Time window and frequency band optimization using regularized neighbourhood component analysis for Multi-View Motor Imagery EEG classification

• Published in: Biomedical signal processing and control Vol. 67; p. 102550
• Main Authors: Singh Malan, Nitesh, Sharma, Shiru
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.05.2021
• Subjects: Brain-computer interface; Dual-tree complex wavelet transform; Electroencephalogram; Motor imagery; Neighbourhood component analysis; Brain-computer interface; Motor imagery; Electroencephalogram; Neighbourhood component analysis; Dual-tree complex wavelet transform
• ISSN: 1746-8094; 1746-8108
• DOI: 10.1016/j.bspc.2021.102550

•A novel multi-view feature selection to optimize time windows and frequency bands.•Proposed method preserves structure of multi-view EEG.•Neural response to motor imagery task is subject-specific.•Obtained classification accuracy 82.1 %, 91.7 %, and 84.5 % for three BCI datasets. Spatial features optimized at frequency bands have been widely used in motor imagery (MI) based brain-computer interface (BCI) systems. However, using a fixed time window of electroencephalogram (EEG) to extract discriminatory features results in suboptimal MI classification performance because time latency during MI tasks is inconsistent between different subjects. Thus, apart from frequency band optimization, time window optimization is equally important to develop a subject-specific MI-BCI. With time windows, extracted feature space becomes a higher-order tensor problem that requires multi-view learning approaches to optimize features. This study proposes a novel multi-view feature selection method based on regularized neighbourhood component analysis to simultaneously optimize time windows and frequency bands. In the experiment, we extracted spatial features using common spatial patterns (CSP) from MI related EEG data at multiple time windows and frequency bands and optimized them using the proposed feature selection method. A support vector machine is trained to classify optimized CSP features to identify MI tasks. The proposed method achieved classification accuracies on three public BCI datasets (BCI competition IV dataset 2a, BCI competition III dataset IIIa, and BCI competition IV dataset 2b), which are 82.1 %, 91.7 %, and 84.5 %, respectively. Obtained results are superior to those obtained using standard competing algorithms. Hence, the proposed multi-view learning approach for simultaneous optimization of time windows and frequency bands of MI signals shows the potential to enhance a practical MI BCI device's performance.