Four signed pre-pulse modes per pair raised mean relative 20–100 ms SEEG forecast error by 19.2% at 94 later sites in five sessions with ridge strength 0.1
We tested whether voltage before a pulse improves a later SEEG waveform forecast. The public source was OpenNeuro ds004457 version 1.0.2. Each site used four earlier pulses for calibration. All shared weights came from earlier sites in the same patient. The added inputs were four signed cosine modes per recording pair from -100 to -20 ms. At 20 to 100 ms, ridge strength 0.1 increased relative test error in all five sessions. The equal-patient mean increase was 19.2% at 94 later sites and 376 test pulses. The descriptive 95% patient bootstrap interval was 6.5% to 36.4%. The two-sided patient sign-flip value was 0.0625. Ridge strength 1.0 gave mixed patient results and a mean increase of 1.4%. This result does not exclude useful models with other features or penalties. The endpoint is recorded binned voltage, with possible background and residual artifact. It does not establish a causal state effect or live BCI benefit. All 110 source files matched hashes. Fresh native extraction and full numeric replay matched saved files exactly. The package preserves data, weights, forecasts, three figures, code, the local plan, source pins, and repeat instructions.
stereoelectroencephalography · waveform forecasting · brain stimulation · ridge regression · temporal validation · pre-stimulus voltage · neural signals
References
6- usesPooled site sets included the correct site for 93.6 percent of later SEEG pulses within five sessions at 20 to 100 ms sci:sha256:16a0821e6ea5302515dea9c46cffb63ec910cd0b438ee578fb5d0ecad4563dc7
Used its original native MEF retrieval and 5 ms extraction recipe through the immediate voltage source. We rebuilt the exact source subset and preserved the MIT notice.
- usesAll earlier pulses gave lower mean SEEG forecast error than four recent pulses in five recordings at 20–500 ms sci:sha256:7572e2f1a54f04ee36ae5756aff12b06f8ed5349904c849467a91c2e6236dbe9
Used its 5 ms voltage bins, public contact and event metadata, fixed recording targets, and pinned source retrieval recipe. Fresh native extraction matched the compact inputs exactly.
- usesA fixed pre-pulse voltage model had a mean 3.8% error increase at later sites in five SEEG sessions sci:sha256:75c9b7d9135d45ecf7f9a8eb06ff6798fe1e048c1713980e8ef4f61e237bf529
Used its quality loader and strict earlier-to-later site split. The prior size forecast motivated this separate signed waveform test. The complete site split matched its saved split.
- usesdoi:10.18112/openneuro.ds004457.v1.0.2
Source dataset used for the five-session SEEG analysis; the recorded version is 1.0.2.
- doi:10.1523/JNEUROSCI.1325-22.2023
- doi:10.1016/j.brs.2026.103118
Cited by
2 publicationsPublic catalog entries on Sciscape.
- usesA four-mode background model had a mean 0.62% increase in 20 to 480 ms SEEG template error after contact order reversal in eight STReEF patients at ridge strength 0.1
Its different-dataset result informed the fixed four-input and ridge comparison. We fitted the present map to background voltage alone. No code or data from this publication entered the present scores.
- usesBackground voltage prediction reduced mean centered SEEG error at 20–100 ms by 3.89 percent compared with four-pulse templates at 94 later sites in five sessions
Used its compact contact voltage inputs, event tables, fixed contact settings, source metadata, loader, and retrieval recipe. Fresh raw extraction matched all five sessions exactly. The new model uses only background voltage for shared weights. Its baseline differs from the earlier model, so their reported gains are not a matched numerical comparison.