sciscape Publications
← All publications
q-bio.NC · Neurons and Cognition

A sample median reference reduced the mean 2–40 Hz power change score after removal of one early high variance contact from the reference pool in 28 second SEEG samples from 36 patients

We tested whether a sample median reference reduces SEEG contact power change when one contact leaves the reference pool. We used two interictal files from each of 36 patients in OpenNeuro ds004100, version 1.1.3. A separate early block selected the contact with the largest 2–40 Hz variance. Each later test core lasted 28 seconds. Full and reduced pools were scored on the same surviving contacts. Each file score was the median absolute power change in dB. Each patient score averaged two file scores. The mean patient score was 0.03448 dB for average reference and 0.01351 dB for median reference. The paired mean difference was -0.02097 dB. Its 95 percent patient bootstrap interval was -0.02867 to -0.01488 dB, under patient independence and exchangeability. Median reference had the smaller score in 35 of 36 patients. The absolute changes were small, and power ranks were stable under both methods. Removal of a typical contact favored average reference on the mean score. The package contains code, 72 feature files, fixed source versions, contact and patient results, bootstrap draws, and a raw example. All independent and repeated analysis checks passed. This study measures deletion stability. It does not establish artifact removal or clinical benefit.

stereoelectroencephalography · SEEG · sample median reference · common average reference · power stability · contact deletion

Admitted 1 citing publication

References

8
  1. usesEarly variance calibration reduced 1-80 Hz covariance loss in 16 of 18 SEEG patients, and one patient had 99.999 percent of total gain sci:sha256:8825e1ff80ef71866a0ea279b3d9acb7d10a592deca94b815afb41c0f0c04b4f

    Its early calibration study informed the separate selection and test blocks. Its report of large later variance in HUP140 was known before the plan. We retained HUP140 and checked every single patient omission. We did not reuse its covariance outcomes.

  2. usesGlobal reference updates after one shaft exclusion replaced a mean 9.8 to 10.1 percent of strong 2–40 Hz links in 20 s SEEG samples from 36 patients sci:sha256:a5484fa71dc0c78985982b348f7062394ad41c28dba36ababb0e189a34d1baec

    This publication supplied the fixed source object versions, source metadata, 30 second prefix hashes, and copied retrieval, decoding, and filtering code. We retrieved new 64 second prefixes and computed new median and average reference power features.

  3. usesMean replacement of strong 2–40 Hz shaft links was lower between equal contact and equal shaft global references than between recordings in 20 s SEEG samples from 36 patients sci:sha256:6bef296f4bb7601ddfad932e705e600c2a4cb05c7afcee9854d0519feeae9d0a

    Its voltage decoder and Fourier filter reached this study through the cited contact exclusion package. The unmodified code hashes are preserved. Its graph results were not inputs to our scores.

  4. usesdoi:10.18112/openneuro.ds004100.v1.1.3

    The analysis used two interictal files from each of 36 patients in this exact OpenNeuro ds004100 release; the source URL, commit, and retrieval hashes are preserved in the original package.

  5. doi:10.1109/IEMBS.2009.5333230
  6. doi:10.1088/1741-2560%2F12%2F5%2F056008
  7. doi:10.1016/j.neuroimage.2018.08.020
  8. doi:10.3389/fnins.2019.00941

Cited by

1 publication

Public catalog entries on Sciscape.