hvEEGNet: a novel deep learning model for high-fidelity EEG reconstruction

• Published in: Frontiers in neuroinformatics Vol. 18; p. 1459970
• Main Authors: Cisotto, Giulia, Zancanaro, Alberto, Zoppis, Italo F., Manzoni, Sara L.
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Research Foundation 20.12.2024; Frontiers Media S.A
• Subjects: Accuracy; Classification; Convulsions & seizures; Datasets; Deep learning; EEG; Electrocardiography; Electroencephalography; Epilepsy; latent representation; motor imagery; Neural networks; VAE; variational autoencoder; motor imagery; VAE; variational autoencoder; EEG; latent representation
• ISSN: 1662-5196; 1662-5196
• DOI: 10.3389/fninf.2024.1459970

Modeling multi-channel electroencephalographic (EEG) time-series is a challenging tasks, even for the most recent deep learning approaches. Particularly, in this work, we targeted our efforts to the high-fidelity reconstruction of this type of data, as this is of key relevance for several applications such as classification, anomaly detection, automatic labeling, and brain-computer interfaces. We analyzed the most recent works finding that high-fidelity reconstruction is seriously challenged by the complex dynamics of the EEG signals and the large inter-subject variability. So far, previous works provided good results in either high-fidelity reconstruction of single-channel signals, or poor-quality reconstruction of multi-channel datasets. Therefore, in this paper, we present a novel deep learning model, called hvEEGNet, designed as a hierarchical variational autoencoder and trained with a new loss function. We tested it on the benchmark Dataset 2a (including 22-channel EEG data from 9 subjects). We show that it is able to reconstruct all EEG channels with high-fidelity, fastly (in a few tens of epochs), and with high consistency across different subjects. We also investigated the relationship between reconstruction fidelity and the training duration and, using hvEEGNet as an anomaly detector, we spotted some data in the benchmark dataset that are corrupted and never highlighted before. Thus, hvEEGNet could be very useful in several applications where automatic labeling of large EEG dataset is needed and time-consuming. At the same time, this work opens new fundamental research questions about (1) the effectiveness of deep learning models training (for EEG data) and (2) the need for a systematic characterization of the input EEG data to ensure robust modeling.