WAVELET2VEC: A Filter Bank Masked Autoencoder for EEG-Based Seizure Subtype Classification

• Published in: Proceedings of the ... IEEE International Conference on Acoustics, Speech and Signal Processing (1998) pp. 1 - 5
• Main Authors: Peng, Ruimin, Zhao, Changming, Xu, Yifan, Jiang, Jun, Kuang, Guangtao, Shao, Jianbo, Wu, Dongrui
• Format: Conference Proceeding
• Language: English
• Published: IEEE 04.06.2023
• Subjects: Brain modeling; Deep learning; EEG; filter bank analysis; Filter banks; masked autoencoder; seizure subtype classification; Self-supervised learning; Training data; Transformers; Wavelet transforms
• ISSN: 2379-190X
• DOI: 10.1109/ICASSP49357.2023.10097183

Electroencephalogram (EEG) based seizure subtype classification plays an important role in clinical diagnostics. However, existing deep learning approaches face two challenges in such applications: 1) convolutional or recurrent neural network based models have difficulty learning long-term dependencies; and, 2) there are not enough labeled seizure sub-type data for training such models. This paper proposes a Transformer-based self-supervised learning model for EEG-based seizure subtype classification, which copes well with these two challenges. Filter bank analysis is first employed to improve Vision Transformer as a Wavelet Transformer (WaT) encoder, which generates multi-grained feature representations of EEG signals. Then, self-supervised learning is used to pre-train WaT from unlabeled EEG data. Experiments on two public datasets demonstrated that Wavelet2Vec outperformed several other supervised and self-supervised models in cross-subject seizure subtype classification.