EEGdash›NeMAR›NM000253

Iss. 253 · 10 subjects · 26 recordings · CC BY 4.0

Dataset Brief · Wang et al. 2024 — Brain Treebank

NM000253: ieeg dataset, 10 subjects#

Wang et al. 2024 — Brain Treebank: Large-scale intracranial recordings from naturalistic language stimuli

Access recordings and metadata through EEGDash.

Citation: Christopher Wang, Adam Yaari, Aaditya K Singh, Vighnesh Subramaniam, Dana Rosenfarb, Jan DeWitt, Pranav Misra, Joseph R Madsen, Scellig Stone, Gabriel Kreiman, Boris Katz, Ignacio Cases, Andrei Barbu (2024). Wang et al. 2024 — Brain Treebank: Large-scale intracranial recordings from naturalistic language stimuli. 10.82901/nemar.nm000253

Modality: ieeg Subjects: 10 Recordings: 26 License: CC BY 4.0 Source: nemar

Metadata: Complete (100%)

10-participant iEEG dataset — Wang et al. 2024 — Brain Treebank: Large-scale intracranial recordings from naturalistic language stimuli.

Data & curation Christopher Wang · Adam Yaari · Aaditya K Singh · Vighnesh Subramaniam · Dana Rosenfarb · Jan DeWitt · …
Senior author Andrei Barbu
Year 2024 · Distributed via NeMAR

iEEG · 164 (8), 136 (3), 190 (3), 166 (3), 156 (3), 218 (2), 248 (2), 108, 158 ch2048 HzBIDS 1.9.0Task · movie

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#

Get Started

Install

pip install eegdash

Access the data

from eegdash.dataset import NM000253

dataset = NM000253(cache_dir="./data")
# Get the raw object of the first recording
raw = dataset.datasets[0].raw
print(raw.info)

Query & Filter

Filter by subject

dataset = NM000253(cache_dir="./data", subject="01")

Advanced query

dataset = NM000253(
    cache_dir="./data",
    query={"subject": {"$in": ["01", "02"]}},
)

Iterate recordings

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

Cite This Dataset

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

BibTeX

@dataset{nm000253,
  title = {Wang et al. 2024 — Brain Treebank: Large-scale intracranial recordings from naturalistic language stimuli},
  author = {Christopher Wang and Adam Yaari and Aaditya K Singh and Vighnesh Subramaniam and Dana Rosenfarb and Jan DeWitt and Pranav Misra and Joseph R Madsen and Scellig Stone and Gabriel Kreiman and Boris Katz and Ignacio Cases and Andrei Barbu},
  doi = {10.82901/nemar.nm000253},
  url = {https://doi.org/10.82901/nemar.nm000253},
}

§ 02Study · The README

About This Dataset#

The Brain Treebank is a large-scale dataset of intracranial electrophysiological (iEEG /

stereo-EEG) recordings collected from 10 epilepsy patients (at Boston Children’s Hospital) while they watched Hollywood movies. Recordings were acquired at 2048 Hz from on average 168 electrodes per subject (1,688 electrodes total), totalling roughly 43 hours across 26 trials (one trial = one movie). The audio of each movie was transcribed and word onsets manually annotated, and each transcript was parsed into Universal Dependencies (UD) syntax trees — making this one of the largest datasets of intracranial recordings grounded in naturalistic language.

This NEMAR record provides the dataset converted to BIDS-iEEG (BrainVision) format:

each trial is one BIDS run (task-movie, run-01 …), with signals in sub-*/ieeg/.

DOI

Brain Treebank: large-scale intracranial (iEEG) recordings from naturalistic language stimuli

Summary

Modality and paradigm

Modality: Intracranial EEG / stereo-EEG (iEEG-BIDS), 2048 Hz, BrainVision

NEMAR Metadata#

[![DOI](https://img.shields.io/badge/DOI-10.82901%2Fnemar.nm000253-blue)](https://doi.org/10.82901/nemar.nm000253) # Brain Treebank: large-scale intracranial (iEEG) recordings from naturalistic language stimuli ## Summary The Brain Treebank is a large-scale dataset of intracranial electrophysiological (iEEG / stereo-EEG) recordings collected from 10 epilepsy patients (at Boston Children’s Hospital) while they watched Hollywood movies. Recordings were acquired at 2048 Hz from on average 168 electrodes per subject (1,688 electrodes total), totalling roughly 43 hours across 26 trials (one trial = one movie). The audio of each movie was transcribed and word onsets manually annotated, and each transcript was parsed into Universal Dependencies (UD) syntax trees — making this one of the largest datasets of intracranial recordings grounded in naturalistic language. This NEMAR record provides the dataset converted to BIDS-iEEG (BrainVision) format: each trial is one BIDS run (task-movie, run-01 …), with signals in sub-*/ieeg/. ## Modality and paradigm - Modality: Intracranial EEG / stereo-EEG (iEEG-BIDS), 2048 Hz, BrainVision

(IEEE_FLOAT_32, multiplexed)

Task / paradigm: Passive naturalistic viewing of Hollywood movies (task-movie), with time-aligned word-level language annotations (transcripts and UD syntax trees in code/transcripts.zip and code/trees.zip)
Population: 10 epilepsy patients undergoing intracranial monitoring

## Participants and data structure - 10 subjects (sub-01 … sub-10), 26 movie-viewing runs in total. - Electrode labels and per-subject electrode information are in each sub-*/ieeg/

(electrodes.tsv, coordsystem.json); corrupted electrodes are flagged as bad in channels.tsv (from the Brain Treebank corrupted_elec.json).

Electrode coordinates are provided as-is from the Brain Treebank localization.zip (code/localization.zip); see each coordsystem.json for important caveats about units and the absence of a published transform to a standard template.

## Original dataset / data paper Please cite the original publication when using this dataset: > Wang, C., Yaari, A. U., Singh, A. K., Subramaniam, V., Rosenfarb, D., DeWitt, J., > Misra, P., Madsen, J. R., Stone, S., Kreiman, G., Katz, B., Cases, I., & Barbu, A. > (2024). Brain Treebank: Large-scale intracranial recordings from naturalistic language > stimuli. Advances in Neural Information Processing Systems 37 (NeurIPS 2024, Datasets > and Benchmarks Track). arXiv:2411.08343. - Preprint / DOI: [arXiv:2411.08343](https://doi.org/10.48550/arXiv.2411.08343) - NeurIPS 2024 proceedings: https://proceedings.neurips.cc/paper_files/paper/2024/hash/aefa2385b3f33abf1526ae4e2c208cd9-Abstract-Datasets_and_Benchmarks_Track.html - Project site / source data: https://braintreebank.dev/ - Original code release: czlwang/brain_treebank_code_release - Ethics: Boston Children’s Hospital / Harvard IRB; all subjects gave informed consent. ## BIDS conversion Per-trial HDF5 recordings from braintreebank.dev were converted to BIDS-iEEG (BrainVision) with the EEGDash conversion script in code/convert_braintreebank.py. Each trial (one movie) becomes one BIDS run. EEG-BIDS / MNE-BIDS were used only for standardisation; the data themselves are from the original Brain Treebank release. Please credit the original creators (Wang et al.) and cite the paper above. ## License CC BY 4.0 (see dataset_description.json).

License: CC BY 4.0

Authors:

Christopher Wang
Adam Yaari
Aaditya K Singh
Vighnesh Subramaniam
Dana Rosenfarb
… and 8 more

Versions:

Version	DOI	Released
`current`	10.82901/nemar.nm000253

§ 03Cohort · Participants

Cohort#

Dataset Statistics#

Channel counts (ch)

108136156158164166190218248

Sampling frequencies: 2048.0 Hz (n=26 recordings)

Total recording duration: 1 h 48 min

§ 04Signal · Electrodes & trace

Signal · Electrodes & live trace#

Fig. 01 Signal & montage 164 (8), 136 (3), 190 (3), 166 (3), 156 (3), 218 (2), 248 (2), 108, 158 ch · iEEG · 2048 Hz · 10 subjects, 26 recordings

Live trace viewer — sub-08 · task-movie · run-01

Showing one representative recording out of 10 subjects and 26 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 · 128 sensors — 128 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.

HED event descriptors word cloud

§ 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…

§ 06API · Programmatic access

API Reference#

Signature

eegdash.dataset

class

eegdash.dataset.NM000253(cache_dir, query=None, s3_bucket=None, **kwargs)

Bases: EEGDashDataset

Author (year)Wang2024_et_al_Brain

Canonical—

Importable asNM000253 · Wang2024_et_al_Brain

Sourceeegdash/dataset/registry.py · [source ↗]

class eegdash.dataset.NM000253(cache_dir: str, query: dict | None = None, s3_bucket: str | None = None, **kwargs)[source]#

Wang et al. 2024 — Brain Treebank: Large-scale intracranial recordings from naturalistic language stimuli

Study:: nm000253 (NeMAR)
Author (year):: Wang2024_et_al_Brain
Canonical:: —

Also importable as: NM000253, Wang2024_et_al_Brain.

Modality: ieeg; Subject type: Unknown. Subjects: 10; recordings: 26; tasks: 1.

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/nm000253 NeMAR dataset: https://nemar.org/dataexplorer/detail?dataset_id=nm000253 DOI: https://doi.org/10.82901/nemar.nm000253

Examples

>>> from eegdash.dataset import NM000253
>>> dataset = NM000253(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 FacePre-bundled mirror at EEGDash/nm000253 · pull with datasets.load_dataset("EEGDash/nm000253").huggingface

Croissant 1.0Machine-readable JSON-LD descriptor — NM000253.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 nm000253 to reproduce the tutorial on this dataset.

Citation

Christopher Wang, Adam Yaari, Aaditya K Singh, Vighnesh Subramaniam, Dana Rosenfarb, … (2024). Wang et al. 2024 — Brain Treebank: Large-scale intracranial recordings from naturalistic language stimuli. 10.82901/nemar.nm000253

Provenance

¹Contributed to nemar in BIDS format.

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

³Persistent identifier: 10.82901/nemar.nm000253.

Related & sibling datasets

ON004696IEEG · 8 subj ON004127IEEG · 8 subj ON005448IEEG · 13 subj ON004370IEEG · 7 subj ON003848IEEG · 6 subj

+ 1 more — see See Also below →

BIDS

BIDS 1.9.0

Sidecars

channels · electrodes · coordsystem

Provenance

CC BY 4.0 · 10.82901/nemar.nm000253

Machine-readable

schema.org/Dataset · Croissant

Mirrors

OpenNeuro · NEMAR · HuggingFace

Dataset ID	`NM000253`
Title	Wang et al. 2024 — Brain Treebank: Large-scale intracranial recordings from naturalistic language stimuli
Author (year)	`Wang2024_et_al_Brain`
Canonical	—
Importable as	`NM000253`, `Wang2024_et_al_Brain`
Year	2024
Authors	Christopher Wang, Adam Yaari, Aaditya K Singh, Vighnesh Subramaniam, Dana Rosenfarb, Jan DeWitt, Pranav Misra, Joseph R Madsen, Scellig Stone, Gabriel Kreiman, Boris Katz, Ignacio Cases, Andrei Barbu
License	CC BY 4.0
Citation / DOI	10.82901/nemar.nm000253
Source links	OpenNeuro \| NeMAR \| Source URL

NM000253: ieeg dataset, 10 subjects#

Quickstart#

About This Dataset#

NEMAR Metadata#

Cohort#

Dataset Statistics#

Signal · Electrodes & live trace#

NEMAR Processing Statistics#

Manifest#

File Explorer#

API Reference#

Citation

Provenance

Related & sibling datasets

See Also#