NM000250: eeg dataset, 87 subjects#

Class for Dreyer2023 dataset management. MI dataset

Access recordings and metadata through EEGDash.

Citation: Pauline Dreyer, Aline Roc, Léa Pillette, Sébastien Rimbert, Fabien Lotte (2021). Class for Dreyer2023 dataset management. MI dataset.

Modality: eeg Subjects: 87 Recordings: 520 License: CC-BY-4.0 Source: nemar

Metadata: Complete (90%)

Quickstart#

Install

pip install eegdash

Access the data

from eegdash.dataset import NM000250

dataset = NM000250(cache_dir="./data")
# Get the raw object of the first recording
raw = dataset.datasets[0].raw
print(raw.info)

Filter by subject

dataset = NM000250(cache_dir="./data", subject="01")

Advanced query

dataset = NM000250(
    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{nm000250,
  title = {Class for Dreyer2023 dataset management. MI dataset},
  author = {Pauline Dreyer and Aline Roc and Léa Pillette and Sébastien Rimbert and Fabien Lotte},
}

About This Dataset#

Class for Dreyer2023 dataset management. MI dataset

Class for Dreyer2023 dataset management. MI dataset.

Dataset Overview

  • Code: Dreyer2023

  • Paradigm: imagery

  • DOI: 10.1038/s41597-023-02445-z

View full README

Class for Dreyer2023 dataset management. MI dataset

Class for Dreyer2023 dataset management. MI dataset.

Dataset Overview

  • Code: Dreyer2023

  • Paradigm: imagery

  • DOI: 10.1038/s41597-023-02445-z

  • Subjects: 87

  • Sessions per subject: 1

  • Events: left_hand=1, right_hand=2

  • Trial interval: [0, 5] s

  • Runs per session: 6

  • Session IDs: calibration, online_training

  • File format: GDF

  • Contributing labs: Inria Bordeaux

Acquisition

  • Sampling rate: 512.0 Hz

  • Number of channels: 27

  • Channel types: eeg=27, emg=2, eog=3

  • Channel names: C1, C2, C3, C4, C5, C6, CP1, CP2, CP3, CP4, CP5, CP6, CPz, Cz, EMGd, EMGg, EOG1, EOG2, EOG3, F3, F4, FC1, FC2, FC3, FC4, FC5, FC6, FCz, Fz, P3, P4, Pz

  • Montage: 10-20

  • Hardware: g.USBAmp (g.tec, Austria)

  • Software: OpenViBE 2.1.0 (Dataset A) / OpenViBE 2.2.0 (Dataset B and C)

  • Reference: left earlobe

  • Ground: FPz

  • Sensor type: active electrodes

  • Line frequency: 50.0 Hz

  • Online filters: none (raw signals recorded without hardware filters)

  • Cap manufacturer: g.tec

  • Auxiliary channels: EOG (3 ch, horizontal, vertical), EMG (2 ch), gsr

Participants

  • Number of subjects: 87

  • Health status: healthy

  • Age: mean=29.0, min=19, max=59

  • Gender distribution: female=41, male=46

  • Handedness: right

  • BCI experience: naive

  • Species: human

Experimental Protocol

  • Paradigm: imagery

  • Number of classes: 2

  • Class labels: left_hand, right_hand

  • Trial duration: 8.0 s

  • Tasks: right_hand_MI, left_hand_MI, resting_state

  • Study design: Graz protocol

  • Feedback type: continuous visual

  • Stimulus type: blue bar varying in length

  • Stimulus modalities: visual, auditory

  • Primary modality: visual

  • Synchronicity: cue-based

  • Mode: online

  • Training/test split: True

  • Instructions: Participants were encouraged to perform kinesthetic imagination and leave them free to choose their mental imagery strategy. Participants were instructed to try to find the best strategy so that the system would show the longest possible feedback bar. Only positive feedback was provided.

HED Event Annotations

Schema: HED 8.4.0 | Browse: https://www.hedtags.org/hed-schema-browser

left_hand

     ├─ Sensory-event, Experimental-stimulus, Visual-presentation
     └─ Agent-action
        └─ Imagine
           ├─ Move
           └─ Left, Hand

right_hand

├─ Sensory-event, Experimental-stimulus, Visual-presentation
└─ Agent-action
   └─ Imagine
      ├─ Move
      └─ Right, Hand

Paradigm-Specific Parameters

  • Detected paradigm: motor_imagery

  • Imagery tasks: right_hand, left_hand

  • Cue duration: 1.25 s

  • Imagery duration: 3.75 s

Data Structure

  • Trials: 240

  • Trials per class: right_hand=120, left_hand=120

  • Blocks per session: 6

  • Block duration: 420.0 s

  • Trials context: per subject (120 per class)

Preprocessing

  • Data state: raw

  • Preprocessing applied: False

  • Bandpass filter: [5.0, 35.0]

  • Filter type: Butterworth

  • Filter order: 5

  • Artifact methods: visual inspection

  • Re-reference: Laplacian (C3, C4 for feature extraction)

  • Notes: The raw signals were recorded without any hardware filters. For online processing, a fifth-order Butterworth filter was applied in a participant-specific discriminant frequency band in the range of 5 Hz to 35 Hz with 0.5 Hz large bins. Impedance could not be measured with active electrodes; EEG signals were visually checked and regularly re-checked to ensure good signal quality.

Signal Processing

  • Classifiers: LDA

  • Feature extraction: CSP, Bandpower

  • Frequency bands: analyzed=[5.0, 35.0] Hz; alpha=[8.0, 13.0] Hz; mu=[8.0, 13.0] Hz; beta=[13.0, 30.0] Hz

  • Spatial filters: CSP, Laplacian

Cross-Validation

  • Method: calibration-feedback

  • Evaluation type: within_session

Performance (Original Study)

  • Accuracy: 63.35%

  • Mean Accuracy Std: 17.36

  • Mean Accuracy R3: 63.14

  • Mean Accuracy R4: 64.82

  • Chance Level Individual: 58.7

  • Chance Level Database: 51.0

BCI Application

  • Applications: rehabilitation, assistive_technology, neurofeedback, user_training

  • Environment: laboratory

  • Online feedback: True

Tags

  • Pathology: Healthy

  • Modality: Motor

  • Type: Motor Imagery

Documentation

  • Description: A large EEG database with users’ profile information for motor imagery brain-computer interface research. Contains electroencephalographic signals from 87 human participants, collected during a single day of brain-computer interface (BCI) experiments, organized into 3 datasets (A, B, and C) that were all recorded using the same protocol: right and left hand motor imagery (MI).

  • DOI: 10.1038/s41597-023-02445-z

  • Associated paper DOI: 10.1038/s41597-023-02445-z

  • License: CC-BY-4.0

  • Investigators: Pauline Dreyer, Aline Roc, Léa Pillette, Sébastien Rimbert, Fabien Lotte

  • Senior author: Fabien Lotte

  • Contact: fabien.lotte@inria.fr

  • Institution: Centre Inria de l’université de Bordeaux

  • Department: LaBRI (Univ. Bordeaux/CNRS/Bordeaux INP)

  • Address: Talence, 33405, France

  • Country: FR

  • Repository: Zenodo

  • Data URL: https://doi.org/10.5281/zenodo.8089820

  • Publication year: 2023

  • Funding: European Research Council (ERC Starting Grant project BrainConquest, grant ERC-2016-STG-714567)

  • Ethics approval: Inria’s ethics committee, the COERLE (Approval number: 2018-13)

  • Keywords: motor imagery, brain-computer interface, EEG, BCI illiteracy, user training, personality profile, cognitive traits, user profile

Abstract

We present and share a large database containing electroencephalographic signals from 87 human participants, collected during a single day of brain-computer interface (BCI) experiments, organized into 3 datasets (A, B, and C) that were all recorded using the same protocol: right and left hand motor imagery (MI). Each session contains 240 trials (120 per class), which represents more than 20,800 trials, or approximately 70 hours of recording time. It includes the performance of the associated BCI users, detailed information about the demographics, personality profile as well as some cognitive traits and the experimental instructions and codes (executed in the open-source platform OpenViBE). Such database could prove useful for various studies, including but not limited to: (1) studying the relationships between BCI users’ profiles and their BCI performances, (2) studying how EEG signals properties varies for different users’ profiles and MI tasks, (3) using the large number of participants to design cross-user BCI machine learning algorithms or (4) incorporating users’ profile information into the design of EEG signal classification algorithms.

Methodology

Participants performed a Graz protocol MI-BCI task with 6 runs (2 calibration runs with sham feedback, 4 online training runs with real feedback). Each run consisted of 40 trials (20 per MI-task) with 8s trial duration. Trial structure: green cross (t=0s), acoustic signal (t=2s), red arrow cue (t=3s, 1.25s duration), continuous visual feedback (t=4.25s, 3.75s duration), inter-trial interval (1.5-3.5s). Signal processing used participant-specific Most Discriminant Frequency Band (MDFB) selection (5-35 Hz range), fifth-order Butterworth filtering, Common Spatial Pattern (CSP) with 3 pairs of spatial filters, and Linear Discriminant Analysis (LDA) classifier trained on calibration data. Participants completed 6 questionnaires assessing demographics, personality (16PF5), cognitive traits, spatial abilities (Mental Rotation test), learning style (ILS), and pre/post-experiment states (NeXT questionnaire).

References

Pillette, L., Roc, A., N’kaoua, B., & Lotte, F. (2021). Experimenters’ influence on mental-imagery based brain-computer interface user training. International Journal of Human-Computer Studies, 149, 102603. Benaroch, C., Yamamoto, M. S., Roc, A., Dreyer, P., Jeunet, C., & Lotte, F. (2022). When should MI-BCI feature optimization include prior knowledge, and which one?. Brain-Computer Interfaces, 9(2), 115-128. Appelhoff, S., Sanderson, M., Brooks, T., Vliet, M., Quentin, R., Holdgraf, C., Chaumon, M., Mikulan, E., Tavabi, K., Hochenberger, R., Welke, D., Brunner, C., Rockhill, A., Larson, E., Gramfort, A. and Jas, M. (2019). MNE-BIDS: Organizing electrophysiological data into the BIDS format and facilitating their analysis. Journal of Open Source Software 4: (1896). https://doi.org/10.21105/joss.01896 Pernet, C. R., Appelhoff, S., Gorgolewski, K. J., Flandin, G., Phillips, C., Delorme, A., Oostenveld, R. (2019). EEG-BIDS, an extension to the brain imaging data structure for electroencephalography. Scientific Data, 6, 103. https://doi.org/10.1038/s41597-019-0104-8 Generated by MOABB 1.5.0 (Mother of All BCI Benchmarks) https://github.com/NeuroTechX/moabb

Dataset Information#

Dataset ID

NM000250

Title

Class for Dreyer2023 dataset management. MI dataset

Author (year)

Dreyer2023

Canonical

—

Importable as

NM000250, Dreyer2023

Year

2021

Authors

Pauline Dreyer, Aline Roc, Léa Pillette, Sébastien Rimbert, Fabien Lotte

License

CC-BY-4.0

Citation / DOI

Unknown

Source links

OpenNeuro | NeMAR | Source URL

Found an issue with this dataset?

If you encounter any problems with this dataset (missing files, incorrect metadata, loading errors, etc.), please let us know!

Report an Issue on GitHub

Technical Details#

Subjects & recordings

  • Subjects: 87

  • Recordings: 520

  • Tasks: 1

Channels & sampling rate

  • Channels: 27

  • Sampling rate (Hz): 512.0

  • Duration (hours): 63.4652734375

Tags

  • Pathology: Healthy

  • Modality: Visual

  • Type: Motor

Files & format

  • Size on disk: 8.8 GB

  • File count: 520

  • Format: BIDS

License & citation

  • License: CC-BY-4.0

  • DOI: —

Provenance

API Reference#

Use the NM000250 class to access this dataset programmatically.

class eegdash.dataset.NM000250(cache_dir: str, query: dict | None = None, s3_bucket: str | None = None, **kwargs)[source]#

Bases: EEGDashDataset

Class for Dreyer2023 dataset management. MI dataset

Study:

nm000250 (NeMAR)

Author (year):

Dreyer2023

Canonical:

—

Also importable as: NM000250, Dreyer2023.

Modality: eeg; Experiment type: Motor; Subject type: Healthy. Subjects: 87; recordings: 520; 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/nm000250 NeMAR dataset: https://nemar.org/dataexplorer/detail?dataset_id=nm000250

Examples

>>> from eegdash.dataset import NM000250
>>> dataset = NM000250(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, overwrite=False)[source]#

Save the dataset to disk.

Parameters:

  • path (str or Path) – Destination file path.

  • overwrite (bool, default False) – If True, overwrite existing file.

Return type:

None

See Also#