Background: In the realm of brain-computer interfaces (BCI), identifying emotions from electroencephalogram (EEG) data is a difficult endeavor because of the volume of data, the intricacy of the signals, and the several channels that make up the signals.
New Methods: Using dual-stream structure scaling and multiple attention mechanisms (LDMGEEG), a lightweight network is provided to maximize the accuracy and performance of EEG-based emotion identification. Reducing the number of computational parameters while maintaining the current level of classification accuracy is the aim. This network employs a symmetric dual-stream architecture to assess separately time-domain and frequency-domain spatio-temporal maps constructed using differential entropy features of EEG signals as inputs.
Result: The experimental results show that after significantly lowering the number of parameters, the model achieved the best possible performance in the field, with a 95.18 % accuracy on the SEED dataset.
Comparison With Existing Methods: Moreover, it reduced the number of parameters by 98 % when compared to existing models.
Conclusion: The proposed method distinct channel-time/frequency-space multiple attention and post-attention methods enhance the model's ability to aggregate features and result in lightweight performance.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1016/j.jneumeth.2024.110223 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Mechanisms and new advances in the efficacy of plant active ingredients in tendon-bone healing.
J Orthop Surg Res
January 2025
The First Affiliated Hospital of Baotou Medical College, Inner Mongolia University of Science and Technology, No.41 Linyin Road, Baotou, Inner Mongolia, 014010, China.
The tendon-bone interface, known as the tenosynovial union or attachment, can be easily damaged by excessive exercise or trauma. Tendon-bone healing is a significant research topic in orthopedics, encompassing various aspects of sports injuries and postoperative recovery. Surgery is the most common treatment; however, it has limited efficacy in promoting tendon-bone healing and carries a risk of postoperative recurrence, necessitating the search for more effective treatments.
View Article and Find Full Text PDFMolecular and cellular characteristics of cerebrovascular cell types and their contribution to neurodegenerative diseases.
Mol Neurodegener
January 2025
The Picower Institute for Learning and Memory, Cambridge, MA, USA.
Many diseases and disorders of the nervous system suffer from a lack of adequate therapeutics to halt or slow disease progression, and to this day, no cure exists for any of the fatal neurodegenerative diseases. In part this is due to the incredible diversity of cell types that comprise the brain, knowledge gaps in understanding basic mechanisms of disease, as well as a lack of reliable strategies for delivering new therapeutic modalities to affected areas. With the advent of single cell genomics, it is now possible to interrogate the molecular characteristics of diverse cell populations and their alterations in diseased states.
View Article and Find Full Text PDFEcofriendly remediation of cadmium, lead, and zinc using dead cells of Microcystis aeruginosa.
Sci Rep
January 2025
Hydrobiology Lab, National Institute of Oceanography and Fisheries (NIOF), Cairo, Egypt.
The utilization of cyanobacteria toxin-producing blooms for metal ions adsorption has garnered significant attention over the last decade. This study investigates the efficacy of dead cells from Microcystis aeruginosa blooms, collected from agricultural drainage water reservoir, in removing of cadmium, lead, and zinc ions from aqueous solutions, and simultaneously addressing the mitigation of toxin-producing M. aeruginosa bloom.
View Article and Find Full Text PDFEnhanced hybrid attention deep learning for avocado ripeness classification on resource constrained devices.
Sci Rep
January 2025
Faculty of Science and Technology, Suan Sunandha Rajabhat University, Bangkok, 10300, Thailand.
Attention mechanisms such as the Convolutional Block Attention Module (CBAM) can help emphasize and refine the most relevant feature maps such as color, texture, spots, and wrinkle variations for the avocado ripeness classification. However, the CBAM lacks global context awareness, which may prevent it from capturing long-range dependencies or global patterns such as relationships between distant regions in the image. Further, more complex neural networks can improve model performance but at the cost of increasing the number of layers and train parameters, which may not be suitable for resource constrained devices.
View Article and Find Full Text PDFHearing impairment (HI) disrupts social interaction by hindering the ability to follow conversations in noisy environments. While hearing aids (HAs) with noise reduction (NR) partially address this, the "cocktailparty problem" persists, where individuals struggle to attend to specific voices amidst background noise. This study investigated how NR and an advanced signal processing method for compensating for nonlinearities in EEG signals can improve neural speech processing in HI listeners.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!