A5005066770- EEG and Brain-Computer Interfaces
- Brain Tumor Detection and Classification
- Emotion and Mood Recognition
- ECG Monitoring and Analysis
- Gaze Tracking and Assistive Technology
- Neural dynamics and brain function
Seoul National University of Science and Technology
2023-2024
Electroencephalography (EEG)-based motor imagery (MI) has potential applications in diverse fields including rehabilitation, drone control, and virtual reality. However, its practical use is hindered by low generalization performance decoding brain signals, primarily due to the subject-dependency of EEG signals. Although multitask autoencoder (MTAE) techniques have recently been used mitigate this issue, these approaches encounter an imbalance problem between loss functions with different...
Emotion recognition has emerged as a active research area, gaining relevance from advancements in deep learning. This study focuses on using electroencephalogram (EEG) data for emotion and addresses the challenge of subject-dependent variability EEG-based by proposing novel architecture that employs multilevel feature fusion multitask autoencoder-based two-phase framework. The first phase generates classspecific data, while second uses these model training. proposed was validated SEED...
Emotion recognition is an emerging technology that employs various types of data modalities to detect and interpret human emotional states. Among them, electroencephalogram (EEG) has gained attention for its ability capture underlying emotions related specific brain activities. In this study, we adopt the attention-temporal convolutional network (ATCNet), previously recognized efficacy in decoding motor imagery EEG data, develop a high-performance emotion method. Our modified ATCNet achieved...
In this paper, we propose a model that combines the multilevel feature fusion algorithm and encoder-decoder structure for evaluation of mental workload using electroencephalogram (EEG) signals. The was used to reduce additive noise subject variations EEG data. encoder is structured by incorporating 3D convolutional neural network (3DCNN) concept, which extracts unified key features from combining low-level high-level features. decoder consists simple 3DCNN layers recover input image latent...