Name
Title | ||
|---|---|---|
Discriminatory Features Based on Wavelet Energy for Effective Analysis of Electroencephalogram During Mental Tasks |
Abstract | ||
|---|---|---|
Mental task categorization using single/limited channel(s) electroencephalogram (EEG) signals is crucial for designing portable brain–computer interface and neurofeedback systems. However, EEG signals are corrupted with ocular artifacts and muscle artifacts, which degrade the robustness of existing features resulting inaccurate categorization of mental tasks. Therefore, in this paper, we propose a real-time mental task categorization method using stationary wavelet transform (SWT)-based novel discriminant features and different classifiers. It consists of four major stages: EEG signal decomposition and reconstruction using SWT into subbands; computation of proposed relative subband energy features (i.e., total energy of all subbands( $$E_t$$ ), relative energy of each subband k ( $$E_{r(k)}$$ ) and sum of relative energy pair ratio and difference); significant feature selection; classification using seven classifiers. The robustness of the proposed method is evaluated using two publicly available datasets and in-house dataset recorded using single-channel EEG headset. Performance evaluation results show that the proposed method with support vector machine classifier achieves the highest average subject-dependent accuracy of 98% for classification of mental/mental and baseline/mental tasks. Extensive comparative performance analysis depicts the superiority of the proposed method as compared to existing techniques in terms of accurate categorization of mental tasks, computational complexity, EEG processing length and robustness under artifact-contaminated EEG signals. |
Year | DOI | Venue |
|---|---|---|
2022 | 10.1007/s00034-022-02057-9 | Circuits, Systems, and Signal Processing |
Keywords | DocType | Volume |
Electroencephalogram, Mental task categorization, Feature extraction, Stationary wavelet transform, Relative energy | Journal | 41 |
Issue | ISSN | Citations |
10 | 0278-081X | 0 |
PageRank | References | Authors |
0.34 | 23 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Saini, Manali | 1 | 0 | 0.34 |
| Satija, Udit | 2 | 29 | 3.70 |
| Upadhayay, Madhur Deo | 3 | 0 | 0.34 |