Motor imagery BCI classification based on novel two‐dimensional modelling in empirical wavelet transform. Issue 25 (6th October 2020)

• Record Type: Journal Article
• Authors: Sadiq, Muhammad Tariq; Yu, Xiaojun; Yuan, Zhaohui; Aziz, Muhammad Zulkifal
• Is Part Of: Electronics letters. Volume 56:Issue 25(2020)
• Page Start: 1367
• Page End: 1369
• Publication Date: 2020-10-06
• Subjects: signal classification -- feedforward neural nets -- signal denoising -- feature extraction -- brain‐computer interfaces -- electroencephalography -- principal component analysis -- neural nets -- wavelet transforms -- medical signal processing; classification check -- total classification accuracy -- motor imagery BCI classification -- empirical wavelet -- brain complexity -- nonstationary nature -- electroencephalography signal -- considerable challenges -- different motor‐imagery tasks -- brain–computer interface -- automated accurate classification -- MI tasks -- raw EEG signals -- multiscale principal component analysis -- denoised signals -- single geometrical feature name -- extracted feature vectors -- feedforward neural network -- cascade forward neural networks; Electronics -- Periodicals; 621.381
• DOI: 10.1049/el.2020.2509
• Languages: English
• ISSNs: 0013-5194

Abstract

Abstract : Brain complexity and non‐stationary nature of electroencephalography (EEG) signal make considerable challenges for the accurate identification of different motor‐imagery (MI) tasks in brain–computer interface (BCI). In the proposed Letter, a novel framework is proposed for the automated accurate classification of MI tasks. First, raw EEG signals are denoised with multiscale principal component analysis. Secondly, denoised signals are decomposed by empirical wavelet transform into different modes. Thirdly, the two‐dimensional (2D) modelling of modes is introduced to identify the variations of different signals. Fourthly, a single geometrical feature name as, a summation of distance from each point relative to a coordinate centre is extracted from 2D modelling of modes. Finally, the extracted feature vectors are provided to the feedforward neural network and cascade forward neural networks for classification check. The proposed study achieved 95.3% of total classification accuracy with 100% outcome for subject with very small training samples, which is outperforming existing methods on the same database.