Multiscale time-frequency method for multiclass Motor Imagery Brain Computer Interface

• Published in: Computers in biology and medicine Vol. 143; p. 105299
• Main Authors: Liu, Guoyang, Tian, Lan, Zhou, Weidong
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.04.2022; Elsevier Limited
• Subjects: Accuracy; Algorithms; Brain; Brain computer interface; Classification; Computational neuroscience; Computer applications; Datasets; Deep learning; Divergence; Divergence framework; EEG; Feature extraction; Feature selection; Human-computer interface; Image segmentation; Implants; Internal Medicine; Mental task performance; Methods; Multiclass motor imagery; Neural networks; Neurology; Other; Rehabilitation; Time-frequency representation; Brain computer interface; Time-frequency representation; Divergence framework; Multiclass motor imagery
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2022.105299

Motor Imagery Brain Computer Interface (MI-BCI) has become a promising technology in the field of neurorehabilitation. However, the performance and computational complexity of the current multiclass MI-BCI have not been fully optimized, and the intuitive interpretation of individual differences on motor imagery tasks is seldom investigated. In this paper, a well-designed multiscale time-frequency segmentation scheme is first applied to multichannel EEG recordings to obtain Time-Frequency Segments (TFSs). Then, the TFS selection based on a specific wrapper feature selection rule is utilized to determine optimum TFSs. Next, One-Versus-One (OvO)-divCSP implemented in divergence framework is used to extract discriminative features. Finally, One-Versus-Rest (OvR)-SVM is utilized to predict the class label based on selected multiclass MI features. Experimental results indicate our method yields a superior performance on two publicly available multiclass MI datasets with a mean accuracy of 80.00% and a mean kappa of 0.73. Meanwhile, the proposed TFS selection method can significantly alleviate the computational burden with little accuracy reduction, demonstrating the feasibility of real-time multiclass MI-BCI. Furthermore, the Motor Imagery Time-Frequency Reaction Map (MI-TFRM) is visualized, contributing to analyzing and interpreting the performance differences between different subjects. •A multiscale segmentation scheme in time and frequency domains is introduced•A wrapper-based time-frequency segment selection approach is proposed•Ideal results can be reached in multiclass MI-BCI with low computational complexity•Visualize the subject-specific motor imagery time-frequency reaction map