Adding three SEEG pairs increased mean 50 ms offline theta phase error after 56 s of ridge fitting in seven rest recordings
We tested whether three extra SEEG pairs improve a 50 ms theta phase forecast after short fitting. We used four selected pairs from each of seven people in public OpenNeuro ds006065 version 1.0.0 rest data. Each linear ridge model used 56 s for fitting and 56 s for tuning. One penalty applied to all coefficients in each model. Models stayed fixed in a later 344 s test. Adding the three pairs raised mean absolute phase error from 57.76 to 62.05 degrees. The mean increase was 4.28 degrees. The descriptive 95% participant bootstrap interval was 3.08 to 5.56 degrees. All seven participant means and 27 of 28 pair means increased. The increase remained with 1 s and 3 s reference filters. All seven full source file hashes passed, and rebuilt selected traces matched exactly. A second analysis run reproduced all 14 numerical result files byte for byte. This exploratory extension reuses clinical records and known single-pair results. The offline reference uses future voltage. No live device or stimulation output was tested. The package preserves selected voltage, fitted models, forecasts, three figures, code, source records, and reproduction instructions.
stereoelectroencephalography · theta phase · phase forecasting · ridge regression · signal processing · brain-computer interfaces
References
5- usesLearned filters reduced mean 50 ms offline theta phase error by 12.06 degrees against a calibrated sinusoid in later SEEG rest from seven people sci:sha256:3adc1388ee2f8a7376426deef4c35ed0f92b2d041c8f0962bbb58c3d1b7fb38a
This publication supplied the exact selected SEEG traces, single-pair model operations, full-source retrieval code, and later test schedule. We reused these components to test added donor inputs. The target-only results matched its results.
- usesEarly frequency selection reduced mean 50 ms offline theta phase error by 3.55 degrees in seven SEEG rest recordings sci:sha256:5c3ae075960de0774e0fcc176ee07a5e3c9a353d86238f1e49023a23aa4d2bfa
This publication supplied the early theta power pair selection rule and MIT selection code. We kept its four contact-disjoint pairs per person and verified the selection from the full public source files.
- usesdoi:10.18112/openneuro.ds006065.v1.0.0
The seven analyzed rest recordings came from OpenNeuro ds006065 v1.0.0; source file hashes and extraction provenance are preserved.
- doi:10.1038/s41467-025-59417-7
- doi:10.7554/eLife.68803
Cited by
3 publicationsPublic catalog entries on Sciscape.
- usesEarly linear calibration with up to three extra SEEG pairs gave a 0.30% mean decrease in future theta forecast error in seven rest records
Provides an earlier extra-pair phase result for comparison of tested scope. It was read after the new model scores.
- usesSeparate penalties and a fallback left higher mean 50 ms offline theta phase error than own history across files from 36 SEEG patients
Its seven-patient phase forecast result motivated the question. We used its centered offline theta phase definition as a method reference. We tested separate own and donor penalties, with an exact own history fallback, on another dataset and another file.
- usesSeparate ridge penalties reduced error from a shared penalty for 50 ms offline theta forecasts in seven SEEG rest records
Uses the seven SEEG input files, fit and tuning intervals, later test interval, model code, and phase reference. The target-only and shared scores were reproduced. This study adds separate group penalties and control searches.