moabb.datasets.Kumar2024#

class moabb.datasets.Kumar2024(subjects=None, sessions=None, *, return_all_modalities=False)[source]#

Bases: BaseDataset

[source]

Dataset Snapshot

Kumar2024

Multi-session longitudinal MI training dataset with 18 BCI-naive subjects over 6 sessions. Demonstrates that inter-subject transfer learning from a single expert promotes acquisition of individual BCI skills via unsupervised domain adaptation.

Motor Imagery, 2 classes (left_hand vs right_hand)

AuthorsSatyam Kumar, Hussein Alawieh, Frigyes Samuel Racz, Rawan Fakhreddine, Jose del R. Millan

🇺🇸 The University of Texas at Austin, US·2024·satyam.kumar@utexas.edu
Motor Imagery Code: Kumar2024 18 subjects 6 sessions 22 ch 512 Hz 2 classes 5.0 s trials CC BY 4.0

Class Labels: left_hand, right_hand

Read Paper Compare Similar Report Issue

Overview

Multi-session longitudinal motor imagery dataset from Kumar et al. 2024.

Dataset from

This dataset contains EEG recordings from 18 healthy, BCI-naive participants (7 female, 11 male, age 23.22 +/- 3.59 years) performing left-hand and right-hand motor imagery over 6 sessions conducted on separate days.

Session 1 was an offline calibration session with 4 bar-feedback runs. Sessions 2-6 were online sessions consisting of bar-feedback runs with continuous visual feedback followed by car racing games. In each bar-feedback run, subjects performed 20 trials (10 left-hand, 10 right-hand MI) in pseudo-random order.

For MOABB, only bar-feedback runs are included (car racing runs are excluded). Session 2 (online session 1) contains 4 bar runs, and sessions 3-6 (online sessions 2-5) each contain 3 bar runs.

EEG was recorded at 512 Hz using an ANT Neuro eego mylab system with 22 EEG electrodes positioned according to the international 10-10 system (reference: CPz, ground: AFz), plus 3 EOG channels. Data is stored in GDF (General Data Format) files.

The two transfer learning training protocols used were:

  • - Generic Recentering (GR): unsupervised domain adaptation (subjects 1-9)
  • Personally Assisted Recentering (PAR): supervised recalibration (subjects 10-18)

Trial structure (bar task):

  • - Fixation cross: 1.0 s
  • Cue presentation: 1.5 s
  • MI + visual feedback: up to 5 s (offline) or 7 s (online)
  • Result display: 2.0 s
  • Inter-trial rest: 1.5 s
Open in Overview tab →

Citation & Impact

Stimulus Protocol
../_images/Kumar2024.svg

5s task window per trial · 2-class motor imagery paradigm · 3 runs/session across 6 sessions

HED Event Tags
HED tags2/2 events annotated

Source: MOABB BIDS HED annotation mapping.

Agent-action
2
Sensory-event
2
left_hand
Sensory-eventAgent-action
right_hand
Sensory-eventAgent-action

HED tree view

Tree · left_hand
├─ Sensory-event
│  ├─ Experimental-stimulus
│  └─ Visual-presentation
└─ Agent-action
   └─ Imagine
      ├─ Move
      └─ Left
         └─ Hand
Tree · right_hand
├─ Sensory-event
│  ├─ Experimental-stimulus
│  └─ Visual-presentation
└─ Agent-action
   └─ Imagine
      ├─ Move
      └─ Right
         └─ Hand
Channel Summary
Total channels22
EEG22 (EEG)
EOG3
Montagestandard_1020
Sampling512 Hz
ReferenceCPz
Notch / line60 Hz

This diagram is automatically generated from MOABB metadata. Please consult the original publication to confirm the experimental protocol details.

Multi-session longitudinal motor imagery dataset from Kumar et al. 2024.

Dataset from [1] [2].

This dataset contains EEG recordings from 18 healthy, BCI-naive participants (7 female, 11 male, age 23.22 +/- 3.59 years) performing left-hand and right-hand motor imagery over 6 sessions conducted on separate days.

Session 1 was an offline calibration session with 4 bar-feedback runs. Sessions 2-6 were online sessions consisting of bar-feedback runs with continuous visual feedback followed by car racing games. In each bar-feedback run, subjects performed 20 trials (10 left-hand, 10 right-hand MI) in pseudo-random order.

For MOABB, only bar-feedback runs are included (car racing runs are excluded). Session 2 (online session 1) contains 4 bar runs, and sessions 3-6 (online sessions 2-5) each contain 3 bar runs.

EEG was recorded at 512 Hz using an ANT Neuro eego mylab system with 22 EEG electrodes positioned according to the international 10-10 system (reference: CPz, ground: AFz), plus 3 EOG channels. Data is stored in GDF (General Data Format) files.

The two transfer learning training protocols used were: - Generic Recentering (GR): unsupervised domain adaptation (subjects 1-9) - Personally Assisted Recentering (PAR): supervised recalibration (subjects 10-18)

Trial structure (bar task): - Fixation cross: 1.0 s - Cue presentation: 1.5 s - MI + visual feedback: up to 5 s (offline) or 7 s (online) - Result display: 2.0 s - Inter-trial rest: 1.5 s

References

[1]

S. Kumar, H. Alawieh, F. S. Racz, R. Fakhreddine, and J. del R. Millan, “Transfer learning promotes acquisition of individual BCI skills,” PNAS Nexus, vol. 3, no. 3, p. pgae076, 2024. DOI: 10.1093/pnasnexus/pgae076

[2]

S. Kumar, H. Alawieh, F. S. Racz, R. Fakhreddine, and J. del R. Millan, “Multi-Session longitudinal MI training dataset,” Zenodo, 2024. DOI: 10.5281/zenodo.10694880

from moabb.datasets import Kumar2024
dataset = Kumar2024()
data = dataset.get_data(subjects=[1])
print(data[1])

Dataset summary

#Subj

18

#Chan

22

#Classes

2

#Trials / class

varies

Trials length

5 s

Freq

512 Hz

#Sessions

6

#Runs

varies

Total_trials

7156

Participants

  • Population: healthy

  • Age: 23.22 years

  • BCI experience: naive

Equipment

  • Amplifier: ANT Neuro eego mylab

  • Electrodes: EEG

  • Montage: standard_1020

  • Reference: CPz

Preprocessing

  • Data state: raw

  • Notes: Raw EEG signals recorded in GDF format. For analysis, signals were bandpass filtered at 8-30 Hz using a second-order Butterworth filter.

Data Access

Experimental Protocol

  • Paradigm: imagery

  • Feedback: continuous visual

  • Stimulus: visual cue and bar feedback

Notes

Added in version 1.2.0.

__init__(subjects=None, sessions=None, *, return_all_modalities=False)[source]#

Initialize function for the BaseDataset.

property all_subjects#

Full list of subjects available in this dataset (unfiltered).

convert_to_bids(path=None, subjects=None, overwrite=False, format='EDF', verbose=None)[source]#

Convert the dataset to BIDS format.

Saves the raw EEG data in a BIDS-compliant directory structure. Unlike the caching mechanism (see CacheConfig), the files produced here do not contain a processing-pipeline hash (desc-<hash>) in their names, making the output a clean, shareable BIDS dataset.

Parameters:

  • path (str | Path | None) – Directory under which the BIDS dataset will be written. If None the default MNE data directory is used (same default as the rest of MOABB).

  • subjects (list of int | None) – Subject numbers to convert. If None, all subjects in subject_list are converted.

  • overwrite (bool) – If True, existing BIDS files for a subject are removed before saving. Default is False.

  • format (str) – The file format for the raw EEG data. Supported values are "EDF" (default), "BrainVision", and "EEGLAB".

  • verbose (str | None) – Verbosity level forwarded to MNE/MNE-BIDS.

Returns:

bids_root – Path to the root of the written BIDS dataset.

Return type:

pathlib.Path

Examples

>>> from moabb.datasets import AlexMI
>>> dataset = AlexMI()
>>> bids_root = dataset.convert_to_bids(path='/tmp/bids', subjects=[1])

See also

CacheConfig

Cache configuration for get_data().

moabb.datasets.bids_interface.get_bids_root

Return the BIDS root path.

Notes

Added in version 1.5.

data_path(subject, path=None, force_update=False, update_path=None, verbose=None)[source]#

Return the path to the extracted ZIP for this dataset.

Downloads the single ZIP from Zenodo and extracts it if needed.

Parameters:

  • subject (int) – Subject number (1-18).

  • path (None | str) – Storage location override.

  • force_update (bool) – Re-download even if local copy exists.

  • update_path (bool | None) – Unused, kept for API compatibility.

  • verbose (bool, str, int, or None) – Verbosity level.

Returns:

Path to the extracted dataset root directory.

Return type:

Path

download(subject_list=None, path=None, force_update=False, update_path=None, accept=False, verbose=None)[source]#

Download all data from the dataset.

This function is only useful to download all the dataset at once.

Parameters:

  • subject_list (list of int | None) – List of subjects id to download, if None all subjects are downloaded.

  • path (None | str) – Location of where to look for the data storing location. If None, the environment variable or config parameter MNE_DATASETS_(dataset)_PATH is used. If it doesn’t exist, the “~/mne_data” directory is used. If the dataset is not found under the given path, the data will be automatically downloaded to the specified folder.

  • force_update (bool) – Force update of the dataset even if a local copy exists.

  • update_path (bool | None) – If True, set the MNE_DATASETS_(dataset)_PATH in mne-python config to the given path. If None, the user is prompted.

  • accept (bool) – Accept licence term to download the data, if any. Default: False

  • verbose (bool, str, int, or None) – If not None, override default verbose level (see mne.verbose()).

get_additional_metadata(subject: str, session: str, run: str) → None | DataFrame[source]#

Load additional metadata for a specific subject, session, and run.

This method is intended to be overridden by subclasses to provide additional metadata specific to the dataset. The metadata is typically loaded from an events.tsv file or similar data source.

Parameters:

  • subject (str) – The identifier for the subject.

  • session (str) – The identifier for the session.

  • run (str) – The identifier for the run.

Returns:

A DataFrame containing the additional metadata if available, otherwise None.

Return type:

None | pd.DataFrame

get_block_repetition(paradigm, subjects, block_list, repetition_list)[source]#

Select data for all provided subjects, blocks and repetitions.

subject -> session -> run -> block -> repetition

See also

BaseDataset.get_data

Parameters:

  • subjects (List of int) – List of subject number

  • block_list (List of int) – List of block number

  • repetition_list (List of int) – List of repetition number inside a block

Returns:

data – dict containing the raw data

Return type:

Dict

get_data(subjects=None, cache_config=None, process_pipeline=None)[source]#

Return the data corresponding to a list of subjects.

The returned data is a dictionary with the following structure:

data = {'subject_id' :
            {'session_id':
                {'run_id': run}
            }
        }

subjects are on top, then we have sessions, then runs. A sessions is a recording done in a single day, without removing the EEG cap. A session is constitued of at least one run. A run is a single contiguous recording. Some dataset break session in multiple runs.

Processing steps can optionally be applied to the data using the *_pipeline arguments. These pipelines are applied in the following order: raw_pipeline -> epochs_pipeline -> array_pipeline. If a *_pipeline argument is None, the step will be skipped. Therefore, the array_pipeline may either receive a mne.io.Raw or a mne.Epochs object as input depending on whether epochs_pipeline is None or not.

Parameters:

  • subjects (List of int) – List of subject number

  • cache_config (dict | CacheConfig) – Configuration for caching of datasets. See CacheConfig for details.

  • process_pipeline (Pipeline | None) – Optional processing pipeline to apply to the data. To generate an adequate pipeline, we recommend using moabb.utils.make_process_pipelines(). This pipeline will receive mne.io.BaseRaw objects. The steps names of this pipeline should be elements of StepType. According to their name, the steps should either return a mne.io.BaseRaw, a mne.Epochs, or a numpy.ndarray(). This pipeline must be “fixed” because it will not be trained, i.e. no call to fit will be made.

Returns:

data – dict containing the raw data

Return type:

Dict

property metadata: DatasetMetadata | None[source]#

Return structured metadata for this dataset.

Returns the DatasetMetadata object from the centralized catalog, or None if metadata is not available for this dataset.

Returns:

The metadata object containing acquisition parameters, participant demographics, experiment details, and documentation. Returns None if no metadata is registered for this dataset.

Return type:

DatasetMetadata | None

Examples

>>> from moabb.datasets import BNCI2014_001
>>> dataset = BNCI2014_001()
>>> dataset.metadata.participants.n_subjects
9
>>> dataset.metadata.acquisition.sampling_rate
250.0