EEGdash›NeMAR›NM000454
Iss. 454 · 15 subjects · 1261 recordings · CC-BY-4.0
Dataset Brief · A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla M…

NM000454: ieeg dataset, 15 subjects#

A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla MRI and intracranial EEG (Penn 3T-7T, Pennsieve 662)

Access recordings and metadata through EEGDash.

Citation: Alfredo Lucas, Mariam Josyula, Nina Petillo, Joost Wagenaar, Joel Stein, Sandhitsu Das, Nishant Sinha, Kathryn Davis, Tamjid Imtiaz (20). A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla MRI and intracranial EEG (Penn 3T-7T, Pennsieve 662). 10.82901/nemar.nm000454

Modality: ieeg Subjects: 15 Recordings: 1261 License: CC-BY-4.0 Source: nemar

Metadata: Complete (100%)

15-participant iEEG dataset — A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla MRI and intracranial EEG (Penn 3T-7T, Pennsieve 662).

Data & curation Alfredo Lucas · Mariam Josyula · Nina Petillo · Joost Wagenaar · Joel Stein · Sandhitsu Das · …
Senior author Tamjid Imtiaz
Year 20 · Distributed via NeMAR
Funding NINDS K99NS138680 / R00NS138680 (PI: Sinha) · NINDS R01NS116504 (PI: Davis) · + 4 more
iEEG · 136 (246), 80 (147), 125 (112), 101 (109), 120 (98), 109 (85), 118 (84), 102 (75), 172 (71), 149 (60), 128 (51), 165 (45), 161 (45), 122 (20), 190 (8), 140 (5) ch256, 500, 512, 1024, 2048 HzBIDS 1.10.01255 tasks
Layer 01Study
What was asked
Hypothesis, independent & dependent variables, paradigm, cohort, and the editorial caveats around what the recordings can and cannot answer.
Layer 02Signal · BIDS
What was recorded
Sidecars, channels & electrodes, coordinate system, event semantics, and quality stats from the NEMAR pipeline when available.
Layer 03Training · ML
What you can train on
Recommended access modes — MNE Raw, braindecode windows, PyTorch DataLoader — plus the targets the metadata makes addressable.
§ 01Access · Get started

Quickstart#

Install

pip install eegdash

Access the data

from eegdash.dataset import NM000454

dataset = NM000454(cache_dir="./data")
# Get the raw object of the first recording
raw = dataset.datasets[0].raw
print(raw.info)

Filter by subject

dataset = NM000454(cache_dir="./data", subject="01")

Advanced query

dataset = NM000454(
    cache_dir="./data",
    query={"subject": {"$in": ["01", "02"]}},
)

Iterate recordings

for rec in dataset:
    print(rec.subject, rec.raw.info['sfreq'])

If you use this dataset in your research, please cite the original authors.

BibTeX

@dataset{nm000454,
  title = {A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla MRI and intracranial EEG (Penn 3T-7T, Pennsieve 662)},
  author = {Alfredo Lucas and Mariam Josyula and Nina Petillo and Joost Wagenaar and Joel Stein and Sandhitsu Das and Nishant Sinha and Kathryn Davis and Tamjid Imtiaz},
  doi = {10.82901/nemar.nm000454},
  url = {https://doi.org/10.82901/nemar.nm000454},
}
§ 02Study · The README

About This Dataset#

Redistribution on NEMAR of Pennsieve Discover dataset 662, version 1 (doi:10.26275/aosz-hj9u, licence CC-BY-4.0),

described in Sci Data 2026, doi:10.1038/s41597-026-07540-5. Authors: Alfredo Lucas, Mariam Josyula, Nina Petillo, Joost Wagenaar, Joel Stein, Sandhitsu Das, Nishant Sinha, Kathryn Davis, Tamjid Imtiaz.

This dataset comprises multimodal paired 3T and 7T MRI scans collected from 30 patients with drug-resistant focal epilepsy along with intracranial EEG data.

DOI

A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla MRI and intracranial EEG

Details (from the paper and the deposit)

c-penn3t7t: Penn 3T-7T Paired Epilepsy Imaging (Pennsieve 662 v1, doi:10.26275/aosz-hj9u)

**Task / paradigm.**30 patients with drug-resistant focal epilepsy with paired 3T and 7T MRI (structural, resting fMRI); for the 15 who later had intracranial EEG, curated ictal and interictal epochs. Ictal epochs from EMU epileptologist notes; interictal epochs randomly selected while awake, >72 h after implantation, >=2 h before seizures and >=2/6/12 h after subclinical/focal/focal-to-bilateral seizures (doi:10.1038/s41597-026-07540-5 (Sci Data 2026, “A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla MRI and intracranial EEG”, PMC13582879), Methods “Intracranial electroencephalography”). The deposit has 1261 EDF epochs under sourcedata/sub-*/ses-clinical01/ieeg with BIDS channels/events/ieeg.json sidecars.

View full README

DOI

A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla MRI and intracranial EEG

Details (from the paper and the deposit)

c-penn3t7t: Penn 3T-7T Paired Epilepsy Imaging (Pennsieve 662 v1, doi:10.26275/aosz-hj9u)

Task / paradigm.**30 patients with drug-resistant focal epilepsy with paired 3T and 7T MRI (structural, resting fMRI); for the 15 who later had intracranial EEG, curated ictal and interictal epochs. Ictal epochs from EMU epileptologist notes; interictal epochs randomly selected while awake, >72 h after implantation, >=2 h before seizures and >=2/6/12 h after subclinical/focal/focal-to-bilateral seizures (doi:10.1038/s41597-026-07540-5 (Sci Data 2026, “A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla MRI and intracranial EEG”, PMC13582879), Methods “Intracranial electroencephalography”). The deposit has 1261 EDF epochs under sourcedata/sub-*/ses-clinical01/ieeg with BIDS channels/events/ieeg.json sidecars. **Recording system. Sampling rates 500, 512 or 1024 Hz (paper Methods). Preprocessing described in the paper: 60 Hz notch (Q=30) and 3rd-order Butterworth 1-120 Hz, forward-backward (paper Methods). Amplifier not stated. Reference scheme.**Not stated in the paper text extracted (n/a); see the *_ieeg.json iEEGReference of the deposit. **Electrode types. ECoG grids and strips, SEEG depth electrodes, or both (paper Methods). SEEG contacts in white matter, outside the brain or noisy were excluded from all recordings (paper Methods). Localisation method. Electrodes manually selected on post-implant CT, mapped to the pre-implant MRI with iEEG-recon; atlas assignment (AAL, Lausanne, DKT) per patient (paper Methods “Electrode localization”). The deposit holds, for 15 subjects, iEEG-recon outputs: electrode names, CT voxel, T1 (T00mri) mm/vox and MNI152NLin2009cAsym mm/vox coordinates, and for 12 subjects module3 atlas assignments (DKT antspynet, Atropos tissue, piriform ASHS).

What this BIDS version contains and what was changed

  • Raw data: the authors’ BIDS tree (deposit files/sourcedata/) is the root: 30 subjects; ses-research3T and ses-research7T (anat, func resting state), ses-clinical01 (iEEG, 15 subjects). Task and run labels of the iEEG epochs (task-ictal<n> / task-interictal<n>, one task label per curated epoch) are kept as released; events.tsv marks each epoch ictal or interictal.

  • iEEG EDF files are the deposit files. Only the EDF header was changed: patient field set to X X X X and the day of the start date set to 01 (year, month and clock time kept); the data records are byte-identical (sha256 checked, sourcedata/pennsieve-662-v1/edf_scrub.tsv).

  • The EDFs are not filtered: a 60 Hz line peak is present and power above 120 Hz is kept (spectral check of 40 files), so they are labelled raw (SoftwareFilters n/a). The notch and 1-120 Hz band-pass in the paper are its analysis steps.

  • _ieeg.json: PowerLineFrequency 60 (paper Methods: 60 Hz notch; USA). _channels.tsv units follow the EDF physical dimension of each channel.

  • Electrode positions: *_space-MNI152NLin2009cAsym_electrodes.tsv (15 subjects) from the deposit’s iEEG-recon outputs (MNI152NLin2009cAsym mm, plus native pre-implant MRI (T00mri) mm columns, DKT/Atropos labels for 12 subjects). region_coarse and atlas_label_aal3_coarse are added by the NEMAR conversion and are derived values.

  • Anatomical MRI: only the images released as defaced are in the BIDS tree; defaced is part of the acq label (acq-3D_defaced became acq-3Ddefaced, images without an acq label got acq-defaced). The MP2RAGE inversion images (inv-1/inv-2) are in sourcedata/pennsieve-662-v1/files/sourcedata/ with their original names, because their sidecars lack RepetitionTimeExcitation/RepetitionTimePreparation, which BIDS requires and which we do not invent. Head images were checked by eye (sagittal projections) before release. Heudiconv placeholders were removed from sidecars (TaskName TODO... -> rest, CogAtlasID TODO dropped).

  • Derivatives from the deposit: derivatives/fmriprep3T, fmriprep7T, freesurfer3T, freesurfer7T, ieeg_recon (same files, regrouped by pipeline).

  • Everything else of the deposit is in sourcedata/pennsieve-662-v1/ with its original relative path (deposit README, manifest, changelog, metadata spreadsheets, clinical CT / pre-implant MRI in iEEG-recon space, electrode voxel files).

  • Calendar dates in text files are set to day 01 (sourcedata/pennsieve-662-v1/date_scrub.tsv).

  • Not redistributed here (1249 files; list in sourcedata/pennsieve-662-v1/withheld.tsv): 556 x fMRIPrep report figures/html (slices of the non-defaced T1w); 220 x FreeSurfer whole-head volume computed from non-defaced input (face visible, eye check); 216 x 7T B0 field maps released without defacing (magnitude images show the face); 119 x fMRIPrep preprocessed T1w (whole head; face visible for at least one subject, eye check); 45 x iEEG-recon head image: defacing incomplete for some subjects (eye check); 30 x iEEG-recon registration figure (slices of head MRI/CT); 18 x MP2RAGE UNIT1 image: removal of the face could not be confirmed on the eye check (noise-filled background); inversion images of the same scan are kept; 15 x clinical pre-implant T1w (T00mri): iEEG-recon copies of this scan still show the face for some subjects (eye check); 15 x clinical post-implant CT: facial skeleton/soft tissue not removed (eye check); 9 x FreeSurfer backup transforms (processing timestamp in the file name); 5 x anatomical image not released as defaced; 1 x spreadsheet has date/identifier-like cells or date formats (privacy; participants.tsv carries the clinical columns). They remain available from the original Pennsieve deposit.

  • Missing values are n/a; nothing was added that the paper or the deposit does not state.

NEMAR Metadata#

[![DOI](https://img.shields.io/badge/DOI-10.82901%2Fnemar.nm000454-blue)](https://doi.org/10.82901/nemar.nm000454) # A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla MRI and intracranial EEG Redistribution on NEMAR of Pennsieve Discover dataset 662, version 1 (doi:10.26275/aosz-hj9u, licence CC-BY-4.0), described in Sci Data 2026, doi:10.1038/s41597-026-07540-5. Authors: Alfredo Lucas, Mariam Josyula, Nina Petillo, Joost Wagenaar, Joel Stein, Sandhitsu Das, Nishant Sinha, Kathryn Davis, Tamjid Imtiaz. ## Description from the deposit This dataset comprises multimodal paired 3T and 7T MRI scans collected from 30 patients with drug-resistant focal epilepsy along with intracranial EEG data. ## Details (from the paper and the deposit) # c-penn3t7t: Penn 3T-7T Paired Epilepsy Imaging (Pennsieve 662 v1, doi:10.26275/aosz-hj9u) Task / paradigm. 30 patients with drug-resistant focal epilepsy with paired 3T and 7T MRI (structural, resting fMRI); for the 15 who later had intracranial EEG, curated ictal and interictal epochs. Ictal epochs from EMU epileptologist notes; interictal epochs randomly selected while awake, >72 h after implantation, >=2 h before seizures and >=2/6/12 h after subclinical/focal/focal-to-bilateral seizures (doi:10.1038/s41597-026-07540-5 (Sci Data 2026, “A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla MRI and intracranial EEG”, PMC13582879), Methods “Intracranial electroencephalography”). The deposit has 1261 EDF epochs under sourcedata/sub-/ses-clinical01/ieeg with BIDS channels/events/ieeg.json sidecars. **Recording system.* Sampling rates 500, 512 or 1024 Hz (paper Methods). Preprocessing described in the paper: 60 Hz notch (Q=30) and 3rd-order Butterworth 1-120 Hz, forward-backward (paper Methods). Amplifier not stated. Reference scheme. Not stated in the paper text extracted (n/a); see the _ieeg.json iEEGReference of the deposit. **Electrode types.* ECoG grids and strips, SEEG depth electrodes, or both (paper Methods). SEEG contacts in white matter, outside the brain or noisy were excluded from all recordings (paper Methods). Localisation method. Electrodes manually selected on post-implant CT, mapped to the pre-implant MRI with iEEG-recon; atlas assignment (AAL, Lausanne, DKT) per patient (paper Methods “Electrode localization”). The deposit holds, for 15 subjects, iEEG-recon outputs: electrode names, CT voxel, T1 (T00mri) mm/vox and MNI152NLin2009cAsym mm/vox coordinates, and for 12 subjects module3 atlas assignments (DKT antspynet, Atropos tissue, piriform ASHS). ## What this BIDS version contains and what was changed - Raw data: the authors’ BIDS tree (deposit files/sourcedata/) is the root: 30 subjects; ses-research3T and

ses-research7T (anat, func resting state), ses-clinical01 (iEEG, 15 subjects). Task and run labels of the iEEG epochs (task-ictal<n> / task-interictal<n>, one task label per curated epoch) are kept as released; events.tsv marks each epoch ictal or interictal.

  • iEEG EDF files are the deposit files. Only the EDF header was changed: patient field set to X X X X and the day of the start date set to 01 (year, month and clock time kept); the data records are byte-identical (sha256 checked, sourcedata/pennsieve-662-v1/edf_scrub.tsv).

  • The EDFs are not filtered: a 60 Hz line peak is present and power above 120 Hz is kept (spectral check of 40 files), so they are labelled raw (SoftwareFilters n/a). The notch and 1-120 Hz band-pass in the paper are its analysis steps.

  • _ieeg.json: PowerLineFrequency 60 (paper Methods: 60 Hz notch; USA). _channels.tsv units follow the EDF physical dimension of each channel.

  • Electrode positions: *_space-MNI152NLin2009cAsym_electrodes.tsv (15 subjects) from the deposit’s iEEG-recon outputs (MNI152NLin2009cAsym mm, plus native pre-implant MRI (T00mri) mm columns, DKT/Atropos labels for 12 subjects). region_coarse and atlas_label_aal3_coarse are added by the NEMAR conversion and are derived values.

  • Anatomical MRI: only the images released as defaced are in the BIDS tree; defaced is part of the acq label (acq-3D_defaced became acq-3Ddefaced, images without an acq label got acq-defaced). The MP2RAGE inversion images (inv-1/inv-2) are in sourcedata/pennsieve-662-v1/files/sourcedata/ with their original names, because their sidecars lack RepetitionTimeExcitation/RepetitionTimePreparation, which BIDS requires and which we do not invent. Head images were checked by eye (sagittal projections) before release. Heudiconv placeholders were removed from sidecars (TaskName TODO… -> rest, CogAtlasID TODO dropped).

  • Derivatives from the deposit: derivatives/fmriprep3T, fmriprep7T, freesurfer3T, freesurfer7T, ieeg_recon (same files, regrouped by pipeline).

  • Everything else of the deposit is in sourcedata/pennsieve-662-v1/ with its original relative path (deposit README, manifest, changelog, metadata spreadsheets, clinical CT / pre-implant MRI in iEEG-recon space, electrode voxel files).

  • Calendar dates in text files are set to day 01 (sourcedata/pennsieve-662-v1/date_scrub.tsv).

  • Not redistributed here (1249 files; list in sourcedata/pennsieve-662-v1/withheld.tsv): 556 x fMRIPrep report figures/html (slices of the non-defaced T1w); 220 x FreeSurfer whole-head volume computed from non-defaced input (face visible, eye check); 216 x 7T B0 field maps released without defacing (magnitude images show the face); 119 x fMRIPrep preprocessed T1w (whole head; face visible for at least one subject, eye check); 45 x iEEG-recon head image: defacing incomplete for some subjects (eye check); 30 x iEEG-recon registration figure (slices of head MRI/CT); 18 x MP2RAGE UNIT1 image: removal of the face could not be confirmed on the eye check (noise-filled background); inversion images of the same scan are kept; 15 x clinical pre-implant T1w (T00mri): iEEG-recon copies of this scan still show the face for some subjects (eye check); 15 x clinical post-implant CT: facial skeleton/soft tissue not removed (eye check); 9 x FreeSurfer backup transforms (processing timestamp in the file name); 5 x anatomical image not released as defaced; 1 x spreadsheet has date/identifier-like cells or date formats (privacy; participants.tsv carries the clinical columns). They remain available from the original Pennsieve deposit.

  • Missing values are n/a; nothing was added that the paper or the deposit does not state.

License: CC-BY-4.0

Authors:

  • Alfredo Lucas

  • Mariam Josyula

  • Nina Petillo

  • Joost Wagenaar

  • Joel Stein

  • … and 4 more

Versions:

Version

DOI

Released

current

10.82901/nemar.nm000454
§ 03Cohort · Participants

Cohort#

Dataset Statistics#

Sex composition

30
subjects
Female
18
Male
12
F : M ratio
1.50 : 1
60% female · n = 30 subjects with reported sex.
HandednessRight · 26Left · 4

Channel counts (ch)

80101102109118120122125128136140149161165172190

Sampling frequencies (Hz)

25650051210242048

Total recording duration: 27 h

§ 04Signal · Electrodes & trace

Signal · Electrodes & live trace#

Fig. 01 Signal & montage 136 (246), 80 (147), 125 (112), 101 (109), 120 (98), 109 (85), 118 (84), 102 (75), 172 (71), 149 (60), 128 (51), 165 (45), 161 (45), 122 (20), 190 (8), 140 (5) ch · iEEG · 256, 500, 512, 1024, 2048 Hz · 15 subjects, 1261 recordings
Live trace viewer — sub-PID02 · ses-clinical01 · task-interictal415326 · run-01

Showing one representative recording out of 15 subjects and 1261 recordings in this dataset. Browse the full set on OpenNeuro; drop any other _ieeg.{set,edf,bdf,vhdr} file onto the viewer (or pass ?ieeg=<url>) to inspect it.

Electrode layout — iEEG · 136 sensors — 136 channels

NEMAR Processing Statistics#

The plots below are generated by NEMAR’s automated EEG pipeline. The histogram shows pipeline success for data cleaning and ICA decomposition, the percentage of data frames and EEG channels retained after artefact removal, line noise per channel (RMS, dB), and the age/gender distribution of participants.

NEMAR pipeline statistics — NM000454
HED event descriptors word cloud HED event descriptors word cloud — NM000454
§ 05Manifest · BIDS tree

Manifest#

File Explorer#

Browse the BIDS file structure of this dataset. Records are fetched on demand from the EEGDash catalog the first time you open the explorer.

Recordings—
Files—
Subjects—
Modalities—
Click to load file structure…
Full dataset metadata table

Dataset ID

NM000454

Title

A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla MRI and intracranial EEG (Penn 3T-7T, Pennsieve 662)

Author (year)

—

Canonical

—

Importable as

NM000454

Year

20

Authors

Alfredo Lucas, Mariam Josyula, Nina Petillo, Joost Wagenaar, Joel Stein, Sandhitsu Das, Nishant Sinha, Kathryn Davis, Tamjid Imtiaz

License

CC-BY-4.0

Citation / DOI

10.82901/nemar.nm000454

Source links

OpenNeuro | NeMAR | Source URL
Copy-paste BibTeX
@dataset{nm000454,
  title = {A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla MRI and intracranial EEG (Penn 3T-7T, Pennsieve 662)},
  author = {Alfredo Lucas and Mariam Josyula and Nina Petillo and Joost Wagenaar and Joel Stein and Sandhitsu Das and Nishant Sinha and Kathryn Davis and Tamjid Imtiaz},
  doi = {10.82901/nemar.nm000454},
  url = {https://doi.org/10.82901/nemar.nm000454},
}
§ 06API · Programmatic access

API Reference#

Signature
eegdash.dataset
class
eegdash.dataset.NM000454(cache_dir, query=None, s3_bucket=None, **kwargs)
Bases: EEGDashDataset
Author (year)—
Canonical—
Importable asNM000454
Sourceeegdash/dataset/registry.py · [source ↗]
class eegdash.dataset.NM000454(cache_dir: str, query: dict | None = None, s3_bucket: str | None = None, **kwargs)[source]#

A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla MRI and intracranial EEG (Penn 3T-7T, Pennsieve 662)

Study:

nm000454 (NeMAR)

Author (year):

—

Canonical:

—

Also importable as: NM000454.

Modality: ieeg; Subject type: Unknown. Subjects: 15; recordings: 1261; tasks: 1255.

Parameters:

  • cache_dir (str | Path) – Directory where data are cached locally.

  • query (dict | None) – Additional MongoDB-style filters to AND with the dataset selection. Must not contain the key dataset.

  • s3_bucket (str | None) – Base S3 bucket used to locate the data.

  • **kwargs (dict) – Additional keyword arguments forwarded to EEGDashDataset.

data_dir#

Local dataset cache directory (cache_dir / dataset_id).

Type:

Path

query#

Merged query with the dataset filter applied.

Type:

dict

records#

Metadata records used to build the dataset, if pre-fetched.

Type:

list[dict] | None

Notes

Each item is a recording; recording-level metadata are available via dataset.description. query supports MongoDB-style filters on fields in ALLOWED_QUERY_FIELDS and is combined with the dataset filter. Dataset-specific caveats are not provided in the summary metadata.

References

OpenNeuro dataset: https://openneuro.org/datasets/nm000454 NeMAR dataset: https://nemar.org/dataexplorer/detail?dataset_id=nm000454 DOI: https://doi.org/10.82901/nemar.nm000454

Examples

>>> from eegdash.dataset import NM000454
>>> dataset = NM000454(cache_dir="./data")
>>> recording = dataset[0]
>>> raw = recording.load()
__init__(cache_dir: str, query: dict | None = None, s3_bucket: str | None = None, **kwargs)[source]#
save(path: str, overwrite: bool = False, offset: int = 0)[source]#

Save datasets to files by creating one subdirectory for each dataset:

path/
    0/
        0-raw.fif | 0-epo.fif
        description.json
        raw_preproc_kwargs.json (if raws were preprocessed)
        window_kwargs.json (if this is a windowed dataset)
        window_preproc_kwargs.json  (if windows were preprocessed)
        target_name.json (if target_name is not None and dataset is raw)
    1/
        1-raw.fif | 1-epo.fif
        description.json
        raw_preproc_kwargs.json (if raws were preprocessed)
        window_kwargs.json (if this is a windowed dataset)
        window_preproc_kwargs.json  (if windows were preprocessed)
        target_name.json (if target_name is not None and dataset is raw)
Parameters:

  • path (str) –

    Directory in which subdirectories are created to store

    -raw.fif | -epo.fif and .json files to.

  • overwrite (bool) – Whether to delete old subdirectories that will be saved to in this call.

  • offset (int) – If provided, the integer is added to the id of the dataset in the concat. This is useful in the setting of very large datasets, where one dataset has to be processed and saved at a time to account for its original position.

Access modesMNE → braindecode → PyTorch → ML
.rawMNE Raw object — standard tools (filter, epoch, ICA, plot_psd).mne
BaseConcatDatasetEach record is a lazy BaseDataset from braindecode — windowed via create_windows_from_events.braindecode
DataLoaderWraps the windowed dataset into a PyTorch DataLoader; supports parallel workers and on-the-fly augmentations.pytorch
Zarr cacheOptional braindecode Zarr mirror for fast resume; persisted to cache_dir.zarr
Hugging FaceNo per-dataset mirror published yet — browse the EEGDash org listing for sibling datasets. See the datasets loader API.huggingface
Croissant 1.0Machine-readable JSON-LD descriptor — NM000454.croissant.json (MLCommons schema, ingestible by PyTorch / TensorFlow / JAX).mlcommons
Examples using EEGDashcurated · start here
Find datasets with the EEGDash APIQuery the catalogue, filter by task or modality, list candidates.Load one EEG recordingResolve a single record to an MNE Raw with channels and events.EEG recording to PyTorch DataLoaderWrap braindecode windows in a DataLoader for model training.Preprocess EEG and create windowsFilter, resample, epoch — and persist the windowed dataset.Save and reload prepared dataCache a windowed dataset to disk and reattach it without recompute.Download a dataset locallyPrefetch BIDS files to a local cache and validate the layout.

Swap any load_dataset(...) call for nm000454 to reproduce the tutorial on this dataset.

Citation

Alfredo Lucas, Mariam Josyula, Nina Petillo, Joost Wagenaar, Joel Stein, … (20). A multimodal epilepsy dataset of paired 3-Tesla and 7-Tesla MRI and intracranial EEG (Penn 3T-7T, Pennsieve 662). 10.82901/nemar.nm000454

Provenance

¹Contributed to nemar in BIDS format.

²Curated & ingested by the EEGDash catalog; see CITATION.cff for canonical reference.

³Persistent identifier: 10.82901/nemar.nm000454.

BIDS
BIDS 1.10.0
Sidecars
events · channels · eeg.json
Provenance
CC-BY-4.0 · 10.82901/nemar.nm000454
Machine-readable
schema.org/Dataset · Croissant
Mirrors
OpenNeuro · NEMAR · HF org

See Also#