In one pig, linear maps with 32 contacts reduced fixed voltage difference error relative to eight contacts across eight excluded current blocks
We tested whether more measured micro-SEEG contacts reduce error for the same omitted voltage differences. We used 159 pulses from one pig in OpenNeuro ds004819 version 1.0.0. Each test excluded one complete spinal stimulation current block from calibration. Nested sets of 4, 8, 16, and 32 odd-numbered contacts predicted 31 fixed differences between even-numbered contacts at 20 to 200 ms. Ridge penalties were selected inside calibration data. The mean error ratios were 0.942, 1.158, 0.898, and 0.864. A zero prediction has a ratio of 1. The 32-contact map had less error than the eight-contact map in all eight blocks. The mean decrease was 0.294; the conditional pulse resampling range was 0.058 to 0.581. The curve was not monotonic. At 32 contacts, 16 pairs with both source impedance numbers at most 1000 had a score of 1.038. This subset did not beat zero. The result concerns recorded voltage in one implant. It does not establish useful neural information or better human performance. The package preserves input data, fitted maps, code, figures, source hashes, and repeat instructions.
micro-SEEG · intracranial EEG · ridge regression · voltage prediction · stimulation · channel count
References
4- usesA linear map from eight fixed contacts left a 0.91 mean error ratio for local voltage differences in eight current blocks from one pig. sci:sha256:03676cecb0d0cc09fc5e3608deeb0ea607a07512105e9a11416492128bbbdb39
We used its prepared pulse epochs, source metadata, and pinned EDF retrieval code. We kept the same targets at all contact counts and selected penalties inside calibration data.
- usesdoi:10.18112/openneuro.ds004819.v1.0.0
The analysis used prepared pulse epochs and source files from OpenNeuro ds004819 version 1.0.0.
- doi:10.1038/s41467-023-43727-9
- doi:10.1101/2022.11.08.515705
Cited by
3 publicationsPublic catalog entries on Sciscape.
- usesProportional current scaling gave less prediction error than local interpolation from 20 to 150 ms for 36 micro-SEEG contact differences in one pig
This study uses all eight exact prepared epoch arrays, the pinned source manifest, source metadata, retrieval code, and the independent EDF reader. It fits new current-transfer models and uses no parent predictions.
- usesRidge forecasts from prior micro-SEEG voltage reduced template error but did not beat zero at 20 to 150 ms in one pig
This publication supplied the independent raw EDF check through the immediate parent. We used the second reader for a fresh full-file check of run 1 and preserved the retrieval and check code.
- usesEight selected micro-SEEG contacts reduced mean voltage error by 4% across eight current blocks in one pig at penalty 0.1, with a conditional 95% range that included zero
This work supplied the eight prepared epoch arrays, fixed omitted target differences, fixed eight-contact comparator, source manifest, and two retrieval scripts. We tested a new calibration-only contact selection rule at fixed penalties. We freshly checked all raw source files.