CLSP-REQA: A Real-Time Quality-Aware Closed-Loop Seizure Prediction Framework with Mamba-BiLSTM and Confidence-Gated Intervention
Researchers have developed CLSP-REQA, a new framework for real-time seizure prediction that incorporates EEG signal quality assessment. This system uses a Mamba-BiLSTM backbone and a quality estimator to modulate prediction confidence, aiming for more reliable performance in real-world conditions. Tested on the CHB-MIT and SIENA databases, CLSP-REQA demonstrated improved accuracy and generalization capabilities compared to existing methods, even with fewer EEG channels. AI
IMPACT Enhances AI's role in real-time medical diagnostics and closed-loop therapeutic systems.