ZJU-MI-EEG: Motor Imagery EEG Trials for Brain-Computer Interface Research

A collection of structured motor imagery EEG datasets for EEG-based MI decoding research. The data, hosted by Jiaheng-Wang on Hugging Face, was last updated on 2026-05-08. Raw EEG signals are band-pass filtered between 0.5 and 48 Hz with a notch filter, and trials are extracted from -1 to 4 seconds relative to cue onset.

Use Cases

• Training machine learning models for motor imagery classification based on preprocessed EEG trials.
• Benchmarking signal processing pipelines for EEG data filtered between 0.5 and 48 Hz.
• Studying the temporal dynamics of motor imagery from -1 to 4 seconds post-cue using structured datasets.

Strengths

• EEG data is preprocessed with a 0.5-48 Hz band-pass filter and a 50/60 Hz notch filter.
• Trial structure is standardized, with data extracted from a consistent -1 s to 4 s window relative to cue onset.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and dataset scale are unknown, which may limit suitability assessment.

Provenance

Source

Jiaheng-Wang on Hugging Face

Collection Method

Likely collected from experimental sessions involving motor imagery tasks.

Freshness

Last updated 2026-05-08 12:48:49; freshness should be verified.