Un marco neuronal multivista con atención para la clasificación de convulsiones epilépticas

Epilepsy is a chronic brain disorder characterized by recurrent seizures due to abnormal neuronal firing. Electroencephalogram (EEG)-based seizure classification has become an important auxiliary tool in clinical practice. This study aims to reduce reliance on expert experience in diagnosis and to improve the automated classification of epileptic seizures using EEG signals.

We propose a novel filter-bank multi-view and attention-based neural network model (FB-AMNet) for seizure classification. The model employs a learnable filter bank to decompose the raw EEG into multiple frequency sub-bands, forming multi-view representations. A multi-branch group convolution network is designed to capture multi-scale frequency-spatial features, while temporal dependencies are extracted through a bidirectional LSTM with an attention mechanism. A shared attention module adaptively emphasizes the most informative sub-bands and time windows for classification.

The proposed model achieves an overall F1 score of 0.7105, a weighted F1 score of 0.8314, and a Cohen's kappa coefficient of 0.6345 on the TUSZ v1.5.2 dataset. Compared with the baseline method FBCNet, the proposed model outperform by 3.22% in overall F1 score (p < 0.05), 1.42% in weighted F1 score (p < 0.05), and 2.87% in Cohen's kappa coefficient (p < 0.05). The best results are also obtained on the CHB-MIT dataset.

These results demonstrate the effectiveness of combining multi-view feature extraction with attention-enhanced temporal modeling.