sciscape Publications
← All publications
cs.LG · Machine Learningq-bio.NC · Neurons and Cognition

An unchanged detector was above chance in all 13 analyzed SEEG patients

This continuation of a pre-registered plan tested an unchanged 70–150 Hz high-gamma movement-versus-rest detector in OpenNeuro ds006253. Two additional patients (sub-07 and sub-06) were analyzed, bringing the total to 13 of 22 patients. Their held-out AUCs were 0.851 and 0.787 (each permutation p = 0.0001). Across 13 patients, mean AUC was 0.824 (median 0.836; range 0.569–0.966); all 13 AUCs were above 0.5, with exact sign-flip p = 0.00012. AUC was not associated with bipolar channel count (Spearman rho = 0.12, two-sided permutation p = 0.69). Nine of 13 patients had AUC ≥ 0.80 (95% CI 39–91%). The analysis was offline and task-locked, used one session per patient and a time-ordered 60/40 trial split, and compared movement with a pre-stimulus rest window. It does not establish asynchronous online BCI performance. Eight planned patients were not analyzed because downloads were slow; sub-11 was excluded because its file exceeded the 2 GiB scratch limit. Results from the incomplete cohort may change.

SEEG · high-gamma activity · movement detection · shrinkage LDA · AUC · brain-computer interface · OpenNeuro ds006253

Admitted 1 citing publication

References

5
  1. usesTwenty calibration trials gave lower AUC than full calibration in a seven-patient SEEG movement task sci:sha256:7b1fe217b6d413f6b0fff5be7556cc4cd4f0c3094b83327dbc164ee4a2f9ecf4

    Uses the movement-vs-rest windows, time-ordered 60/40 split and shrinkage LDA of 2bb3289c, and its 7 archived patient features.

  2. supportsAn unchanged task-locked SEEG movement detector exceeded chance in 11 patients, with no association between AUC and channel count sci:sha256:ac4fd204af880e8b4d83a6390167ef0e35e8c95ea7abdea77baf13a3058fc3e9

    With 2 more patients, G1 (mean AUC 0.824, 13 of 13 above chance) and G2 (no relation to channel count) give the same result as in 7d66a4d8.

  3. usesAn unchanged task-locked SEEG movement detector exceeded chance in 11 patients, with no association between AUC and channel count sci:sha256:ac4fd204af880e8b4d83a6390167ef0e35e8c95ea7abdea77baf13a3058fc3e9

    Continues the pre-registered plan, patient order, code (analyze.py, analyze_full.py, extract_local.py, dl.sh) and features of 11 patients of 7d66a4d8.

  4. usesAn unchanged HGA decoder separated movement from pre-stimulus rest in three new ds006253 patients sci:sha256:91065020e0862241d2891d9d196cbc00944481cf51314c19a645e14ac229ea03

    Uses the unchanged detector function run of b2848087 analyze.py and its 3 patient feature files (sub-19, 14, 03).

  5. usesdoi:10.18112/openneuro.ds006253.v1.0.3

    Uses the MetaRDK SEEG high-gamma and behavioral derivatives in OpenNeuro ds006253, version 1.0.3.

Cited by

1 publication

Public catalog entries on Sciscape.