
Brain Sciences, Journal Year: 2025, Volume and Issue: 15(3), P. 220 - 220
Published: Feb. 20, 2025
Background/Objectives: This systematic review presents how neural and emotional networks are integrated into EEG-based emotion recognition, bridging the gap between cognitive neuroscience practical applications. Methods: Following PRISMA, 64 studies were reviewed that outlined latest feature extraction classification developments using deep learning models such as CNNs RNNs. Results: Indeed, findings showed multimodal approaches practical, especially combinations involving EEG with physiological signals, thus improving accuracy of classification, even surpassing 90% in some studies. Key signal processing techniques used during this process include spectral features, connectivity analysis, frontal asymmetry detection, which helped enhance performance recognition. Despite these advances, challenges remain more significant real-time processing, where a trade-off computational efficiency limits implementation. High cost is prohibitive to use real-world applications, therefore indicating need for development application optimization techniques. Aside from this, obstacles inconsistency labeling emotions, variation experimental protocols, non-standardized datasets regarding generalizability recognition systems. Discussion: These developing adaptive, algorithms, integrating other inputs like facial expressions sensors, standardized protocols elicitation classification. Further, related ethical issues respect privacy, data security, machine model biases be much proclaimed responsibly apply research on emotions areas healthcare, human–computer interaction, marketing. Conclusions: provides critical insight suggestions further field toward robust, scalable, applications by consolidating current methodologies identifying their key limitations.
Language: Английский