One waveform learned from two pulses per site missed repeated site differences in five SEEG sessions

1. usesA fixed covariance rule gave higher SEEG site accuracy than a diagonal rule in offline tests from five patient sessions sci:sha256:1407cd9ac712a4bcd551a3738cc90c3fab309b871572cbd6fc644b41555f7886

   This archive supplies all five patients, hash-checked 5 ms voltage bins, recording pair and site rules, raw extraction code, and source provenance. The new test measures repeat power outside a direction learned from separate pulses.

2. usesA learned time basis did not show lower mean one-pulse SEEG forecast error at 94 new sites in five patient sessions sci:sha256:7aa30d9799432d4754152e931d764d9f390ea9b971b9c6e176501180aabbeebf

   This report defines the prior learned-time-basis forecast test at new sites. It helped define the distinct scope of this test: compression of repeated differences at the same stimulation sites.

3. usesA voltage model trained on pulses gave mean error decreases of 2.08 percent after pulses and 2.69 percent without pulses at 95 later sites in five SEEG sessions sci:sha256:a25622ee5a1f9f974c61821bdc73d48fc47797bc227fa8730ce7cfb912ac2357

   This report found voltage forecast gains during periods without pulses in the same source dataset. It supplies context for the matched background control and the limit on causal interpretation.

4. usesdoi:10.18112/openneuro.ds004457.v1.0.2

   The analysis uses hash-checked archived voltage bins and source metadata for OpenNeuro ds004457 version 1.0.2.

5. doi:10.1523/JNEUROSCI.1325-22.2023
6. doi:10.1073/pnas.2506535122

Public catalog entries on Sciscape.

• usesA fixed SEEG model increased shape error by a mean of 15.3 percent in five sessions at 20 to 100 ms

  Used voltage bins, contact targets, and the source retrieval recipe. Fresh extraction from the public source files matched all archived array fields. The forecast test and error audit are new.

• 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 voltage bins, source metadata, and cohort selection code. Its learned-waveform residual result motivated a moment estimate that does not fit a waveform.

• usesSets of eight sites selected with four calibration pulses per site gave 97.5% mean SEEG site accuracy on later pulses in five sessions

  Its test showed repeated site differences outside one learned waveform direction in this cohort. That result helped us choose a fixed-size site selection question. We did not copy data or code from that package.

• usesAt ranks set by 80% pre-pulse fit energy, pre-pulse PCA retained less repeated site voltage than response PCA in five SEEG sessions at 20–500 ms

  This publication supplied the two-pulse fit and two separate three-pulse scoring groups, plus the repeated site cross-product method. The new analysis fits pre-pulse and response PCA bases at the same rank. No prior fitted mode or score was copied.

• usesFour-pulse peaks showed no clear mean error gain over shared peaks in five SEEG sessions within 20 to 500 ms

  We reused the five-session ds004457 voltage bins, BIDS metadata, fixed contact targets, and site eligibility rules. We recalculated peak times and fit a new shared median comparison. We did not reuse fitted waveform bases or claim an independent patient cohort.

• usesMean estimated SEEG site map alignment was 0.28 for 20 to 100 ms and 100 to 500 ms in five sessions under stable means and independent group errors

  Used its full preserved voltage bins, event tables, source metadata, fixed cohort selection, and retained pulse membership. The new analysis estimates agreement between early and late site maps. It does not repeat its waveform compression test.

• usesFirst and final four-pulse SEEG means had a mean ratio of 1.071 against balanced splits in five sessions at 20 to 500 ms

  Supplies the preserved 5 ms contact voltage bins, event tables, BIDS metadata, fixed recording-contact list, source manifest, and raw-code lineage. All 45 reused data and metadata files matched its archived hashes. This study uses the same five patients.

• usesFour pulse groups give an unbiased fixed SEEG map score under stable means and independent errors

  The earlier study shows that residual power outside a learned first waveform can be positive for a true rank-one map. This limitation sets the question for a four-group score that uses no fitted direction.

• usesPre-pulse ridge updates had lower mean relative forecast error than early updates in five SEEG replays with earlier fits from other shafts

  Its fixed readout selection was inherited through the direct data parent. The selection and MIT notice are preserved.

• usesTime order raised mean site accuracy in five offline SEEG sessions with four fit pulses and normalized voltage at 100 to 500 ms

  This record supplied the fixed readout and site cohort rules through the input publication. We kept those rules and used the same five patient sessions. This is reused data, not an independent cohort.

• usesMean SEEG forecast gains fell from 14.1 to 5.6 percent when whole shafts were excluded in five sessions

  This publication defined the cohort and first-four/final-four pulse rules inherited through the input package. The new test keeps those rules.

• usesEnergy shared across four pulses gave no clear offline site-rank gain over total energy in five SEEG sessions at 100 to 500 ms

  The fixed readout targets and eligible site and pulse rules came from this cohort through the direct input publication. The five people are reused. No fitted waveform basis from that work is used here.

• usesA 90% calibration voltage map left positive test contrast outside its space in five offline SEEG sessions

  Its separate-group residual test supplied the closest prior question and interpretation limit. We extend the fixed rank-one voltage test to the 90% rank rule and compare voltage with calibration noise geometry.

• usesWith four pulses per site, late calibration gave higher late SEEG scores in five offline sessions

  We used its inherited target-contact and pulse-selection rules through the immediate input package. We retained the first four and final four eligible pulses, then added exact 6 mA and whole-target-shaft exclusions.

• usesA five-point fourth-pulse rule lost 7.9 percentage points in one of five SEEG sessions at 20 to 100 ms

  Its repeated-response rank results motivated the question of selecting the number of response directions from calibration data. We read it before saving the local plan. Its data or code did not supply the new selector outcomes.

• usesVariation in four early pulses gave lower SEEG error-limit scores than response energy at 20–500 ms in five patients

  The inherited cohort rule and bipolar pair selection were adapted from this work. We kept the selection fixed for this new scale comparison.

• usesA 90% timing prior gave no clear accuracy gain over nine shifted SEEG templates in five offline sessions

  This is a credited upstream source for the preserved voltage bins, source hashes, and MEF extraction pipeline. We rebuilt the pinned raw source and checked all bin arrays and event fields.

• usesFull covariance did not establish a mean increase in SEEG response spread at 20 to 100 ms in five sessions under stable means and independent errors with zero mean

  Used the upstream pinned input provenance and good-contact cohort rules received through b2613210. Preserved its MIT code notice and checked the input hashes. The same source sessions were reused.

• usesLate SEEG features gave no clear accuracy gain at eight values per pair in five offline sessions

  We received voltage bins, events, metadata, extraction code, and the base cohort rules from this work.

• usesWindow mean voltage supplied 98.2% of the mean 20 to 100 ms SEEG order score at fixed readouts in five 6 mA sessions

  This earlier study describes the fixed recording pairs reused here. The new outcome is later-pulse order similarity. No contact was selected with the new scores.

• usesEight learned shapes increased mean relative SEEG prediction error with two fit pulses in five offline sessions at 20–500 ms

  Supplied the hash-checked voltage bins, fixed cohort rules, and raw extraction recipe. Its response-shape result helped define the new two-pulse test.