Five recent voltage means reduced mean 50 ms offline theta phase error by 6.73 degrees after causal input filtering in seven SEEG rest records
Five means from the last 40 ms of raw bipolar voltage reduced mean 50 ms offline theta phase error after causal input filtering. We used 28 fixed SEEG pairs in seven public rest records from OpenNeuro ds006065 version 1.0.0. The added path retained 25 causal filtered values and added five recent means. Mean error fell from 62.66 to 55.93 degrees. The paired mean reduction was 6.73 degrees. All seven people improved. The descriptive 95 percent person bootstrap interval was 6.06 to 7.42 degrees. The exact two-sided sign-flip value was 0.015625 under independent, symmetric person differences. Thirty filtered history values and five older raw means did not give the gain. A recent 20 ms control had 0.63 degrees less mean error than the main 40 ms method. The model used 30 inputs instead of 250, an 88 percent reduction. Filter cost raised total counted multiplications at the tested 10 Hz output rate. The package includes source arrays, all models and forecasts, three figures, a chunk-based reader, source hashes, and reproduction code. A second solver checked all 252 models. A complete replay matched all numerical outputs. Fresh downloads checked seven complete raw files and exactly reproduced all 3,360,000 stored voltage values. This is an exploratory extension on a reused cohort. The phase reference uses future data. Hardware deadlines, energy, stimulation effects, behavior, and transfer to new sessions were not tested.
theta phase prediction · stereoelectroencephalography · SEEG · intracranial EEG · causal filtering · ridge regression · voltage means · offline phase reference
References
4- usesA causal 15 Hz filter with 50 Hz inputs increased mean 50 ms offline theta phase error by 5.69 degrees in seven SEEG rest records sci:sha256:b0bc5d8c2166828653850e2e97e734d11101148069b5e205847620816027148a
We use the causal 15 Hz filter baseline, reference phase definition, ridge code, calibration and test blocks, and archived baseline forecasts. We add recent raw means and equal-input-count controls. Four inherited paths reproduce 129,920 prior forecasts within 1.24e-12 radians.
- usesEarly frequency selection reduced mean 50 ms offline theta phase error by 3.55 degrees in seven SEEG rest recordings sci:sha256:5c3ae075960de0774e0fcc176ee07a5e3c9a353d86238f1e49023a23aa4d2bfa
We use all seven selected voltage arrays, all 28 fixed pairs, the pinned source metadata and hashes, and the retrieval code. Fresh complete raw-file checks reproduce every stored voltage value and selected pair.
- usesdoi:10.18112/openneuro.ds006065.v1.0.0
The analysis used one pre-stimulation rest record from each of seven people in this exact OpenNeuro dataset version; the selected arrays and full source hashes are preserved.
- doi:10.1038/s41467-025-59417-7
Cited by
2 publicationsPublic catalog entries on Sciscape.
- usesSeparate delay fits reduced mean offline theta phase error by 1.45 degrees at five known delays in seven SEEG rest records after stimulation
This publication supplied the hybrid40 causal input rule, centered phase reference, and ridge fit rule. We adapted those rules to compare delay rotation with separate delay fits on later post-rest voltage.
- usesA 100 ms voltage history gave 1.16 degrees less mean offline theta phase error than 500 ms in seven SEEG rest records at 25 inputs and a 50 ms forecast
We use all seven selected voltage arrays, the source provenance and retrieval code, the ridge fit, the offline phase reference, and the calibration and test blocks. We add ten history methods and three forecast horizons. The two inherited 500 ms paths reproduce 64,960 archived forecasts within 1.24e-12 radians.