La 44e Conférence internationale d’ingénierie en médecine et biologie de l’IEEE (EMBC 2022) s’est tenue à Glasgow du 11/7 au 15/7. Victor a présenté un article intitulé « A Saliency Based Feature Fusion Model for EEG Emotion Estimation ». Luca a présenté une affiche intitulée « A Hybrid Framework for ERP Preprocessing in EEG Experiments ».

Titre

A Saliency Based Feature Fusion Model for EEG Emotion Estimation

Auteures

Victor Delvigne, Antoine Facchini, Hazem Wannous, Thierry Dutoit, Laurence Ris, Jean-Philippe Vandeborre

Résumé

Among the different modalities to assess emotion, electroencephalogram (EEG), representing the electrical brain activity, achieved motivating results over the last decade. Emotion estimation from EEG could help in the diagnosis or rehabilitation of certain diseases. In this paper, we propose a dual model considering two different representations of EEG feature maps: 1) a sequential based representation of EEG band power, 2) an image-based representation of the feature vectors. We also propose an innovative method to combine the information based on a saliency analysis of the image-based model to promote joint learning of both model parts. The model has been evaluated on four publicly available datasets: SEED-IV, SEED, DEAP and MPED. The achieved results outperform results from state-of-the-art approaches for three of the proposed datasets with a lower standard deviation that reflects higher stability. For sake of reproducibility, the codes and models proposed in this paper are available at https://github.com/VDelv/Emotion-EEG.

Link: https://arxiv.org/abs/2201.03891

Titre

A Hybrid Framework for ERP Preprocessing in EEG Experiments

Auteures

Luca La Fisca, Bernard Gosselin

Résumé

Methods to derive information on neural processes from Electroencephalographic (EEG) signals become increasingly complex, especially with the introduction of deep learning algorithms. However, considering the low Signal-to-Noise Ratio (SNR) of raw EEG signals, the input data should be properly preprocessed. Common preprocessing algorithms using single method struggle to reduce several types of artifacts/noises without affecting the useful parts of the signal. We therefore propose a hybrid preprocessing framework to combine strengths of multiple state-of-the-art approaches. The latter method provides output signals of better quality than a common Independent Component Analysis (ICA) based pipeline on simulated data. All codes are available at https://github.com/numediart/PreprocEEG

Une actualité sur Twitter: https://twitter.com/IsiaLab/status/1564251495989649409