EEGdash›OpenNeuro›DS004120
Iss. 4120 · 109 subjects · 131 recordings · CC0
Dataset Brief · BCIT Baseline Driving

DS004120: eeg dataset, 109 subjects#

BCIT Baseline Driving

Citation: Jonathan Touryan (data and curation), Greg Apker (data), Brent Lance (data), Scott Kerick (data), Anthony Ries (data), Kaleb McDowell (data), Tony Johnson (curation), Kay Robbins (curation) (19). BCIT Baseline Driving. 10.18112/openneuro.ds004120.v1.0.0

109-participant EEG dataset — BCIT Baseline Driving.

Data & curation Jonathan Touryan (data and curation) · Greg Apker (data) · Brent Lance (data) · Scott Kerick (data) · Anthony Ries (data) · Kaleb McDowell (data) · …
Year 19 · Distributed via OpenNeuro
Funding This research was sponsored by the Army Research Laboratory and was accomplished under Cooperative Agreement Number W911NF-10-0-0002.
EEG · 266 (81), 74 (50) ch1024 Hz · mixedBIDS 1.7.0HED ✓Task · DriveWithSpeedChange3 sessionsHealthyVisualAttention
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 DS004120

dataset = DS004120(cache_dir="./data")
# Get the raw object of the first recording
raw = dataset.datasets[0].raw
print(raw.info)

Filter by subject

dataset = DS004120(cache_dir="./data", subject="01")

Advanced query

dataset = DS004120(
    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{ds004120,
  title = {BCIT Baseline Driving},
  author = {Jonathan Touryan (data and curation) and Greg Apker (data) and Brent Lance (data) and Scott Kerick (data) and Anthony Ries (data) and Kaleb McDowell (data) and Tony Johnson (curation) and Kay Robbins (curation)},
  doi = {10.18112/openneuro.ds004120.v1.0.0},
  url = {https://doi.org/10.18112/openneuro.ds004120.v1.0.0},
}
§ 02Study · The README

About This Dataset#

Overview: The Baseline Driving study was designed to collect extended time-on-task measurements of

subjects performing a driving task in a simulated environment in order to assess fatigue-based performance through novel biomarkers. The Baseline Driving study was intended to identify periods of driver fatigue via predictive algorithms formulated from the analysis of driver EEG data, in comparison to the objective performance measures, and in contrast with the (non-fatigued) Calibration driving session for the subject.

Baseline driving data sets were designed to be the second component of every recording session within the

BCIT program, which featured multiple studies investigating fatigue.

BCIT Baseline Driving

Introduction

Collectively, the Baseline Driving recordings comprise a virtual study, in which long time-on-task driving performance can be analyzed for fatigue-related EEG biomarkers based on measured driving performance degradation. Further information is available on request from cancta.net.

The task was performed using identical systems at three different sites:

View full README

BCIT Baseline Driving

Introduction

Collectively, the Baseline Driving recordings comprise a virtual study, in which long time-on-task driving performance can be analyzed for fatigue-related EEG biomarkers based on measured driving performance degradation. Further information is available on request from cancta.net.

The task was performed using identical systems at three different sites: - Army Research Laboratory, Aberdeen MD (T1) - Teledyne Corporation, Durham, NC (T2) - Science Applications International Corporation (SAIC), Louisville, CO (T3)

All sites used identical driving simulator setups.

The data collected at site T1 used a 64-channel Biosemi EEG headset as did the data collected at site T2, while site T3 used a 256-channel Biosemi EEG headset.

Data from site T1 has legacy subject IDs in the range 1000 to 1999. Data from site T2 has legacy subject IDs in the range 2000 to 2999.

Data from site T3 has legacy subject IDs in the range 3000 to 3999. Legacy subject IDs are unique across the entire BCIT program.

Methods

Subjects: Subjects at Aberdeen Proving Grounds were recruited, on a voluntary basis from among the scientists and engineers working at APG.

Subjects recruited by Teledyne and SAIC were found via advertising and community outreach efforts, and primarily consisted of local college students. Apparatus: Driving simulator with steering wheel and brake / foot pedals (Real Time Technologies; Dearborn, MI); Video Refresh Rate (VRR) = 900 Hz; Vehicle data log file Sampling Rate (SR) = 100 Hz); EEG (BioSemi 256 (+8) channel systems with 4 eye and 2 mastoid channels recorded; SR=1024 Hz); Eye Tracking (Sensomotoric Instruments (SMI); REDEYE250). Eye tracking data is not included in this dataset. Initial setup: Upon arrival to the lab, subjects were given an introduction to the primary study for which they were recruited and provided informed consent and provided demographics information.

This was followed by a practice session, to acclimate the subject to the driving simulator. The driving practice task lasted 10-15 min, until asymptotic performance in steering and speed control was demonstrated and lack of motion sickness was reported. Subjects were then outfitted and prepped for eye tracking and EEG acquisition. Task organization: Subjects always began recording sessions by performing a Calibration Driving task, which was a 15-minute drive where the subject controlled only the steering (and speed was controlled by the simulator).

Following this, subjects would perform the Baseline Driving task and the Guard Duty task, with counter-balancing used across subjects as to which of them came first.

The Baseline Driving run was 60 minutes of driving, performed in 6 blocks of 10 minutes each, with subjects responsible for speed and steering control.

The subject was instructed to stay within the boundaries of the right-most lane, and to drive at the posted speed limits. The vehicle was periodically subject to lateral perturbing forces, which could be applied to either side of the vehicle, pushing the vehicle out of the center of the lane; and the subject was instructed to execute corrective steering actions to return the vehicle to the center of the lane. Independent variables: For T1 (ARL) and T3 (SAIC) there were no independent variables.

For T2 data sets (Teledyne), independent variables were Visual Complexity (high vs. low), Perturbation Frequency (high vs. low). Dependent variables: Reaction times to perturbations, continuous performance based on vehicle log (steering wheel angle, lane position, heading error, etc.), Task-Induced Fatigue Scale (TIFS), Karolinska Sleepiness Scale (KSS), Visual Analog Scale of Fatigue (VAS-F).

Note: questionnaire data is available upon request from cancta.net. Additional data acquired: Participant Enrollment Questionnaire, Subject Questionnaire for Current Session, Simulator Sickness Questionnaire. Experimental Locations: Army Research Laboratory, Aberdeen MD (site T1); Teledyne Corporation, Durham, NC (site T2); Science Applications International Corporation (SAIC), Louisville, CO (site T3). Note 1: This dataset has a corresponding dataset in the BCIT Calibration Driving ds004118 which has the 15 minute driving task performed prior to this one. Note 2: Some of the subjects in this dataset performed either the BCIT Basic Guard Duty Task (ds004118) or the BCIT Advanced Guard Duty Task (ds004106) counterbalanced during the same session.

§ 03Cohort · Participants

Cohort#

Dataset Statistics#

Channel counts (ch)

74266

Sampling frequencies (Hz)

10242048
§ 04Signal · Electrodes & trace

Signal · Electrodes & live trace#

Fig. 01 Signal & montage 266 (81), 74 (50) ch · EEG · 1024 Hz · mixed · 109 subjects, 131 recordings
Live trace viewer — sub-13 · ses-01 · task-DriveWithSpeedChange · run-1

Showing one representative recording out of 109 subjects and 131 recordings in this dataset. Browse the full set on OpenNeuro; drop any other _eeg.{set,edf,bdf,vhdr} file onto the viewer (or pass ?eeg=<url>) to inspect it.

Electrode layout — EEG · 256 sensors — 256 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 — DS004120
HED event descriptors word cloud HED event descriptors word cloud — DS004120
§ 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

DS004120

Title

BCIT Baseline Driving

Author (year)

Touryan2022_BCIT_Baseline

Canonical

—

Importable as

DS004120, Touryan2022_BCIT_Baseline

Year

19

Authors

Jonathan Touryan (data and curation), Greg Apker (data), Brent Lance (data), Scott Kerick (data), Anthony Ries (data), Kaleb McDowell (data), Tony Johnson (curation), Kay Robbins (curation)

License

CC0

Citation / DOI

doi:10.18112/openneuro.ds004120.v1.0.0

Source links

OpenNeuro | NeMAR | Source URL
Copy-paste BibTeX
@dataset{ds004120,
  title = {BCIT Baseline Driving},
  author = {Jonathan Touryan (data and curation) and Greg Apker (data) and Brent Lance (data) and Scott Kerick (data) and Anthony Ries (data) and Kaleb McDowell (data) and Tony Johnson (curation) and Kay Robbins (curation)},
  doi = {10.18112/openneuro.ds004120.v1.0.0},
  url = {https://doi.org/10.18112/openneuro.ds004120.v1.0.0},
}
§ 06API · Programmatic access

API Reference#

Signature
eegdash.dataset
class
eegdash.dataset.DS004120(cache_dir, query=None, s3_bucket=None, **kwargs)
Bases: EEGDashDataset
Author (year)Touryan2022_BCIT_Baseline
Canonical—
Importable asDS004120 · Touryan2022_BCIT_Baseline
Sourceeegdash/dataset/registry.py · [source ↗]
class eegdash.dataset.DS004120(cache_dir: str, query: dict | None = None, s3_bucket: str | None = None, **kwargs)[source]#

BCIT Baseline Driving

Study:

ds004120 (OpenNeuro)

Author (year):

Touryan2022_BCIT_Baseline

Canonical:

—

Also importable as: DS004120, Touryan2022_BCIT_Baseline.

Modality: eeg; Experiment type: Attention; Subject type: Healthy. Subjects: 109; recordings: 131; 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/ds004120 NeMAR dataset: https://nemar.org/dataexplorer/detail?dataset_id=ds004120 DOI: https://doi.org/10.18112/openneuro.ds004120.v1.0.0 NEMAR citation count: 0

Examples

>>> from eegdash.dataset import DS004120
>>> dataset = DS004120(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/ds004120 · pull with datasets.load_dataset("EEGDash/ds004120").huggingface
Croissant 1.0Machine-readable JSON-LD descriptor — DS004120.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 ds004120 to reproduce the tutorial on this dataset.

Citation

Jonathan Touryan (data and curation), Greg Apker (data), Brent Lance (data), Scott Kerick (data), Anthony Ries (data), … (19). BCIT Baseline Driving. 10.18112/openneuro.ds004120.v1.0.0

Provenance

¹Contributed to openneuro in BIDS format.

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

³Persistent identifier: 10.18112/openneuro.ds004120.v1.0.0.

BIDS
BIDS 1.7.0
Sidecars
events · channels · electrodes · coordsystem · eeg.json
Machine-readable
schema.org/Dataset · Croissant

See Also#