Considerate motion imagination classification method using deep learning

• Published in: PloS one 2022-10, Vol.17 (10), p.e0276526
• Main Authors: Yan, Zhaokun, Yang, Xiangquan, Jin, Yu
• Format: Article
• Language: English
• Subjects: Ablation; Algorithms; Analysis; Biology and Life Sciences; Brain; Brain-Computer Interfaces; Classification; Computational linguistics; Computer and Information Sciences; Computer applications; Correlation; Deep Learning; EEG; Electrodes; Electroencephalography; Electroencephalography - methods; Emotion recognition; Engineering and Technology; Euclidean geometry; Euclidean space; Feature extraction; Graph representations; Graphical representations; Human-computer interface; Imagination; Imagination (Philosophy); Implants; Interfaces; Language processing; Machine learning; Medicine and Health Sciences; Mental task performance; Methods; Natural language interfaces; Neural networks; Rehabilitation; Research and Analysis Methods; Signal classification; Social Sciences; User interface; Wavelet transforms
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0276526

In order to improve the classification accuracy of motion imagination, a considerate motion imagination classification method using deep learning is proposed. Specifically, based on a graph structure suitable for electroencephalography as input, the proposed model can accurately represent the distribution of electroencephalography electrodes in non-Euclidean space and fully consider the spatial correlation between electrodes. In addition, the spatial-spectral-temporal multi-dimensional feature information was extracted from the spatial-temporal graph representation and spatial-spectral graph representation transformed from the original electroencephalography signal using the dual branch architecture. Finally, the attention mechanism and global feature aggregation module were designed and combined with graph convolution to adaptively capture the dynamic correlation intensity and effective feature of electroencephalography signals in various dimensions. A series of contrast experiments and ablation experiments on several different public brain-computer interface datasets demonstrated that the excellence of proposed method. It is worth mentioning that, the proposed model is a general framework for the classification of electroencephalography signals, which is suitable for emotion recognition, sleep staging and other fields based on electroencephalography research. Moreover, the model has the potential to be applied in the medical field of motion imagination rehabilitation in real life.