Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Resistance analysis of the Tobacco Variety SA1214 against M. incognita based on physiological and biochemical assays.

BMC plant biology·2026
Same author

Differentiating high-frequency and high-severity hotspots: A robust risk-evolution-volume (REV) framework.

Accident; analysis and prevention·2026
Same author

[Mechanisms and applications of magnesium ion-regulated stem cell functions in promoting tendon-bone interface healing].

Sheng wu yi xue gong cheng xue za zhi = Journal of biomedical engineering = Shengwu yixue gongchengxue zazhi·2026
Same author

[Clinical and genetic analysis of a Chinese pedigree affected with MRXS34 syndrome due to variant of NONO gene].

Zhonghua yi xue yi chuan xue za zhi = Zhonghua yixue yichuanxue zazhi = Chinese journal of medical genetics·2026
Same author

Structure and antioxidant characteristics of diacylglycerol-based oleogels: A comparison with triacylglycerol systems.

Food chemistry·2026
Same author

Gut-brain axis crosstalk in patients with Alzheimer's disease and high body mass index: Gut dysbiosis, blood-brain barrier disruption, and neuroinflammation.

Journal of Alzheimer's disease : JAD·2026

Related Experiment Video

Updated: Jan 15, 2026

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.3K

LMSA-net: a lightweight multi-scale attention network for eeg-based emotion recognition.

Hao Yue1,2, Hengrui Ruan1,2, Yawu Zhao3

  • 1College of Computer Science and Technology, China University of Petroleum (East China), Qingdao 266580, People's Republic of China.

Biomedical Physics & Engineering Express
|January 13, 2026
PubMed
Summary

This study introduces LMSA-Net, a lightweight model for emotion recognition using electroencephalogram (EEG) signals. It achieves high accuracy by learning spatial-temporal features directly from raw EEG data, enabling practical applications.

Keywords:
BCIattention mechanismsconvolutional neural networkelectroencephalogramemotion recognitionlightweight

More Related Videos

Applications of EEG Neuroimaging Data: Event-related Potentials, Spectral Power, and Multiscale Entropy
11:15

Applications of EEG Neuroimaging Data: Event-related Potentials, Spectral Power, and Multiscale Entropy

Published on: June 27, 2013

34.4K
Author Spotlight: Advancing Large-Scale Neural Dynamics Through HD-MEA Technology
09:44

Author Spotlight: Advancing Large-Scale Neural Dynamics Through HD-MEA Technology

Published on: March 8, 2024

5.8K

Related Experiment Videos

Last Updated: Jan 15, 2026

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.3K
Applications of EEG Neuroimaging Data: Event-related Potentials, Spectral Power, and Multiscale Entropy
11:15

Applications of EEG Neuroimaging Data: Event-related Potentials, Spectral Power, and Multiscale Entropy

Published on: June 27, 2013

34.4K
Author Spotlight: Advancing Large-Scale Neural Dynamics Through HD-MEA Technology
09:44

Author Spotlight: Advancing Large-Scale Neural Dynamics Through HD-MEA Technology

Published on: March 8, 2024

5.8K

Area of Science:

  • Affective Computing
  • Neuroscience
  • Machine Learning

Background:

  • Electroencephalogram (EEG) signals are crucial for emotion recognition but are challenging due to non-stationarity and noise.
  • Existing end-to-end models face difficulties in achieving high performance with lightweight architectures for practical deployment.
  • Manual feature engineering in EEG analysis is complex and time-consuming.

Purpose of the Study:

  • To develop a lightweight, interpretable, and end-to-end model for direct spatio-temporal feature learning from raw EEG signals.
  • To enhance the performance and efficiency of EEG-based emotion recognition systems.
  • To enable practical deployment of emotion recognition technology on edge devices.

Main Methods:

  • Proposed LMSA-Net (Lightweight Multi-Scale Attention Network) architecture.
  • Integration of learnable channel weighting for adaptive spatial encoding.
  • Utilized multi-scale temporal separable convolution for rhythm-specific feature extraction.
  • Incorporated Sim Attention Module for parameter-free saliency enhancement.

Main Results:

  • Achieved top performance on the SEED dataset (65.53% accuracy) and competitive results on SEED-IV (48.52% accuracy).
  • Demonstrated strong performance in arousal classification on the DEAP dataset, indicating good generalization.
  • Confirmed the effectiveness of each module through ablation studies and frequency analysis, showing specialization in EEG rhythms.
  • Showcased a lightweight design with minimal parameters (7.64K) and low latency, suitable for edge deployment.

Conclusions:

  • LMSA-Net offers an efficient, interpretable, and high-performing solution for EEG-based emotion recognition.
  • The model's design aligns with neurophysiological principles, extracting rhythm-specific features.
  • The lightweight and interpretable nature of LMSA-Net facilitates practical applications in affective computing and human-computer interaction.