 36 patients. CAR was formed before contact removal. We ranked ordinary partial correlations between different named shafts, using identical candidate links for each comparison. At 1–80 Hz, changing the removed contact replaced 12.94 percent of the strongest 5 percent of links after CAR. The acquisition-reference value was 7.33 percent. The mean patient excess was 5.61 percentage points, with a 95 percent bootstrap interval of 4.74 to 6.60. All 36 patient differences were positive. CAR file change caused more mean replacement than removal choice. We preserve covariance files, code, all case results, matrix identities, and pinned raw retrieval instructions. An independent implementation checked link rankings. Eight raw EDF files reproduced source covariances exactly. This method audit does not test clinical accuracy.
stereoelectroencephalography · common average reference · partial correlation · covariance · contact removal · network sensitivity
References
5- usesRidge partial correlation tends to minus one after common average reference with one covariance null mode in 36 public SEEG patients sci:sha256:6d1ad9052276490dd1e37896e7af8b6cf5858bcd5c300bb389769d788719bfeb
The prior CAR constraint result and its single removed-contact control motivated this test across many removal choices. The present result adds a matched acquisition-reference baseline and file-change comparison.
- usesTop 5% 1–80 Hz partial correlation links between retained shafts had lower mean replacement after one shaft omission than file change in 35 SEEG patients sci:sha256:7f1ad56600ec1f1edf0ef02d02a80966bb26fc88b203dbc3e15d366cefcee29f
This study adapts the matched candidate-link comparison of omission with file change and the equal patient weights. The new test uses individual contact choice in two reference systems.
- usesEdges that share a contact are more common among the strongest 5 percent of 1–80 Hz bipolar SEEG correlations in 36 epilepsy patients sci:sha256:7dc3370736cdefb57323429971522e882b644b700faccfda51b916b3d865f58f
This study uses all 72 lossless contact covariance files, metadata, source hashes, and pinned EDF retrieval code. It calculates new contact removal choice comparisons in two reference systems.
- usesdoi:10.18112/openneuro.ds004100.v1.1.3
The analysis uses two interictal files from each of 36 patients in HUP iEEG Epilepsy Dataset version 1.1.3.
- doi:10.1093/braincomms%2Ffcae165
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesContrast scores gave more file replacement than fixed CAR partial correlations at 1–80 Hz and 5% in 36 SEEG patients
This publication supplies the 72 lossless covariance files, metadata, selected contact cases, and source checking scripts. The ordinary CAR and acquisition reference results are reproduced exactly. The new analysis compares contrast file replacement with those results.