TFFBN-HDLF: Hybrid Deep Learning Framework for Epileptic Seizure Detection Using Time-Frequency Functional Brain Networks

This study presents a hybrid deep learning framework combining time-frequency functional brain networks with CNN and Transformer modules to improve epilepsy seizure detection in elderly patients from EEG signals. The method uses Pearson correlation and phase lag index to construct brain networks that capture synchronous neural interactions, converting complex EEG data into clinically useful features for automated seizure detection.