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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Purpose:  Emotion is the reflection of individual's perception and understanding of various things, which needs the synergy of various brain regions. A large number of emotion decoding methods based on electroencephalogram (EEG) have been proposed. But extracting the most discriminative and cognitive features to construct a model is yet to be determined. This paper aims to construct a model that can extract the most discriminative and cognitive features. Materials and methods:  Here, we collected EEG signals from 24 subjects in a musical emotion induction experiment. Then, an end-to-end branch LSTM-CNN (BLCNN) was used to extract emotion features from the laboratory dataset and DEAP dataset for emotion decoding. Finally, the extracted features were visualized on the laboratory dataset using saliency map. Result:  The classification results showed that the accuracy of the three classification of the laboratory dataset was 95.78% ± 1.70%, and the accuracy of the four classification of the DEAP dataset was 80.97% ± 7.99%. We found that the discriminating features of positive emotion were distributed in the left hemisphere, at the same time, negative emotion features were distributed in the right hemisphere, where mainly in the frontal, parietal and occipital lobes. Conclusion:  In this paper, we proposed a neural network model, namely BLCNN. The model obtained good results in laboratory dataset and DEAP dataset. Through the visual analysis of the features extracted by BLCNN, it was found that the features were consistent with emotional cognition. Therefore, this paper provided a new perspective for the practical application of human-computer emotional interaction.

Deep learning with convolutional neural networks for EEG-based music emotion decoding and visualization