Razzak Imran, A Hameed Ibrahim, Xu Guandong
1University of TechnologySydneyNSW2007Australia.
2Norwegian University of Science and Technology7491TrondheimNorway.
IEEE J Transl Eng Health Med. 2019 Oct 2;7:2000508. doi: 10.1109/JTEHM.2019.2942017. eCollection 2019.
EEG signals are extremely complex in comparison to other biomedical signals, thus require an efficient feature selection as well as classification approach. Traditional feature extraction and classification methods require to reshape the data into vectors that results in losing the structural information exist in the original featured matrix.
The aim of this work is to design an efficient approach for robust feature extraction and classification for the classification of EEG signals.
In order to extract robust feature matrix and reduce the dimensionality of from original epileptic EEG data, in this paper, we have applied robust joint sparse PCA (RJSPCA), Outliers Robust PCA (ORPCA) and compare their performance with different matrix base feature extraction methods, followed by classification through support matrix machine. The combination of joint sparse PCA with robust support matrix machine showed good generalization performance for classification of EEG data due to their convex optimization.
A comprehensive experimental study on the publicly available EEG datasets is carried out to validate the robustness of the proposed approach against outliers.
The experiment results, supported by the theoretical analysis and statistical test, show the effectiveness of the proposed framework for solving classification of EEG signals.
与其他生物医学信号相比，脑电图（EEG）信号极其复杂，因此需要高效的特征选择和分类方法。传统的特征提取和分类方法需要将数据重塑为向量，这会导致丢失原始特征矩阵中存在的结构信息。
本研究的目的是设计一种高效的方法，用于对EEG信号进行稳健的特征提取和分类。
为了从原始癫痫性EEG数据中提取稳健的特征矩阵并降低其维度，本文应用了稳健联合稀疏主成分分析（RJSPCA）、离群点稳健主成分分析（ORPCA），并将它们的性能与不同的基于矩阵的特征提取方法进行比较，随后通过支持矩阵机进行分类。联合稀疏主成分分析与稳健支持矩阵机的结合因其凸优化特性，在EEG数据分类中表现出良好的泛化性能。
对公开可用的EEG数据集进行了全面的实验研究，以验证所提方法对离群点的稳健性。
实验结果在理论分析和统计检验的支持下，表明了所提框架在解决EEG信号分类问题上的有效性。
