Windows selected from calibration gave higher mean site accuracy than 500 ms windows in five offline SEEG sessions
We tested whether calibration can choose a shorter SEEG observation window. We used five public sessions with 183 stimulation sites. Four early pulses per site set the window. Four separate later pulses per site tested it. The rule chose ends from 40 to 150 ms, with a mean of 86 ms. Mean site accuracy was 40.8 percent, compared with 34.4 percent at 500 ms and 40.6 percent at 100 ms. All five patients had higher accuracy than at 500 ms. The mean gain was 6.4 percentage points. The exact patient test gave p = 0.0625. The small sample does not establish a population gain. The mean window end was 414 ms earlier than 500 ms. Across sites, calibration and test times overlapped. Source quality selection used full recorded epochs. These are offline results within sessions. We did not measure a live interface benefit. Raw reconstruction matched all input hashes. All 11,712 decisions passed a separate distance check. The package preserves three figures, input bins, models, pulse predictions, source code, provenance, and reproduction instructions.
stereoelectroencephalography · SEEG · electrical stimulation · response window · calibration · site classification · covariance decoder · brain-computer interface
References
4- usesA fixed covariance rule gave higher SEEG site accuracy than a diagonal rule in offline tests from five patient sessions sci:sha256:1407cd9ac712a4bcd551a3738cc90c3fab309b871572cbd6fc644b41555f7886
We used the fixed cohort, voltage bins, contact list, eligibility code, source recipe, and covariance decoder. This study adds a window choice from calibration data and tests it on separate later pulses.
- usesAfter five calibration pulses, a 100 ms endpoint gave mean later AUC of 0.951 for fixed SEEG pairs in eight sessions sci:sha256:bfd74ee6170a5883952f42e0898b29d958be62240dad7866aafd0c75d34aa016
Its endpoint question helped define this test. We used another dataset and a site classification measure. We did not combine its patients or scores with our sample.
- usesdoi:10.18112/openneuro.ds004457.v1.0.2
This is the recorded OpenNeuro ds004457 version 1.0.2 source used for the five SEEG sessions; saved source hashes and retrieval records identify the analyzed files.
- doi:10.1523/JNEUROSCI.1325-22.2023
Cited by
1 publicationPublic catalog entries on Sciscape.
- usesSelection from three windows and one early window gave the same mean lowest site accuracy in five offline SEEG sessions
Its comparison of response windows in the same cohort informed the question and the local plan. We knew its report before scoring. We tested fixed-duration windows with later starts and did not use its fitted model files.