lasfk's EEG-fNIRS-based Handwriting Trajectory Dataset contains raw multimodal signals and metadata for a competition. The dataset includes synchronized EEG and fNIRS recordings, organized into labeled training trials and unlabeled test trials. It was last updated on 2026-04-25.
Use Cases
- Classify imagined handwriting characters based on synchronized EEG and fNIRS signals.
- Train multimodal neural decoders for motor imagery tasks.
- Benchmark machine learning models for time-series classification of brain activity.
- Develop end-to-end competition workflows for BCI challenges.
Strengths
- Dataset is structured for an end-to-end competition workflow with separate training and test splits.
- Contains synchronized recordings from two complementary neuroimaging modalities: EEG and fNIRS.
- Includes raw signals alongside metadata files for labeled and unlabeled trials.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- Description metadata is limited; actual data quality requires manual inspection after download.
Provenance
- Source
- lasfk
- Freshness
- Last updated 2026-04-25 07:39:48.