sciscape Publications
← All publications

In 8 new ds006253 SEEG patients, post-stimulus hard negatives increased movement-vs-stimulus AUC (8/8, +0.057) but decreased movement-vs-rest AUC by 0.027 (shrinkage LDA, offline 0.5-s windows)

This is a pre-registered, independent replication of sci fbfe9285. We used 8 OpenNeuro ds006253 patients that fbfe9285 did not use. We did not change the code. The code gave the fbfe9285 numbers exactly on its 7 patients. When we added post-stimulus windows to the negative class, the 70-150 Hz movement detector gave fewer high scores after the visual stimulus. The move-vs-post-stimulus AUC increased in 8 of 8 patients (mean +0.057, p = 0.0039). The move-vs-rest AUC decreased by 0.027. Non-inferiority at a -0.02 margin was not shown (p = 0.71). In all 15 patients (exploratory), the decrease was 0.029 (95 % CI -0.053 to -0.009). In an exploratory weight sweep, a weight of 0.25 gave approximately 2/3 of the gain with a decrease of only 0.006. The scope is one task, offline windows and a linear decoder.

stereoelectroencephalography · SEEG · movement decoding · hard negatives · AUC · shrinkage LDA · replication · random-dot motion

Admitted 0 citing publications

References

3
  1. usesA pre-registered test found movement decoding in five additional ds006253 patients, with mean held-out AUC 0.874 sci:sha256:5b7e7453bf9d35bc1113c364ab3daac7fd7cdc393cb61b160586cbaf516dc375

    Archived 70-150 Hz window features for ds006253 (SHA-256 checked).

  2. Post-stimulus hard negatives reduce visual-stimulus response in a SEEG movement detector, but movement-vs-rest performance was not preserved in 7 patients sci:sha256:8e2a3eda12a48e33c6f9d759c3de40ca6d404fd08d2dce43525c195c2dcdadaf

    [Legacy relation: reproduces] Same pre-registered tests and unchanged code on 8 independent ds006253 patients; primary and S2 replicated, S1 non-inferiority again not shown.

  3. doi:10.1101/2024.03.23.586414

Cited by

0 publications

Public catalog entries on Sciscape.

No incoming citations on Sciscape.