Four-pulse noise correction reduced forecast error for later SEEG energy estimates in three of five patient sessions
We tested whether correction for trial noise improves response forecasts from four early pulses per stimulation site. We used OpenNeuro ds004457 version 1.0.2. The test included five patient sessions, nine bipolar recording pairs, and 324 pair-site combinations. The target used products from separate later trials in a 30 to 250 ms window. Correction reduced error in three patients and increased it in two. The geometric patient error ratio was 0.957. Its descriptive 95 percent bootstrap interval was 0.858 to 1.068. The two-sided sign-flip probability was 0.5625. These data do not establish a general benefit. Setting negative forecasts to zero gave a secondary error ratio of 0.909 and lower error in four patients. This result is exploratory. The package preserves voltage epochs, all forecasts, figures, code, source hashes, and a raw retrieval recipe. Fresh raw extraction matched all 63 saved arrays exactly. An independent calculation confirmed the primary scores.
stereoelectroencephalography · SEEG · evoked responses · trial noise correction · forecast error · response calibration
References
4- usesThree early valid 6 mA pulses per site selected sites with greater later SEEG voltage energy than random expectation in five patient sessions sci:sha256:2986e9c79b14eb86379a6faea58142d182aaac35b734ad23f0dd1db049ae222a
We used its nine voltage epoch files, source metadata, eligibility rules, and unchanged raw retrieval and extraction code. These files enabled the separate-trial energy forecast test. Fresh raw extraction reproduced all 63 saved arrays exactly.
- usesFour pulses gave better mean response ranks than one pulse in five offline bipolar SEEG tests sci:sha256:c9015b4b89ceffc195f0e36e740e7eae7d03e41d8ed41a831cef61e3e6bd2754
Its four-pulse response-rank test and proposed target with noise correction motivated our numerical forecast test. Our later target uses distinct trial products and our primary outcome is patient-level squared forecast error.
- usesdoi:10.18112/openneuro.ds004457.v1.0.2
The study used the 1.0.2 dataset files and metadata for the five SEEG patient sessions; the fixed version-specific DOI and CC0 license are recorded in the saved dataset metadata.
- doi:10.1523/JNEUROSCI.1325-22.2023
Cited by
2 publicationsPublic catalog entries on Sciscape.
- usesCorrection with four pulses kept all 15 site choices in five offline SEEG sessions from 20 to 500 ms
This prior study supplied the pulse-product energy correction and the separate later-pulse scoring concept. It tested numerical forecast error in a different contact panel and time window. The present study tests the resulting top site choices in 15 pairs from 20 to 500 ms. It does not claim to invent the estimator.
- usesClipped trial products reduced geometric mean squared forecast error for later 20–500 ms SEEG energy estimates in eight sessions
This study tests the earlier exploratory clipping proposal with a fixed analysis in a different public dataset. It uses five early and five later pulses in ds004696, rather than the earlier four-pulse ds004457 design.