NM000342: eeg dataset, 12 subjects#

CastillosCVEP40

Access recordings and metadata through EEGDash.

Citation: Kalou Cabrera Castillos, Simon Ladouce, Ludovic Darmet, Frédéric Dehais (2023). CastillosCVEP40. 10.1016/j.neuroimage.2023.120446

Modality: eeg Subjects: 12 Recordings: 12 License: CC-BY-4.0 Source: nemar

Metadata: Complete (100%)

Quickstart#

Install

pip install eegdash

Access the data

from eegdash.dataset import NM000342

dataset = NM000342(cache_dir="./data")
# Get the raw object of the first recording
raw = dataset.datasets[0].raw
print(raw.info)

Filter by subject

dataset = NM000342(cache_dir="./data", subject="01")

Advanced query

dataset = NM000342(
    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{nm000342,
  title = {CastillosCVEP40},
  author = {Kalou Cabrera Castillos and Simon Ladouce and Ludovic Darmet and Frédéric Dehais},
  doi = {10.1016/j.neuroimage.2023.120446},
  url = {https://doi.org/10.1016/j.neuroimage.2023.120446},
}

About This Dataset#

CastillosCVEP40

c-VEP and Burst-VEP dataset from Castillos et al. (2023)

Dataset Overview

Code: CastillosCVEP40 Paradigm: cvep DOI: https://doi.org/10.1016/j.neuroimage.2023.120446

View full README

CastillosCVEP40

c-VEP and Burst-VEP dataset from Castillos et al. (2023)

Dataset Overview

Code: CastillosCVEP40 Paradigm: cvep DOI: https://doi.org/10.1016/j.neuroimage.2023.120446 Subjects: 12 Sessions per subject: 1 Events: 0=100, 1=101 Trial interval: (0, 0.25) s File format: EEGLAB .set Number of contributing labs: 1

Acquisition

Sampling rate: 500.0 Hz Number of channels: 32 Channel types: eeg=32 Channel names: C3, C4, CP1, CP2, CP5, CP6, Cz, F10, F3, F4, F7, F8, F9, FC1, FC2, FC5, FC6, Fp1, Fp2, Fz, O1, O2, Oz, P10, P3, P4, P7, P8, P9, Pz, T7, T8 Montage: standard_1020 Hardware: BrainProducts LiveAmp 32 Reference: FCz Ground: FPz Sensor type: EEG Line frequency: 50.0 Hz Online filters: {‘line_noise_filter’: ‘IIR cut-band filter 49.9-50.1 Hz, order 16’} Impedance threshold: 25.0 kOhm Cap manufacturer: BrainProducts Cap model: Acticap Electrode type: active

Participants

Number of subjects: 12 Health status: healthy Age: mean=30.6, std=7.1 Gender distribution: female=4, male=8 Species: human

Experimental Protocol

Paradigm: cvep Task type: reactive BCI Number of classes: 2 Class labels: 0, 1 Trial duration: 2.2 s Tasks: visual_attention Study design: factorial design Study domain: brain-computer interface Feedback type: none Stimulus type: visual flicker Stimulus modalities: visual Primary modality: visual Synchronicity: synchronous Mode: offline Training/test split: False Instructions: focus on targets that were cued sequentially in a random order for 0.5 s, followed by a 2.2 s stimulation phase Stimulus presentation: cue_duration=500 ms, stimulation_duration=2200 ms, inter_trial_interval=700 ms, cue_type=red-bordered square around target stimulus, display=Dell P2419HC, 1920×1080 pixels, 265 cd/m², 60 Hz

HED Event Annotations

Schema: HED 8.4.0 | Browse: https://www.hedtags.org/hed-schema-browser 0

     ├─ Sensory-event
     ├─ Experimental-stimulus
     ├─ Visual-presentation
     └─ Label/intensity_0

1

├─ Sensory-event
├─ Experimental-stimulus
├─ Visual-presentation
└─ Label/intensity_1

Paradigm-Specific Parameters

Detected paradigm: cvep Code type: m-sequence Number of targets: 4 Cue duration: 0.5 s

Data Structure

Trials: 60 Blocks per session: 15 Trials context: 15 blocks x 4 trials per block = 60 trials per subject for m-sequence c-VEP at 40% amplitude

Preprocessing

Data state: raw

Signal Processing

Classifiers: CNN (Convolutional Neural Network) Feature extraction: sliding windows, bitwise decoding

Cross-Validation

Evaluation type: offline

Performance (Original Study)

Accuracy: 95.6% Burst 100 Accuracy 17.6S Calibration: 90.5 Burst 100 Accuracy 52.8S Calibration: 95.6 Burst 40 Accuracy: 94.2 Mseq 100 Accuracy 17.6S Calibration: 71.4 Mseq 100 Accuracy 52.8S Calibration: 85.0 Mean Selection Time: 1.5

BCI Application

Applications: reactive BCI Environment: laboratory Online feedback: False

Tags

Pathology: Healthy Modality: EEG Type: reactive, code-VEP, visual

Documentation

Description: Burst c-VEP Based BCI: Optimizing stimulus design for enhanced classification with minimal calibration data and improved user experience DOI: 10.1016/j.neuroimage.2023.120446 Associated paper DOI: 10.1016/j.neuroimage.2023.120446 License: CC-BY-4.0 Investigators: Kalou Cabrera Castillos, Simon Ladouce, Ludovic Darmet, Frédéric Dehais Senior author: Frédéric Dehais Contact: kalou.cabrera-castillos@isae-supaero.fr Institution: Institut Supérieur de l’Aéronautique et de l’Espace (ISAE-SUPAERO) Department: Human Factors and Neuroergonomics Address: 10 Av. Edouard Belin, Toulouse, 31400, France Country: FR Repository: Zenodo Data URL: https://zenodo.org/record/8255618 Publication year: 2023 Ethics approval: University of Toulouse CER approval number 2020-334 Keywords: Code-VEP, Reactive BCI, CNN, Amplitude depth reduction, Visual comfort

External Links

Abstract

The utilization of aperiodic flickering visual stimuli under the form of code-modulated Visual Evoked Potentials (c-VEP) represents a pivotal advancement in the field of reactive Brain–Computer Interface (rBCI). This study introduces an innovative variant of code-VEP, referred to as ‘Burst c-VEP’, involving the presentation of short bursts of aperiodic visual flashes at a deliberately slow rate (2-4 flashes per second). The study tested an offline 4-classes c-VEP protocol involving 12 participants with factorial design manipulating pattern (burst and m-sequences) and amplitude (100% or 40% depth modulations). Full amplitude burst c-VEP sequences exhibited higher accuracy (90.5% with 17.6s calibration to 95.6% with 52.8s calibration) compared to m-sequence (71.4% to 85.0%). Mean selection time was 1.5s. Lowering intensity to 40% decreased accuracy slightly to 94.2% while improving user experience substantially.

Methodology

Factorial experimental design with 12 participants. Four conditions: burst vs m-sequence × 100% vs 40% amplitude depth. Participants seated comfortably, presented with 15 blocks of 4 trials for each condition. Each trial: 0.5s cue (red-bordered square), 2.2s stimulation, 0.7s inter-trial interval. Four disc targets (150 pixels) on Dell monitor (60 Hz). Background: medium grey (50% max luminance, 124 lux). 100% condition: modulation to brightest white (168 lux). 40% condition: 40% of grey-to-white range (142 lux). EEG recorded with BrainProducts LiveAmp (32 channels, 500 Hz), impedance <25kΩ. Analysis on subset: O1, O2, Oz, Pz, P3, P4, P8, P9. Preprocessing: average re-reference, IIR notch filter (49.9-50.1 Hz, order 16), epoching (0-2.2s), baseline removal. Classification: CNN architecture with sliding windows for bitwise decoding.

References

Kalou Cabrera Castillos. (2023). 4-class code-VEP EEG data [Data set]. Zenodo.(dataset). DOI: https://doi.org/10.5281/zenodo.8255618 Kalou Cabrera Castillos, Simon Ladouce, Ludovic Darmet, Frédéric Dehais. Burst c-VEP Based BCI: Optimizing stimulus design for enhanced classification with minimal calibration data and improved user experience,NeuroImage,Volume 284, 2023,120446,ISSN 1053-8119 DOI: https://doi.org/10.1016/j.neuroimage.2023.120446 Notes .. versionadded:: 1.1.0 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

NM000342

Title

CastillosCVEP40

Author (year)

Castillos2023_CastillosCVEP40

Canonical

CastillosCVEP40

Importable as

NM000342, Castillos2023_CastillosCVEP40, CastillosCVEP40

Year

2023

Authors

Kalou Cabrera Castillos, Simon Ladouce, Ludovic Darmet, Frédéric Dehais

License

CC-BY-4.0

Citation / DOI

doi:10.1016/j.neuroimage.2023.120446

Source links

OpenNeuro | NeMAR | Source URL
Copy-paste BibTeX
@dataset{nm000342,
  title = {CastillosCVEP40},
  author = {Kalou Cabrera Castillos and Simon Ladouce and Ludovic Darmet and Frédéric Dehais},
  doi = {10.1016/j.neuroimage.2023.120446},
  url = {https://doi.org/10.1016/j.neuroimage.2023.120446},
}

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: 12

  • Recordings: 12

  • Tasks: 1

Channels & sampling rate

  • Channels: 32

  • Sampling rate (Hz): 500.0

  • Duration (hours): 0.8488822222222221

Tags

  • Pathology: Healthy

  • Modality: Visual

  • Type: Attention

Files & format

  • Size on disk: 145.3 MB

  • File count: 12

  • Format: BIDS

License & citation

  • License: CC-BY-4.0

  • DOI: doi:10.1016/j.neuroimage.2023.120446

Provenance

API Reference#

Use the NM000342 class to access this dataset programmatically.

class eegdash.dataset.NM000342(cache_dir: str, query: dict | None = None, s3_bucket: str | None = None, **kwargs)[source]#

Bases: EEGDashDataset

CastillosCVEP40

Study:

nm000342 (NeMAR)

Author (year):

Castillos2023_CastillosCVEP40

Canonical:

CastillosCVEP40

Also importable as: NM000342, Castillos2023_CastillosCVEP40, CastillosCVEP40.

Modality: eeg; Experiment type: Attention; Subject type: Healthy. Subjects: 12; recordings: 12; 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/nm000342 NeMAR dataset: https://nemar.org/dataexplorer/detail?dataset_id=nm000342 DOI: https://doi.org/10.1016/j.neuroimage.2023.120446

Examples

>>> from eegdash.dataset import NM000342
>>> dataset = NM000342(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#