From EEG Cleaning to Decoding: The Role of Artifact Rejection in MI-based BCIs

TL;DR

This paper introduces FAAR, a lightweight automated artifact rejection method for EEG-based motor imagery BCIs, improving robustness and consistency across datasets and real-time applications.

Contribution

The authors propose FAAR, a novel automated artifact rejection technique that adaptively thresholds EEG data without prior knowledge, enhancing BCI reliability and reducing variability.

Findings

FAAR performs well across 13 datasets, especially in low-SNR conditions.

FAAR reduces inter-subject performance variability.

FAAR supports real-time EEG artifact rejection without manual tuning.

Abstract

Motor imagery (MI) BCIs are sensitive to EEG artifacts, yet the practical impact of automated artifact rejection on downstream MI decoding performance remains unclear. While most work focuses on decoder design, the contribution of data curation, particularly automated rejection policies, has received comparatively less attention, despite its importance for robust ML pipelines. Here, we propose Fast Automatic Artifact Rejection (FAAR), a lightweight method that computes a compact set of artifact-sensitive features, derives an epoch-level Signal Quality Index, adaptively selects rejection thresholds, and automatically rejects contaminated epochs without requiring prior knowledge of artifact types or manual threshold tuning. We evaluate FAAR on 13 publicly available MI datasets and compare it to a no-rejection baseline, AutoReject, and Isolation Forest. We show rejection effects are strongly subject- and regime-dependent, with the largest gains in low-baseline/low-SNR conditions, so it should be used adaptively. FAAR reduces inter-subject performance variability, an important property for MI-BCI reliability and BCI-illiteracy, without aggressive data removal. Finally, FAAR's lightweight and fully automated thresholding yields consistent rejection behavior across offline curation, training, and online filtering, and supports real-time BCI constraints.