NeuroTechX / NeuroTechX/moabb

[Dataset] Add OpenNeuro Motor Imagery dataset (ds004362)

Open
#908 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
1.1k
Forks
264
Avg merge
1d 13m
Merged PRs (30d)
23

Description

Dataset Information

Link Status

Link Status
OpenNeuro https://openneuro.org/datasets/ds004362/versions/1.0.0
PhysioNet (original) https://physionet.org/content/eegmmidb/1.0.0/

Publication Details

Field Value
Title BCI2000: A General-Purpose Brain-Computer Interface (BCI) System
Authors Schalk G, McFarland DJ, Hinterberger T, Birbaumer N, Wolpaw JR
Year 2004
Journal IEEE Transactions on Biomedical Engineering
Source Wadsworth Center's BCI R&D Program

Technical Specifications

Specification Value
Subjects 109
Channels 64
Sampling Rate 160 Hz
Sessions 14 runs per subject
Data Format EDF+ (with separate annotation .event files)
Electrode System International 10-10 system
BIDS Compatible Yes (OpenNeuro version)

Experimental Design

  • Paradigm: Motor Movement and Motor Imagery
  • Recording Sessions: 14 experimental runs per subject (~1500 total recordings)
    • 2 baseline runs (1 minute each): eyes open and eyes closed
    • 12 task runs (2 minutes each): three repetitions of four different motor/imagery tasks
Four Core Tasks
  1. Real left/right fist movement – subject opens/closes fist matching screen target location
  2. Imagined left/right fist movement – subject imagines fist motion without physical action
  3. Real bilateral fist or foot movement – subject performs both fists (top target) or both feet (bottom target)
  4. Imagined bilateral fist or foot movement – subject imagines bilateral motion
Event Markers
  • T0: Rest state
  • T1: Motion onset (left fist or bilateral fists)
  • T2: Motion onset (right fist or bilateral feet)

Description

This is the well-known PhysioNet EEG Motor Movement/Imagery Dataset, now also available on OpenNeuro in BIDS format. It contains EEG recordings from 109 subjects performing motor execution and motor imagery tasks. The dataset is widely used for benchmarking motor imagery classification algorithms and is one of the largest publicly available motor imagery datasets. The OpenNeuro version provides BIDS-compliant formatting for easier integration with modern neuroimaging pipelines.

Notes

  • This dataset may already be partially supported in MOABB via PhysioNet
  • The OpenNeuro version offers BIDS compatibility
  • Consider checking for overlap with existing PhysioNetMI dataset in MOABB

Related

This issue is a sub-issue of #1 (Discover new datasets)

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by checking MOABB's existing PhysioNetMI support and the dataset-discovery work tracked in issue #1, then compare it with the OpenNeuro ds004362 details provided here. Done means the dataset is added in a way that avoids duplicating existing PhysioNet coverage and supports the stated BIDS-formatted data.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
data, machine-learning
Issue type
Feature
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
45/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.