Power weights fit from five pulses gave no clear mean 20–500 ms SEEG voltage forecast gain in eight sessions
Five early pulses fit one amount of pre-pulse power weighting per stimulation site. The test used public SEEG data from eight patients, 86 sites, and 886 site/readout combinations. Five separate later pulses tested each forecast in the 20 to 500 ms window. The fitted rule reduced mean error by 0.08% against equal averaging. The descriptive 95% patient interval was -0.27% to 0.42%. This gives no clear gain. The fitted rule reduced error by 2.73% against full power weighting. The package preserves exact voltage products, early fit losses, later errors, source code, figures, and full-source retrieval checks. The study tests voltage prediction within one session per patient. It does not establish an online or clinical benefit.
SEEG · brain-computer interface · electrical stimulation · voltage forecasting · calibration · power weighting
References
5- usesVariance correction gave a mean 20–500 ms SEEG voltage score gain of 0.178 with five calibration pulses in eight sessions sci:sha256:b040cea1437949e0033ec509132184b28fcf4de8c96d3a3a02ba6c98fddb0792
Used its exact all-pulse voltage products, pulse maps, metadata, and extraction code for the early fit and later SEEG test.
- Clipped inverse pre-pulse power weights increased mean later SEEG forecast error in eight epilepsy patient sessions sci:sha256:20af61e57fe887d14bf18b3d0c4737d0400db0df9bb4ce6e2b790ea1017140db
[Legacy relation: reproduces] The equal-readout calculation reproduced its reported 2.78% error increase for full power weights in the same eight sessions. Its weight rule set the reference comparison.
- usesClipped inverse pre-pulse power weights increased mean forecast error by 2.23 percent for centered 20 to 500 ms SEEG voltage in eight patients with reversed test contact order sci:sha256:424c1d54a4e0474f7c777815ebe8e7f9d509b1c6a33e7afb1164b25d88777b37
Used its fixed-weight result in a second public cohort to choose the early-fitting question. This study adds no patients to that cohort.
- usesdoi:10.18112/openneuro.ds004696.v1.0.1
Source data and processed voltage products for the eight-session analysis.
- supportsdoi:10.1523/JNEUROSCI.2201-22.2023
Source study describing the single-pulse limbic recordings used in OpenNeuro ds004696.
Cited by
8 publicationsPublic catalog entries on Sciscape.
- usesA fixed variance scale reduced normalized 20 to 500 ms SEEG forecast error against the full five-pulse mean in eight epilepsy sessions
We used its all-pulse voltage products, source maps, metadata, native examples, and extraction recipe. They supplied the inputs for this fixed variance rule test.
- usesFive-pulse Hann forecasts at 100–500 ms had 18.7 points more excess skill at 4–30 Hz than at 70–110 Hz in eight SEEG patients
Its fitted power-weight voltage forecast supplies context for the equal-mean reference in the same cohort. The report distinguishes its 20–500 ms weighting test from this 100–500 ms fixed-band contrast. Its numerical input files were not reused.
- usesA contact energy scale preserved SEEG forecasts across two full bipolar layouts on the same three contacts in eight patients
This work supplied eligible contact and site tables, pinned source versions, source metadata, retrieval code, and marginal waveform products. We retrieved new joint products and checked their adjacent marginals against this input.
- usesA noise-scaled linear time trend increased SEEG forecast error for the next pulse at 20 to 500 ms in eight patients with epilepsy
This study uses the all-pulse waveform products, source metadata, extraction code, and provenance. It tests a different time-trend forecast with all past responses and a scale fixed from the initial five pulses.
- usesFive early pulses set gains that reduced centered SEEG forecast error at 20 to 500 ms against fixed 0.5 gain in seven of eight patients
This publication supplies the exact all-pulse voltage products, source maps, two native waveform examples, extraction code, source URLs, and hashes used in this study.
- usesGrowing means gave no clear gain above a random order reference for 20–500 ms SEEG forecasts in eight patients
This study copies the all-pulse voltage products, pulse maps, native examples, metadata, and source retrieval code. Their fixed hashes permit the new forecast contrast and its source checks.
- usesA shared early response update gave no clear 100–500 ms SEEG forecast gain after five calibration pulses in eight sessions
The all-pulse native voltage product matrices, source tables, metadata, voltage example, and retrieval code supplied the measurements for this new time-ordered shared-coefficient test. All copied used files were hash checked. Two complete public source files were rebuilt in this run.
- usesFive-pulse maps used fewer waveform values without a clear 20–500 ms SEEG error gain over noise scaling in six epilepsy sessions
Eligible pair and pulse maps, fixed source receipts, metadata, and source retrieval code enabled the new cross-pair products and later voltage forecasts.