EEG-Fusion: Failure-Informed Source-Free Expert Routing for Robust Motor Imagery EEG Decoding

2026-09-27

AI Summary

This study addresses the overconfident predictions and single-class collapse arising from distribution shifts in cross-subject decoding of motor imagery EEG. To mitigate these issues, we propose a decision-level fusion framework based on failure information, which innovatively incorporates unlabeled stream diagnosis and reliability gating to reformulate expert routing as dynamic reliability estimation. By integrating Euclidean alignment, test-time adaptation, and multi-expert ensemble algorithms, this approach transcends the limitations of conventional static adaptation. Experimental results across multiple BCI datasets demonstrate that the proposed method significantly improves macro F1 scores while reducing the collapse index, thereby validating the effectiveness of unlabeled reliability estimation in suppressing subject-level failure modes.

Abstract

Subject-independent motor-imagery (MI) EEG decoding can exhibit subject-level failures even when average performance appears acceptable: under subject shift, a decoder can become an overconfident near-one-class predictor. This is especially problematic in source-free deployment, where target-user labels are unavailable during adaptation and expert selection. We present \textit{EEG-Fusion}, a failure-informed decision-level fusion framework that treats source-free MI decoding as label-free reliability estimation over heterogeneous experts. EEG-Fusion applies subject-wise Euclidean alignment and normalization-only test-time adaptation, then routes each target subject to a neural, covariance-based, or physiological-feature expert using a reliability gate trained on source-held-out folds to predict expert performance and collapse risk from label-free stream diagnostics. The gate uses confidence, entropy, prediction diversity, expert agreement, and predicted class balance; collapse is measured as the maximum predicted class fraction. In 9-fold leave-one-subject-out (LOSO) evaluation with three seeds, relative to a no-alignment raw EEGNet source-free anchor, EEG-Fusion improves subject macro-F1 from 0.417 to 0.529 on BCI IV-2a local protocol, from 0.314 to 0.482 on BNCI2014-001, and from 0.607 to 0.708 on BNCI2014-004; corresponding collapse-index reductions are 0.199, 0.227, and 0.169. In a 9-subject Cho2017 external subset, EEG-Fusion improves macro-F1 from 0.516 to 0.630. These results suggest that label-free reliability estimation can reduce subject-level failure modes in source-free MI-EEG deployment.

Problem

• Motor Imagery EEG
• Subject-independent decoding
• Source-free adaptation
• Subject shift
• Model collapse

Innovation

• Source-Free Domain Adaptation
• Expert Routing
• Motor Imagery EEG
• Test-Time Adaptation
• Reliability Estimation