Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 6, 2023Frontiers in NeuroscienceOpen Access

The proposed spatial-temporal transformer model achieved an accuracy of 97.3% for binary emotion classification and 97.1% for ternary emotion classification from multi-channel EEG signals.

View Full Paper
Ask AI
Bookmark
Share

Why the study?

A key challenge with automated emotion recognition in EEG signals is extracting and selecting discriminating features to classify different emotions accurately.

Does a spatial-temporal transformer model improve emotion classification accuracy from multi-channel EEG signals compared to state-of-the-art techniques?

Population

20 subjects during listening to music

Comparison

Proposed spatial-temporal Transformer model vs state-of-the-art techniques

Design

Comparative deep learning model evaluation study

Key result

The proposed spatial-temporal transformer model achieved an accuracy of 97.3% for binary emotion classification and 97.1% for ternary emotion classification from multi-channel EEG signals.

Authors

YZYanan ZhouJLJian Lian

Discussion

Loading...

Member takes

Overview

May advance EEG emotion classification algorithms; leaves open clinical validation before diagnostic use in emotional disorders.

Structured PICO

Does a spatial-temporal transformer model improve emotion classification accuracy from multi-channel EEG signals compared to state-of-the-art techniques?

P
Population
32 healthy university students aged 18-31 without mental illness or brain damage participated in an EEG study to classify emotions evoked by music.
I
Intervention
Spatial-temporal transformer model for identifying emotions from multi-channel EEG signals while listening to music.
C
Comparator
State-of-the-art deep learning methods (e.g., U-Net, Mask R-CNN, ViT, Swin Transformer).
O
Outcome
Accuracy of binary (positive/negative) and ternary (positive/negative/neutral) emotion classification.surrogate

A novel spatial-temporal transformer model achieved high accuracy (>97%) in classifying emotions from multi-channel EEG signals, outperforming existing deep learning methods.

Limitations

  • Used a private dataset in the primary experiments.
  • Weighting parameters were selected using a trial-and-error strategy, which lacks interpretability.
  • The training process of the proposed approach was time-consuming.
  • Requires lightweight parameter configuration to decrease computation resources and execution time for practical applications.

Cite This Study

Zhou et al. (2023) studied Healthy (Emotion recognition) (n=32). Spatial-temporal transformer model vs. State-of-the-art deep learning techniques was evaluated on Accuracy of binary emotion classification (positive vs. negative). The proposed spatial-temporal transformer model achieved an accuracy of 97.3% for binary emotion classification and 97.1% for ternary emotion classification from multi-channel EEG signals.

synapsesocial.com/papers/6a6f4de02163a0a01bc39fc0https://doi.org/10.3389/fnins.2023.1188696
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 3 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Evaluating the three-network theory of creativity: Effects of music listening on resting state EEG2022 · 10 citations
  2. 2Music and emotion: Electrophysiological correlates of the processing of pleasant and unpleasant music2007 · 612 citations
  3. 3Music and emotions: from enchantment to entrainment2015 · 234 citations