English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

With the recent success of artificial intelligence in neuroscience, a numberof deep learning (DL) models were proposed for classification, anomalydetection, and pattern recognition tasks in electroencephalography (EEG). EEGis a multi-channel time-series that provides information about the individualbrain activity for diagnostics, neuro-rehabilitation, and other applications(including emotions recognition). Two main issues challenge the existingDL-based modeling methods for EEG: the high variability between subjects andthe low signal-to-noise ratio making it difficult to ensure a good quality inthe EEG data. In this paper, we propose two variational autoencoder models,namely vEEGNet-ver3 and hvEEGNet, to target the problem of high-fidelity EEGreconstruction. We properly designed their architectures using the blocks ofthe well-known EEGNet as the encoder, and proposed a loss function based ondynamic time warping. We tested the models on the public Dataset 2a - BCICompetition IV, where EEG was collected from 9 subjects and 22 channels.hvEEGNet was found to reconstruct the EEG data with very high-fidelity,outperforming most previous solutions (including our vEEGNet-ver3 ).Furthermore, this was consistent across all subjects. Interestingly, hvEEGNetmade it possible to discover that this popular dataset includes a number ofcorrupted EEG recordings that might have influenced previous literatureresults. We also investigated the training behaviour of our models and relatedit with the quality and the size of the input EEG dataset, aiming at opening anew research debate on this relationship. In the future, hvEEGNet could be usedas anomaly (e.g., artefact) detector in large EEG datasets to support thedomain experts, but also the latent representations it provides could be usedin other classification problems and EEG data generation.

hvEEGNet: exploiting hierarchical VAEs on EEG data for neuroscience applications