CLSP-REQA: A Real-Time Quality-Aware Closed-Loop Seizure Prediction Framework with Mamba-BiLSTM and Confidence-Gated Intervention
For clinicians and patients relying on closed-loop neurostimulation, this work provides a more realistic and robust seizure prediction framework that accounts for real-world EEG quality variability.
CLSP-REQA integrates a real-time EEG quality estimator with a Mamba-BiLSTM predictor to improve seizure prediction under variable signal quality, achieving AUC-ROC of 0.7426 on CHB-MIT and 0.7012 on SIENA, outperforming prior cross-patient baselines by 5-9% with fewer channels and no domain adaptation.
Reliable seizure prediction is a prerequisite for closed-loop neurostimulation therapy, yet existing methods rarely account for the variability in EEG signal quality encountered in real-world deployment, and the overwhelming majority adopt non-strict evaluation protocols that overestimate generalisation performance. We propose CLSP-REQA (Closed-Loop Seizure Prediction with Real-time EEG Quality Assessment), a unified framework that embeds a lightweight signal quality estimator directly within the prediction pipeline. A Real-time EEG Quality Assessment (REQA) module runs in parallel with a Mamba-BiLSTM backbone, producing a scalar quality score q in [0,1] that modulates output confidence through a tiered non-linear fusion function (ECLO). Under strict cross-patient evaluation on the CHB-MIT Scalp EEG Database (n = 23 subjects, 198 seizures), CLSP-REQA achieves an AUC-ROC of 0.7426 +- 0.0199, outperforming the unadapted cross-patient baseline of 0.69 reported by Jemal et al., using only 16 EEG channels compared to 23 in prior work, and without requiring any target-patient data or domain adaptation. On the SIENA Scalp EEG Database (n = 14 subjects, 47 seizures), CLSP-REQA achieves AUC 0.7012 +- 0.0249, substantially surpassing the best domain-adapted cross-patient result of 0.61 on the same dataset, demonstrating strong cross-dataset generalisation. The framework outputs a structured four-tuple (p, q, c, Phi_SHAP) directly compatible with closed-loop neurostimulator interfaces.