EEGdash›NeMAR›NM000158
Iss. 158 · 50 subjects · 50 recordings · CC-BY-4.0
Dataset Brief · Liu, Lv et al. 2023 — EEG datasets of stroke patients (motor…

NM000158: eeg dataset, 50 subjects#

Liu, Lv et al. 2023 — EEG datasets of stroke patients (motor imagery)

Citation: Haijie Liu, Penghu Wei, Haochong Wang, Xiaodong Lv, Wei Duan, Meijie Li, Yan Zhao, Qingmei Wang, Xinyuan Chen, Gaige Shi, Bo Han, Junwei Hao (2022). Liu, Lv et al. 2023 — EEG datasets of stroke patients (motor imagery). 10.82901/nemar.nm000158

50-participant EEG dataset — Liu, Lv et al. 2023 — EEG datasets of stroke patients (motor imagery).

Data & curation Haijie Liu · Penghu Wei · Haochong Wang · Xiaodong Lv · Wei Duan · Meijie Li · …
Year 2022 · Distributed via NeMAR
Funding National Natural Science Foundation of China (grant nos. 82090043 and 81825008)
EEG · 29 ch500 HzBIDS 1.9.0Task · imageryOtherMultisensoryMotor
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 NM000158

dataset = NM000158(cache_dir="./data")
# Get the raw object of the first recording
raw = dataset.datasets[0].raw
print(raw.info)

Filter by subject

dataset = NM000158(cache_dir="./data", subject="01")

Advanced query

dataset = NM000158(
    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{nm000158,
  title = {Liu, Lv et al. 2023 — EEG datasets of stroke patients (motor imagery)},
  author = {Haijie Liu and Penghu Wei and Haochong Wang and Xiaodong Lv and Wei Duan and Meijie Li and Yan Zhao and Qingmei Wang and Xinyuan Chen and Gaige Shi and Bo Han and Junwei Hao},
  doi = {10.82901/nemar.nm000158},
  url = {https://doi.org/10.82901/nemar.nm000158},
}
§ 02Study · The README

About This Dataset#

Dataset [1]_ from the study on motor imagery [2]_.

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

DOI

Dataset [1] from the study on motor imagery [2]

left_hand
View full README

DOI

Dataset [1] from the study on motor imagery [2]

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: left_hand, right_hand

  • Cue duration: 2.0 s

  • Imagery duration: 4.0 s

Data Structure

  • Trials: 40

  • Trials per class: left_hand=20, right_hand=20

  • Trials context: 40 trials per subject total (20 left-hand, 20 right-hand), alternating. Each trial: 8s total (instruction + 4s MI + break). Training/test split: 60%/40%.

Preprocessing

  • Data state: preprocessed

  • Preprocessing applied: True

  • Steps: baseline removal (mean removal method), FIR filtering (0.5-40 Hz)

  • Highpass filter: 0.5 Hz

  • Lowpass filter: 40.0 Hz

  • Bandpass filter: [0.5, 40.0]

  • Filter type: FIR

  • Epoch window: [0.0, 8.0]

  • Notes: Preprocessed with EEGLAB toolbox in MATLAB R2019b. Filtered data split into trials x channels x time-samples format by marker ‘1’. Some motion artifacts present in subjects 4, 5, 13, 14, 18, 24, 28, 33, 42, 43, 47, 48, 49.

Signal Processing

  • Classifiers: CSP+LDA, FBCSP+SVM, TSLDA+DGFMDRM, TWFB+DGFMDM

  • Feature extraction: CSP, FBCSP, ERD/ERS, Riemannian geometry (SCMs on SPD manifolds), Tangent Space, Time-Frequency (Morlet wavelet), TWFB (Time Window Filter Bank)

  • Frequency bands: alpha=[8.0, 15.0] Hz; beta=[15.0, 30.0] Hz; analyzed=[8.0, 30.0] Hz

  • Spatial filters: CSP, FBCSP, Discriminant Geodesic Filtering

Cross-Validation

  • Method: 10-fold cross-validation

  • Folds: 10

  • Evaluation type: within_subject

Performance (Original Study)

  • Csp+Lda Accuracy: 55.57

  • Fbcsp+Svm Accuracy: 57.57

  • Tslda+Dgfmdrm Accuracy: 61.2

  • Twfb+Dgfmdm Accuracy: 72.21

  • Twfb+Dgfmdm Kappa: 0.4442

  • Twfb+Dgfmdm Precision: 0.7543

  • Twfb+Dgfmdm Sensitivity: 0.7845

BCI Application

  • Applications: rehabilitation

  • Environment: hospital

  • Online feedback: False

Tags

  • Pathology: Stroke

  • Modality: Motor

  • Type: Motor Imagery

Documentation

  • Description: EEG motor imagery dataset from 50 acute stroke patients performing left- and right-handed hand-grip imagination tasks. First open dataset addressing left- and right-handed motor imagery in acute stroke patients.

  • DOI: 10.1038/s41597-023-02787-8

  • License: CC-BY-4.0

  • Investigators: Haijie Liu, Penghu Wei, Haochong Wang, Xiaodong Lv, Wei Duan, Meijie Li, Yan Zhao, Qingmei Wang, Xinyuan Chen, Gaige Shi, Bo Han, Junwei Hao

  • Senior author: Junwei Hao

  • Contact: haojunwei@vip.163.com

  • Institution: Xuanwu Hospital Capital Medical University

  • Department: Department of Neurology

  • Address: Beijing, 100053, China

  • Country: CN

  • Repository: Figshare

  • Data URL: https://doi.org/10.6084/m9.figshare.21679035.v5

  • Publication year: 2024

  • Funding: National Natural Science Foundation of China (grant nos. 82090043 and 81825008)

  • Ethics approval: Ethics Committee of Xuanwu Hospital of Capital Medical University (No. 2021-236)

  • Keywords: motor imagery, BCI, brain-computer interface, stroke patients, EEG, rehabilitation, acute stroke, hand-grip imagery, databases, scientific data

Abstract

The brain-computer interface (BCI) is a technology that involves direct communication with parts of the brain and has evolved rapidly in recent years; it has begun to be used in clinical practice, such as for patient rehabilitation. Patient electroencephalography (EEG) datasets are critical for algorithm optimization and clinical applications of BCIs but are rare at present. We collected data from 50 acute stroke patients with wireless portable saline EEG devices during the performance of two tasks: 1) imagining right-handed movements and 2) imagining left-handed movements. The dataset consists of four types of data: 1) the motor imagery instructions, 2) raw recording data, 3) pre-processed data after removing artefacts and other manipulations, and 4) patient characteristics. This is the first open dataset to address left- and right-handed motor imagery in acute stroke patients.

Methodology

50 acute stroke patients (1-30 days post-stroke) performed 40 trials of hand-grip motor imagery (20 left, 20 right). Each 8s trial included instruction, 4s video-guided imagery, and rest phases. EEG recorded with ZhenTec NT1 wireless system (29 EEG + 2 EOG channels) at 500 Hz. Data organized in EEG-BIDS format with raw (.mat) and preprocessed (.edf) versions. Clinical assessments: NIHSS (mean=4.16, SD=2.85), MBI (mean=70.94, SD=18.22), mRS (mean=2.66, SD=1.44). 23 patients right hemiplegia, 27 left hemiplegia.

References

Liu, Haijie; Lv, Xiaodong (2022). EEG datasets of stroke patients. figshare. Dataset. DOI: https://doi.org/10.6084/m9.figshare.21679035.v5 Liu, Haijie, Wei, P., Wang, H. et al. An EEG motor imagery dataset for brain computer interface in acute stroke patients. Sci Data 11, 131 (2024). DOI: https://doi.org/10.1038/s41597-023-02787-8 Notes To add the break and instruction events, set the break_events and instr_events parameters to True while instantiating the class. .. versionadded:: 1.1.1 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) NeuroTechX/moabb

§ 03Cohort · Participants

Cohort#

Dataset Statistics#

Age distribution by gender (n=50, range 71–71 yr, mean 71.0 yr)

70
Other · 50

Channel counts: 29 ch (n=50 recordings)

Sampling frequencies: 500.0 Hz (n=50 recordings)

Total recording duration: 4 h 26 min

§ 04Signal · Electrodes & trace

Signal · Electrodes & live trace#

Fig. 01 Signal & montage 29 ch · EEG · 500 Hz · 50 subjects, 50 recordings
Live trace viewer — sub-13 · ses-0 · task-imagery · run-0

Showing one representative recording out of 50 subjects and 50 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 · 29 sensors — 29 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 — NM000158
HED event descriptors word cloud HED event descriptors word cloud — NM000158
§ 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

NM000158

Title

Liu, Lv et al. 2023 — EEG datasets of stroke patients (motor imagery)

Author (year)

Liu2024

Canonical

—

Importable as

NM000158, Liu2024

Year

2022

Authors

Haijie Liu, Penghu Wei, Haochong Wang, Xiaodong Lv, Wei Duan, Meijie Li, Yan Zhao, Qingmei Wang, Xinyuan Chen, Gaige Shi, Bo Han, Junwei Hao

License

CC-BY-4.0

Citation / DOI

10.82901/nemar.nm000158

Source links

OpenNeuro | NeMAR | Source URL
Copy-paste BibTeX
@dataset{nm000158,
  title = {Liu, Lv et al. 2023 — EEG datasets of stroke patients (motor imagery)},
  author = {Haijie Liu and Penghu Wei and Haochong Wang and Xiaodong Lv and Wei Duan and Meijie Li and Yan Zhao and Qingmei Wang and Xinyuan Chen and Gaige Shi and Bo Han and Junwei Hao},
  doi = {10.82901/nemar.nm000158},
  url = {https://doi.org/10.82901/nemar.nm000158},
}
§ 06API · Programmatic access

API Reference#

Signature
eegdash.dataset
class
eegdash.dataset.NM000158(cache_dir, query=None, s3_bucket=None, **kwargs)
Bases: EEGDashDataset
Author (year)Liu2024
Canonical—
Importable asNM000158 · Liu2024
Sourceeegdash/dataset/registry.py · [source ↗]
class eegdash.dataset.NM000158(cache_dir: str, query: dict | None = None, s3_bucket: str | None = None, **kwargs)[source]#

Liu, Lv et al. 2023 — EEG datasets of stroke patients (motor imagery)

Study:

nm000158 (NeMAR)

Author (year):

Liu2024

Canonical:

—

Also importable as: NM000158, Liu2024.

Modality: eeg; Experiment type: Motor; Subject type: Other. Subjects: 50; recordings: 50; 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/nm000158 NeMAR dataset: https://nemar.org/dataexplorer/detail?dataset_id=nm000158 DOI: https://doi.org/10.82901/nemar.nm000158

Examples

>>> from eegdash.dataset import NM000158
>>> dataset = NM000158(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 — NM000158.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 nm000158 to reproduce the tutorial on this dataset.

Citation

Haijie Liu, Penghu Wei, Haochong Wang, Xiaodong Lv, Wei Duan, … (2022). Liu, Lv et al. 2023 — EEG datasets of stroke patients (motor imagery). 10.82901/nemar.nm000158

Provenance

¹Contributed to nemar in BIDS format.

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

³Persistent identifier: 10.82901/nemar.nm000158.

BIDS
BIDS 1.9.0
Sidecars
events · events.json · channels · eeg.json
Provenance
CC-BY-4.0 · 10.82901/nemar.nm000158
Machine-readable
schema.org/Dataset · Croissant
Mirrors
OpenNeuro · NEMAR · HF org

See Also#