sciscape Publications
← All publications
eess.SP · Signal Processing

Early linear SEEG corrections increased mean relative squared error by 66.5% over cue and early sensor forecasts during grasps in four people

This study tested whether early SEEG improved forecasts of a later grasp sensor voltage. It used public NEMAR nm000397 version 1.0.0 data from four people with epilepsy. There were 472 eligible events, with 236 reserved for test. Inputs ended at 1.5 seconds after the cue. The outcome window started at 2.5 seconds. A separate block selected the penalty for a linear correction from eight neural amplifier inputs. The reference used the target cue and early sensor value. Mean relative test squared error increased by 66.5%, with equal person and grasp weights. Five groups had more error; three selected zero correction. Planned method checks also had more mean error. Keeping the original input list did not restore a mean gain. The result applies to these linear readers and this small reused cohort. The package preserves features, raw audit trials, code, four figures, source records, and reproduction instructions.

stereoelectroencephalography · SEEG · grasp forecasting · brain-computer interface · sensor voltage · linear regression · forecast error

Admitted 0 citing publications

References

8
  1. usesFixed SEEG readers scored 80% mean balanced accuracy early and 54% later across executed grasps in four people sci:sha256:9317803e3e59d0db5ce4f76b02220e1344da603fb253dca80e67acfbce823766

    The direct feature parent used its extraction code, audit windows, and source records. This package preserves and adapts that extraction code and keeps its license notice.

  2. usesSEEG readers with eight inputs increased squared sensor error over target estimates during late grasps in four people sci:sha256:36e4b7ae7abe3acc0dcb12020c08825c3fb662eb2187c32e1470bc56dd013466

    Its continuous sensor result supplied prior evidence for the study choice. This study tests forecasts across a time gap after the target cue and early sensor are known.

  3. usesRest calibration reduces mean task accuracy in fixed SEEG imagery tests from two people sci:sha256:b2d5a1ce684c63615eb48aace1fd2a51169cfeb20f8e08e74a6759aa6964ec06

    The inherited extraction method and A/B features descend from this work. This package preserves its source and license notice through the direct feature parent.

  4. usesFixed early SEEG readers with eight inputs had 40% mean balanced accuracy for three force target ranks across executed grasps in four people sci:sha256:083c092fa74094556a2561f396868b02c01154dd262296239970e931558ede13

    This study reused its 32 feature blocks, eight raw audit trials, source metadata, retrieval receipts, and extraction code. These files supplied all 472 eligible grasp events.

  5. usesLinear SEEG readers with eight inputs and paired rest power exceeded target mean error in four people during late grasps sci:sha256:25704fffea7aae6c3e1c703f89b40b73c2a7f49bfe02bb8f11898b46dbe36c95

    Its same-window sensor test supplied the paired rest-power method and the target-conditioned force question. This study adds an earlier sensor reference, a future outcome window, and a separate block for penalty choice.

  6. usesdoi:10.82901/nemar.nm000397

    The analysis used the pinned NEMAR nm000397 version 1.0.0 conversion; source URLs, hashes, and retrieval records are preserved.

  7. usesdoi:10.5061/dryad.nq4fs

    The NEMAR conversion is derived from these source recordings, which are identified as Dryad version 1 in the saved source records.

  8. doi:10.1371/journal.pone.0150359

Cited by

0 publications

Public catalog entries on Sciscape.

No incoming citations on Sciscape.