EEG Data for Muscular Artifact Removal via Subspace ICA

This dataset supports a study on suppressing overlearning in Independent Component Analysis (ICA) for removing muscular artifacts from electroencephalographic (EEG) records. The research demonstrates a subspace projection technique that improves artifact separation for short, high-density EEG signals compared to standard ICA methods. The author is Jan Sebek, with the data last updated in June 2020.

Use Cases

• Benchmark the proposed subspace ICA method against standard ICA for muscular artifact removal in short EEG records.
• Analyze the separation performance of artifact-free EEG components from muscular artifact components in high-density channel setups.
• Validate the suppression of ICA overlearning in short signal segments using the provided EEG data.

Strengths

• Data is associated with a peer-reviewed methodological paper addressing a specific signal processing challenge (ICA overlearning).
• The dataset is openly available under a CC0 1.0 public domain license, allowing unrestricted use.

Limitations

• The specific volume, format, and structure of the data (e.g., row count, columns) are not described in the provided metadata.
• The temporal and geographic coverage of the underlying EEG recordings is unknown, limiting contextual analysis.

Provenance

Source

Dryad digital repository.

Collection Method

EEG records processed with a novel subspace ICA technique for muscular artifact removal, as described in the associated paper.

Freshness

Last updated on 2020-06-24.