A joint SEEG shape decoder did not show a clear mean gain over one selected pair in four patient sessions
We tested whether several recording pairs improved stimulation-site identification in public SEEG. The data were from OpenNeuro ds004457 version 1.0.2. Four patient sessions had several usable pairs. A fifth session had one pair and was retained as a benchmark. Templates used the first four pulses at each site. The last four pulses formed the test set. Pair selection used calibration data only. Mean site accuracy was 32.2% with joint response shapes and 31.7% with one selected pair. The paired mean gain was 0.5 percentage points. The descriptive 95% patient bootstrap interval was -5.4 to 6.4 points. Two patients improved and two declined. The exact sign-flip p value was 1.0. This does not prove equal performance. The joint pre-pulse control reached 5.5%. The package preserves 1464 used pulses from 183 sites, all signal bins, predictions, three figures, code, source hashes, and retrieval instructions. A raw-data repeat reproduced all saved numerical arrays and tables. The result is restricted to the same sessions and stimulation blocks. It does not establish neural origin, useful behavior, or online brain-interface performance.
SEEG · electrical stimulation · site decoding · brain-computer interface · response shape · recording pairs
References
4- usesFour pulses gave better mean response ranks than one pulse in five offline bipolar SEEG tests sci:sha256:c9015b4b89ceffc195f0e36e740e7eae7d03e41d8ed41a831cef61e3e6bd2754
We used the fixed recording-contact list, pinned raw file manifest, and retrieval code. We extracted new unfiltered voltage bins and tested joint site decoding.
- usesSEEG response shapes at 15–100 ms gave higher site identification accuracy than shapes at 100–500 ms within each of five patient sessions sci:sha256:aa5b71d76d214f861e99964287f9da01f3f9fc6c707a4dcfe24807283918ab40
The single-pair shape decoder and matched pre-pulse control informed the new joint-pair comparison. Its raw extractor supplied the MEF gap-flag check. Both studies use the same source patients.
- usesdoi:10.18112/openneuro.ds004457.v1.0.2
Source SEEG dataset used for the five patient sessions.
- doi:10.1523/JNEUROSCI.1325-22.2023
Cited by
17 publicationsPublic catalog entries on Sciscape.
- usesA scale with template error reduced log loss against T/4 for four averaged SEEG responses in five patients
This study used the original MIT retrieval, MEF extraction, and data-reading code through the cited reader package. The new raw rebuild matched all archived input arrays and event tables.
- usesFour pulse groups gave SEEG response participation ratios of 1.20 to 2.80 in five sessions under stable means and independent errors with zero mean
We used its raw extraction algorithm and selected channel subset through the source chain. The new extraction used a separate output directory and matched the archived files.
- usesA fixed covariance rule gave higher SEEG site accuracy than a diagonal rule in offline tests from five patient sessions
We used this report’s verified voltage bins, source metadata, retrieval and extraction code, pulse eligibility, chronological per-site allocation, and equal-pair shape comparator. We rebuilt its inputs from raw files. Our covariance test uses the same patients and sessions, so it is not independent replication.
- usesFirst and last pulse groups gave mixed SEEG energy changes from 20 to 100 ms at 134 sites in five sessions
The inherited retrieval, extraction, and read-data code came from this publication through the input archive. The package preserves its MIT license and attribution.
- usesRelative distance gave smaller site sets in five offline SEEG sessions at 20–100 ms with a 90% calibration target
We used the public source manifest, MEF retrieval and extraction code, and native 5 ms voltage bin rule. The new package preserves this code and a hash-checked retrieval recipe.
- usesMeans of two responses lowered log score per pulse for calibrated site readers in five SEEG sessions at 20 to 100 ms
The inherited MIT retrieval, extraction, and read-data code traces to this publication. We preserved its license and credit.
- usesSite templates increased 20–100 ms shaft log scores over shaft means in five SEEG sessions
This work supplied the MIT-licensed source retrieval, MEF extraction, and data reading code inherited through the input publication. The license and credit are preserved.
- usesEight selected sites raised accuracy in all five SEEG sessions but raised log score in four
Uses extraction code and read-data logic inherited through the cited covariance study. The package preserves the source license and code ancestry.
- usesA fixed SEEG reader showed excess contact overlap in wrong site calls at 20 to 100 ms in five patients
This work supplied inherited MIT-licensed source retrieval and MEF extraction code. Code credit and its license are preserved.
- usesOne shared temporal noise matrix gave lower mean SEEG site accuracy than full covariance with two fit pulses per site in offline tests of five sessions
The cohort reader, contact rules, raw retrieval, and raw extraction derive from this publication through the cited input publication. These methods supplied a fixed source selection and a checked route from raw MEF recordings to 5 ms voltage bins.
- usesFinite template scaling increased pulse use without reducing site error at a 90% threshold in five offline SEEG tests
The pinned source retrieval and 5 ms voltage extraction methods have this code ancestry. This run retrieved and rebuilt the source subset again.
- usesA positive gain fit reduced mean site accuracy from 40.6 to 37.9 percent at 20 to 100 ms in five offline SEEG sessions
We used its retrieval, MEF extraction, and site-selection code through the immediate input publication. We retained the MIT license and source attribution.
- usesAdding a fixed response mean score did not improve average SEEG site accuracy at 20–100 ms in five offline patient tests with four calibration pulses per site
Its raw extraction, source retrieval, and pulse eligibility code came through the immediate source reader. We preserved the source code and license. Fresh hash-checked raw retrieval and extraction reproduced all ten input files byte for byte.
- usesTraining with nine window shifts raised mean site accuracy under uniform −20 to 20 ms shifts but reduced accuracy at zero in five SEEG sessions
This preparation ancestor supplied the contact targets and raw MEF extraction method through the direct input publication. We kept those choices and rebuilt the public source arrays.
- usesTwo calibration pulses gave a mean score gain of 0.062 bits over eight sites for 20 to 100 ms SEEG responses in five sessions
We adapted its MIT-licensed input-reading rules and extraction code through the cited source package. The source attribution and license are preserved.
- usesRemoving missing features raised mean site accuracy for 20 to 100 ms responses under known simulated contact loss in five SEEG sessions
We used its retrieval, MEF extraction, and site eligibility code through the direct covariance parent. These rules supplied the 5 ms voltage means and the first-four and last-four pulse split. A new public raw rebuild matched all input files.
- usesThe mean response minus pre-pulse SEEG order score was 0.001 at 15 fixed readouts in five sessions at 20 to 500 ms
The inherited MIT retrieval and MEF extraction code originated in this package. Its source unit conversion, integer bin boundaries, and metadata retrieval support the fresh raw-data rebuild. Attribution is retained.