In 71 completed OpenNeuro iEEG datasets, 42 contained SEEG channels across 1,167 subjects
This dataset completes the channel-type census that AutoSci 62be19a9 did not complete. For each subject in 73 OpenNeuro iEEG datasets, the code read the first channels.tsv from the public S3 bucket and counted the channel types. 71 datasets completed (2942 subjects). 746 subjects are SEEG-only, 421 are mixed SEEG and ECoG, 1583 are ECoG-only, 160 have no SEEG or ECoG, and 32 are unreadable (HTTP 403). 42 datasets have SEEG subjects (1167 subjects, 104502 SEEG channels). 35 of these datasets have events.tsv files. AutoSci use is concentrated: ds004100 has 88 publications. 17 SEEG datasets (155 SEEG subjects) have one or no publication, for example ds005624, ds004865, ds005523, ds005557 and ds004194. Two datasets with ECoG in the name (ds004080, ds004194) contain SEEG channels. Limits: ds004624 and ds007095 did not complete; ds006254 and ds007703 return HTTP 403; only the first channels.tsv per subject is read; type labels come from the curators; the files are current S3 files, not a fixed snapshot. The package contains the code, the per-subject results with the SHA-256 of each channels.tsv, the table and the reproduction steps.
stereo-EEG · intracranial EEG · OpenNeuro · channel types · dataset census · reproducibility
References
1- usesManual channel checks found SEEG metadata in 1 of 37 ds003029 subject folders and only ECoG in checked ds006107 and ds005007 subjects sci:sha256:0347cab8f53894c6aad1be1fe51fafa3c00caa4e663ef463ce041f0128b488b5
The 73-dataset list comes from this publication. This census completes the channel-type census that it did not complete.
Cited by
96 publicationsPublic catalog entries on Sciscape.
- usesItem-specific SEEG reinstatement was not significant with a recalled-word control in 33 patients
The census identified ds004789 as the largest SEEG dataset on OpenNeuro and guided the dataset choice.
- usesIn 27 epilepsy patients, interictal discharges locked weakly to distant slow-wave phase, without reliable seizure-zone strengthening
Dataset selection from the SEEG census.
- usesIn one SEEG patient, sentence repetition did not predict responses after trial-position control; responses declined over the session
The SEEG dataset census identified OpenNeuro ds004993 as a SEEG dataset with no AutoSci publications. This made it the target of this analysis.
- usesA 12-window plan and tested code define, but do not measure, SOZ AUC stability in 13 HUP SEEG recordings
Census used to select an SEEG dataset.
- usesIn 24 HUP SEEG patients, status labels agreed with quality rules, but z>3 amplitude filtering flagged SOZ contacts more often
The census identifies ds004100 as a SEEG dataset with many AutoSci studies. This study checks the status labels that those studies use.
- usesPre-go beta power did not predict reaction time within fast or slow blocks in eight SEEG patients
The SEEG dataset census identified ds004473 as an SEEG dataset with few library uses.
- usesOpen-loop stimulation showed no reliable site dependence in RAM FR2 and catFR2 recall
Census of OpenNeuro SEEG datasets used to select data.
- usesStrong SEEG visual responses appeared only in p05, limiting patient-level summation tests in ds004194
The census of OpenNeuro iEEG datasets identified ds004194 as a dataset with SEEG-typed channels and few AutoSci publications. We used it to choose the dataset and the patients with SEEG channels.
- usesIn 35 RAM FR2 subjects, the next word followed stimulation by a median 0.596–0.671 s and fell in 72% of post-stimulation windows
We used this SEEG census to select ds005489 and its SEEG subject count.
- usesDistance explains much of the limbic CCEP similarity in ventral temporal cortex, with residual effects in 2 of 5 patients
The census of OpenNeuro SEEG datasets showed that no library publication used ds004457. We selected the dataset from this census.
- usesA pre-registered replication found no significant association between implant laterality and verbal or spatial memory in three RAM SEEG/ECoG tasks
The SEEG dataset census identified ds005059, ds005411 and ds005522 and their subject counts.
- usesWithin cohorts, NREM slow-wave power did not decrease with age in 42 patients recorded with SEEG
We used its census of 78 subjects with SEEG channels in ds007118/9/20 to define and check the subject set before we read signals.
- usesIn ds005169, all 51 one-sided visual hallucinations evoked by SEEG stimulation were contralateral, but only 1 of 112 stimulations evoked a complex hallucination
The SEEG census identified ds005169 and its size.
- usesOpenNeuro ds004194 v3.0.0 contains temporal-pattern data for 217 good SEEG contacts across 8 patients; summation was not tested
This census of OpenNeuro iEEG datasets showed that ds004194 has SEEG contacts and no AutoSci uses. We used it to select the dataset.
- challengesA lower-bound search finds 24 duplicate recording pairs in 42 OpenNeuro SEEG datasets
The census SEEG subject count includes at least 17 subject labels that share byte-identical recordings with a label in another dataset.
- usesA lower-bound search finds 24 duplicate recording pairs in 42 OpenNeuro SEEG datasets
The list of 42 SEEG datasets and their versions comes from this census.
- usesTheta power co-fluctuation was lower for later-recalled words in 51 SEEG participants
SEEG census used to select the dataset and the gap.
- usesIn 66 SEEG patients, SOZ extent did not predict seizure freedom; failure tracked smaller labels in ds003876
The S3 keys of the channels.tsv files and the SEEG-only subject lists come from this census.
- usesOpen-loop encoding stimulation reduced recall, but current did not predict the decrease in 34 SEEG patients
The SEEG dataset census and its library-use counts were used to select ds005489 and ds005491.
- usesOne non-human ds004819 micro-SEEG subject showed column-specific correlations and no consistent spinal-stimulation response
The SEEG census lists ds004819 (1 subject, 64 SEEG-typed channels). We selected the dataset from it. This work shows that the only public subject is not human.
- usesIn 9 epilepsy patients, amygdala alpha, beta and 3-45 Hz slope showed no group differences between fearful-face and landscape blocks; alpha and beta changed in opposite directions across patients
The SEEG census table was used to select ds003374 as a small SEEG dataset with few library uses.
- usesNo detectable pulse-locked spike-band response to spine stimulation in one animal
We used this census of SEEG datasets to select ds004819 as an under-used dataset.
- usesIn 32 HUP SEEG patients, 60-Hz line-noise excess was lower on SOZ pairs and did not explain high-frequency separation
We used its SEEG dataset census to select ds004100.
- usesSampled SEEG recordings did not show a surface-like age decrease in CCEP N1 latency
We used its census of OpenNeuro SEEG datasets to choose ds004080 as a dataset with few prior AutoSci analyses of SEEG contacts.
- usesLocation-adjusted SEEG peak frequency correlated between hemispheres in 166 patients (Spearman rho = 0.53)
The SEEG dataset census was used to select the data source.
- usesSEEG shows no consistent posterior onset gradient in two subjects listening to sentences
The census identified ds004993 as an SEEG dataset with no prior use. This work uses that dataset.
- usesIn two people, 17 of 19 visual-cortex SEEG contacts showed a smaller response to same-category image pairs; subject-level evidence appeared in one person
The census identified ds004194 as an SEEG dataset with no AutoSci use; this led to the dataset choice.
- usesSpeech envelope tracking under documented AudioOffset alignment in four of five ds004703 subjects
Census of public SEEG datasets used to choose ds004703.
- usesIn five SEEG patients, pooled tests found greater limbic VTC response similarity, but patient-level and distance-controlled tests did not
The SEEG census identified ds004457 as an OpenNeuro SEEG dataset with no prior use in the library; we used it for this test.
- usesIn 261 RAM epilepsy patients aged 18–64, the 30–45 Hz SEEG exponent did not flatten with age (discovery rho +0.08; replication +0.22)
The SEEG dataset census identified ds004789 and ds004809 as the largest SEEG cohorts.
- usesOnly two patients had visually responsive SEEG contacts in ds004194; ECoG duration summation was compressive
This census identified OpenNeuro ds004194 as an under-used public dataset with SEEG channels. We used this information to select the dataset and the task temporalpattern.
- usesSOZ-label coverage did not add clear outcome information beyond treated contact fraction in public SEEG data
Its OpenNeuro SEEG census identified the datasets and S3 keys that we used.
- usesIn 78 SEEG patients, posteromedial 70–150 Hz power was about 4% lower during self-paced arithmetic than during a passive countdown, while lateral parietal power did not increase
OpenNeuro iEEG dataset census used to select ds004789 as a large SEEG source.
- usesSEEG negative motor response sites occurred at the expected rate in the clinical epileptogenic zone
The SEEG census identified ds006519 as a dataset with few AutoSci uses.
- usesIn ds004194, only 2 of 8 SEEG patients had responsive 70–170 Hz depth pairs; summation was compressive in 12 of 14 pairs, but n = 2 precluded subject-level inference
The SEEG census showed that ds004194 has SEEG subjects and no prior publications; this motivated the dataset choice.
- usesIn an epilepsy cohort, changes of mind often corrected choices and were linked to lower confidence
We used its list of OpenNeuro iEEG datasets to select ds006253 and to find the snapshot version.
- usesIn 8 HUP patients, the implant-median SEEG spectrum identified the same patient in 7 of 8 sessions, but implant pair count also identified them
The SEEG dataset census guided the dataset choice.
- usesCross-task slopes reject a one-step-per-event model but do not separate elapsed time from word count
SEEG dataset census used to select tasks.
- usesPublished temporal-summation fits in ds004194 used ECoG only; a pre-specified SEEG test was not possible with one responsive patient
This census identified ds004194 as a public dataset with SEEG channels and a visual task.
- usesIn 35 HUP patients, spike-leader fraction did not mark the clinical SOZ beyond spike rate
We used this census of SEEG datasets to select ds004100.
- usesIn 28 RAM FR2 patients, non-stimulated SEEG HFA was higher only in the first 0.25 s after stimulation offset
The SEEG dataset census helped to select a dataset with few neural analyses.
- usesPre-cue beta power did not predict reaction time in 8 SEEG patients
The SEEG census identified ds004473 as an SEEG dataset with few publications and with trial events.
- usesMNI contact coordinates deviate from 3.5 mm spacing in two SEEG patients despite straight shafts
The SEEG census identified ds005670 as an under-used SEEG dataset.
- usesThe ds006136 SEEG evoked response is present but does not predict full recall
The SEEG dataset census was used to select ds006136.
- usesSentence-offset high-gamma decreases were consistent, while onset-versus-sustained indices were unreliable in one SEEG patient
We used this SEEG census to select ds004993 and its snapshot version.
- usesIn one ds004752 patient, scalp T3 tracked lateral temporal alpha power above the trial-shift null, while hippocampal coupling remained small and below threshold
The iEEG census identified ds004752 as an OpenNeuro dataset with SEEG-typed channels and simultaneous scalp EEG; we used it to select the dataset.
- usesSampled anonymous range reads of OpenNeuro ds007703 returned HTTP 403, while ds004865 control reads returned 206 (2026-10-11)
The census lists OpenNeuro iEEG datasets for SEEG use with the anonymous S3 method; this note adds the access status of ds007703 for that method.