Filters trained to omit gaps did not lower observed mean theta phase error after simulated 50 ms gaps in seven SEEG rest records
We tested whether a filter trained to omit a known signal gap gives better theta phase forecasts after sensing resumes. We used one public SEEG rest record from each of seven people in OpenNeuro ds006065 version 1.0.0. Early data set all model parameters. The primary test used a simulated 50 ms gap, a 20 ms forecast delay, and four recovery offsets from 0 to 100 ms. Mean absolute offline phase error was 57.07 degrees with the filter trained to omit gaps and 56.66 degrees with linear interpolation. Interpolation error minus omit-gap error was -0.41 degrees. The descriptive 95 percent person bootstrap interval was -0.85 to 0.15 degrees. The exact two-sided sign-flip probability was 0.21875. Two of seven people improved with gap training. The result does not prove equal performance. No-gap error was 54.12 degrees. Holding the last value gave 61.24 degrees. Both offline references kept the same mean method order. Mean error could return near its no-gap value while the phase output still differed. The study reuses an earlier group and tests simulated sample loss. It does not measure biological phase truth or a stimulation benefit. The package preserves source samples, provenance, all predictions, fitted filters, source code, three figures, and exact numeric repeat checks.
theta phase · SEEG · signal prediction · signal gaps · interpolation · ridge regression · brain-computer interfaces
References
4- usesTraining for a 50 ms delay reduced mean offline theta phase disagreement in seven SEEG rest records sci:sha256:f768857b688f4dba8d3e9a0af3ef4aebb78c20d0e3cf4dd44b133d69dbdbae01
This publication supplied preserved contact samples, audit segments, source hashes, the source retrieval recipe, the finite complex phase reference, and the ridge method. We added known-gap masks, gap-specific filters, interpolation and hold comparisons, recovery measures, and independent checks. The seven people and later test period are reused.
- usesAn early amplitude gate selects lower offline theta phase disagreement for a fixed 20 ms SEEG forecast in seven people sci:sha256:1a24afec3bc8510b7da72f27f008188f73fd7121b505514dcd06427be493d7e5
This earlier work established the seven metadata-selected adjacent SEEG pairs and the early amplitude threshold rule. We retained that source-pair selection through the cited phase study. The high-amplitude secondary control uses the unchanged clean forecast, rather than an online gate for gap input.
- usesdoi:10.18112/openneuro.ds006065.v1.0.0
The publication used OpenNeuro ds006065 version 1.0.0; this is the recorded version-specific dataset DOI.
- doi:10.1038/s41467-025-59417-7
Cited by
2 publicationsPublic catalog entries on Sciscape.
- usesSimulated pulse schedules matched with and without 25-sample gaps at a 301-sample interval in seven SEEG rest records with fixed readers and delays
We used its description of forecast recovery after hypothetical sensing gaps to define the new gap question. Its tests used separate gaps at fixed query times. We instead tested each policy with its own repeated output-caused gaps and measured exact pulse schedules. We did not use its voltage arrays or fitted filters.
- usesThree-bit error feedback reduced changes in a fixed 50 ms phase forecast in seven SEEG rest records
We reused its compact original contact arrays, pinned source manifest, metadata, retrieval and source-check code, and basic complex ridge implementation. We checked all selected values against fresh public source ranges. We fit a new fixed 50 ms reader and tested a new digital encoding question.