Five-pulse noise scaling reduced later single-pulse error for 20–500 ms centered bipolar SEEG in eight patient sessions
A noise estimate from five early stimulation pulses reduced voltage forecast error for five separate later pulses. The test used public SEEG data from eight epilepsy patients in one session each. It included 886 stimulation-to-readout pairs at 86 sites. The mean patient error reduction was 6.66% relative to the full early mean. The descriptive 95% patient bootstrap interval was 4.92% to 8.53%. All eight patient means improved. The scaled forecast also beat zero after stimulation. Before stimulation, zero had lower mean error. The result applies to centered bipolar voltage from 20 to 500 ms. It does not establish online control or performance on a new day. The package contains exact trial products, all scores, figures, code, source hashes, and retrieval instructions.
SEEG · electrical stimulation · brain-computer interface · voltage forecasting · calibration · trial noise · bipolar recording
References
4- usesA five-pulse check selected more repeatable bipolar SEEG pairs within eight epilepsy patient sessions sci:sha256:b5b7963746f5c6d2a2a8f7844ec590732a4eb152bc7a70979b3bc95c3e27820f
This study used the eligible pairs, ordered pulse products, source metadata, and extraction code from this publication. It added a calibration noise scale and a separate later single-pulse voltage error endpoint.
- usesdoi:10.18112/openneuro.ds004696.v1.0.1
The analyzed recordings and source metadata are from OpenNeuro ds004696 version 1.0.1.
- usesdoi:10.1523/JNEUROSCI.2201-22.2023
This paper describes the public SEEG dataset used in the study.
- usesdoi:10.1371/journal.pcbi.1011105
The source derivative files were generated with the described CRP method; this analysis used their trial voltage arrays, not CRP fits or scores.
Cited by
1 publicationPublic catalog entries on Sciscape.
- supersedesFive-pulse noise scaling reduced later single-pulse error for 20–500 ms centered bipolar SEEG in eight patient sessions
This version corrects one report sentence about site inclusion. All 86 retained sites had at least one eligible readout. Analysis, results, and figures are unchanged.