Automated diagnosis of epileptic electroencephalogram using independent component analysis and discrete wavelet transform for different electroencephalogram durations

Abstract

Epilepsy is a disorder of the brain depicted by recurrent seizures. Electroencephalogram signals can be used to study the characteristics of epileptic seizures. In this study, we propose a method for the automated classification of electroencephalogram into normal, interictal and ictal classes using 6, 12, 18 and 23.6 s of data. We employed discrete wavelet transform to decompose electroencephalogram signals into frequency sub-bands. These discrete wavelet transform coefficients were then subjected to independent component analysis for reducing the data dimension. The independent component analysis features were then fed to six classifiers, namely, decision tree, K-nearest neighbor, probabilistic neural network, fuzzy, Gaussian mixture model and support vector machine to select the best classifier. We observed that the support vector machine classifier with radial basis function kernel function gave the best results with an average accuracy of 96%, sensitivity of 96% and specificity of 97% for 23.6 s of electroencephalogram data. Our results show that as the duration of the data increases, the classification accuracy increases. This proposed technique can be used as an automatic seizure monitoring software to aid the doctors in providing timely quality care for the patients suffering from epilepsy.

Deep learning-based EEG emotion recognition: Current trends and future perspectives

“…The feature extraction methods mainly include time domain analysis, frequency domain analysis, time frequency domain analysis, multivariate statistical analysis, and nonlinear dynamic analysis (Donos et al, 2018;Rasheed et al, 2020). Principal component analysis, linear discriminant analysis (LDA; Alotaiby et al, 2017), and independent component analysis (Ur et al, 2013;Acharya et al, 2018a;Maimaiti et al, 2021) are widely used unsupervised time-domain methods to summarize EEG data. Frequency domain features include spectral center, coefficient of variation, power spectral density, signal energy, spectral moment, and spectral skewness, which can provide key information about data changes (Yuan et al, 2018;Acharya et al, 2018a).…”

“…Wavelet transformation (WT) is usually used to decompose EEG signals into their frequency components to express the relationship between signal information and time. Time frequency signal processing algorithms, such as discrete wavelet transform analysis and continuous wavelet transform, are a necessary means to solve different EEG behavior, which can be described in the time and frequency domains (Martis et al, 2012;UR et al, 2013). Statistical parameters, such as mean, variance, skewness, and kurtosis, have been widely used to extract feature information from EEG signals.…”

Epileptic-Seizure onset detection using PARAFAC model with Cross-Wavelet Transformation on Multi-Channel EEG

“…Their method yielded a sensitivity of 81.08% and specificity of 82.23%. Acharya in 2013 [9], employed discrete wavelet transform based methods and independent component analysis for reducing the data dimension and effectively extract features which were then fed to classifiers to develop an automated classifier for diagnosing epileptic seizures. The proposed techniques had high accuracy and sensitivity for even short data duration.…”

Balanced Communication-Avoiding Support Vector Machine when Detecting Epilepsy based on EEG Signals

“…Authors in [11][12][13][14][15][16][17] applied the Gaussian Mixture Model to the EEG signals. An accuracy of 93.1% was reached in [11].…”

“…An accuracy of 93.1% was reached in [11]. In [12][13][14], the authors used the Support Vector Machine as the main idea. The wavelet transformer obtained an accuracy of 96.3%, the recurrence quantification analysis achieved an accuracy of 95.6%, and the discrete wavelet transformation an accuracy of 96%.…”