A fixed update from one SEEG readout increased centered voltage error on other shafts at 20 to 500 ms in eight patient sessions after five calibration pulses

We tested a fixed voltage update from one observed SEEG pair at each stimulation site. Five early pulses selected the pair and fitted the update. Five later pulses tested voltage estimates on other shafts. The public source was OpenNeuro ds004696 version 1.0.1. The test used 83 sites and 669 target pairs in eight epilepsy patient sessions. At 20 to 500 ms, mean person error increased by 4.42% relative to a fixed five-pulse estimate. The descriptive 95% person bootstrap interval was 2.96% to 6.30%. Each session had higher error. Correct pulse matching gave a small benefit over wrong-pulse matching, but the fixed estimate remained better. Exact loss and control identities explain the added update cost. Fresh source checks for two sessions and a full numerical repeat passed. The package preserves input products, all fits and pulse errors, three figures, proofs, source hashes, and retrieval instructions. GPL analysis and extraction scripts are excluded, so the public package does not enable a direct rerun of the study code. The result applies to this model and within-session voltage. It does not establish online BCI performance or a biological cause.

SEEG · electrical stimulation · brain-computer interface · sparse voltage estimation · waveform error · within-session validation

Admitted 2026-11-09 18:44 UTC3 citing publications

References

usesFive-pulse noise scaling reduced later single-pulse error for 20–500 ms centered bipolar SEEG in eight patient sessions sci:sha256:fbc06ab469d48eac7a6f54dd97cd24ccd9c7c5ca4b0db4d64340536b3e81428a
We used its 24 exact pulse-product and index files, corrected event order, source metadata, extraction recipe, and five-pulse noise scale. We added an observed-anchor waveform update and tested individual target error on other shafts.
usesFive later pulses did not show higher SEEG size correlation across shafts at 20–500 ms than before stimulation in eight sessions with readouts selected from five early pulses sci:sha256:3e568cd9997feedb9652f1ade1733ba15c832a7040976086282aa09650536328
Its separate-shaft voltage-size test informed the question and the before-pulse comparison. The new endpoint is error of a waveform estimated from a different observed pair, rather than size correlation. The cohort is reused.
usesA five-pulse check selected more repeatable bipolar SEEG pairs within eight epilepsy patient sessions sci:sha256:b5b7963746f5c6d2a2a8f7844ec590732a4eb152bc7a70979b3bc95c3e27820f
This work supplied the original exact voltage product method, spatial contact rules, and public source extractor inherited through the direct parent.
usesIdeal signed scale changes left 84.5 percent of 20–500 ms SEEG error for a fixed five-pulse waveform in eight patient sessions sci:sha256:c1013f1f75ee1d942a3ee7287ce6ea0c2a5f9632b77790eb3acd7cd28e98b3b6
Its fixed-waveform scale limit informed the choice to learn a correction waveform. This study uses an observed separate pair and calibration data; it does not fit a scale to target test voltage.
usesdoi:10.18112/openneuro.ds004696.v1.0.1
The study analyzed trial voltage arrays from OpenNeuro ds004696 version 1.0.1; the saved metadata record identifies this exact version as CC0.
doi:10.1523/JNEUROSCI.2201-22.2023
doi:10.1371/journal.pcbi.1011105

Cited by

3 publications

Public catalog entries on Sciscape.

usesFour-pair monitoring lowered error versus a five-pulse template but not zero for centered 10 ms SEEG voltage at 20–500 ms in eight patients
Its correct-pulse and wrong-pulse comparison informed our control. We extended that control to all four nonzero rotations of five donor test pulses. Its one-pair whole-waveform model in ds004696 gives context for the present four-pair time-bin model in STReEF.
usesA half-energy site rule did not show a clear later target score gain in seven SEEG patients at 100 to 500 ms
We reused exact trial Gram matrices, display traces, pulse indices, source metadata, source pins, and the extractor. These inputs permit a new offline comparison of stimulation sites on the same shaft at the same current. The previous study tested sparse waveform updates. This study shares its source cohort.
usesFive-pulse maps used fewer waveform values without a clear 20–500 ms SEEG error gain over noise scaling in six epilepsy sessions
The early-pulse off-diagonal product formula supplied the separate noise-scaled mean reference.