Automated eye blink artefact removal from EEG using support vector machine and autoencoder

Electroencephalography (EEG) has been used extensively for classifying cognitive tasks. Many investigators have demonstrated classification accuracies well over 90% for some combinations of cognitive tasks, signal transformations, and classification methods. Unfortunately, EEG data is prone to significant interference from a wide variety of artifacts, particularly eye blinks. Most methods for classifying cognitive tasks with EEG data simply discard time windows containing eye blink artifacts. However, future applications of EEG-based cognitive task classification should not be hindered by eye blinks. The value of an EEG-controlled human-computer interface, for instance, would be severely diluted if it did not work in the presence of eye blinks. Fortunately, recent advances in blind signal separation algorithms and their applications to EEG data mitigate the artifact contamination issue. In this paper, we show how independent components analysis (ICA) and its extension for sub-Gaussian sources, extended ICA (eICA), can be applied to accurately classify cognitive tasks with eye blink contaminated EEG recordings.

Electroencephalogram (EEG) plays an important role in E-healthcare systems, especially in the mental healthcare area, where constant and unobtrusive monitoring is desirable. In the context of OPTIMI project, a novel, low cost, and light weight wearable EEG sensor has been designed and produced. In order to improve the performance and reliability of EEG sensors in real-life settings, we propose a method to evaluate the quality of EEG signals, based on which users can easily adjust the connection between electrodes and their skin. Our method helps to filter invalid EEG data from personal trials in both domestic and office settings. We then apply an algorithm based on Discrete Wavelet Transformation (DWT) and Adaptive Noise Cancellation (ANC) which has been designed to remove ocular artifacts (OA) from the EEG signal. DWT is applied to obtain a reconstructed OA signal as a reference while ANC, based on recursive least squares, is used to remove the OA from the original EEG data. The newly produced sensors were tested and deployed within the OPTIMI framework for chronic stress detection. EEG nonlinear dynamics features and frontal asymmetry of theta, alpha, and beta bands have been selected as biological indicators for chronic stress, showing relative greater right anterior EEG data activity in stressful individuals. Evaluation results demonstrate that our EEG sensor and data processing algorithms have successfully addressed the requirements and challenges of a portable system for patient monitoring, as envisioned by the EU OPTIMI project.

Detection of eye blink artifacts from single prefrontal channel electroencephalogram

BACKGROUND: Electroencephalogram (EEG) signals are sometimes contaminated by cardiac artifacts (CAs). The artifacts resulted by electrical activities of heart, named electrocardiogram (ECG), appear in EEG recordings as spiky potentials that may obscure the information in EEG data and reduce their interpretability. OBJECTIVE: Real-time removal of CAs is of great importance in several applications of EEG, and particularly brain-computer interface (BCI). The process is, however, often neglected due to the time-consuming computations. METHODS: This paper applies a new real-time hybrid approach to remove ECG artifacts from EEG signals. The method is based on the combination of independent component analysis (ICA) and adaptive noise cancellation (ANC), referred to as ICA-ANC. ICA is applied to a few EEG signals in order to extract the reference signal for ANC. The method so utilizes a few EEG channels without synchronous ECG channel, and thus is suited to portable BCI applications. RESULTS: ICA-ANC is evaluated for datasets of five different subjects. CAs are efficiently removed while preserving the cerebral information. The approach is shown to outperform a state of the art method. CONCLUSION: The proposed new algorithm is capable of real-time cardiac artifacts removal using a few EEG channels.

In this study, normal (healthy), pre-seizure and seizure states were analyzed using EEG data from BONN database. In this paper, we propose a new method of using SVM, KNN and Decision Tree for classification analysis in order to improve the detection accuracy of seizure. First, due to the presence of significant noise in the EEG signals, for signal pre-processing we performed noise removal. Second, two methods of frequency space transformation were used such as Discrete Wavelet Transform (DWT), separating signal into sub-bands and Empirical Mode Decomposition (EMD) technique to decompose signal into the Intrinsic Mode Functions (IMFs). Before the classification, the statistical moments of the signals in the frequency domain were obtained for feature extraction. Using these features, the performances of Support Vector Machines (SVM), Decision Tree, K-nearest Neighborhood (KNN) classifiers were analyzed. The experiment results show that the most accurate detection of epilepsy was obtained by applying EMD method with classifiers SVM, KNN and Decision Tree, and such algorithm with EMD can achieve accuracy for normal, pre-seizure, and seizure equal to 100%, 100% and 96.67%, respectively.

Eye blink artifact detection based on multi-dimensional EEG feature fusion and optimization

Since the multi-channel coil of magnetic resonance imaging (MRI) became a mainstream, methods to reduce inhomogeneity in MRI without using the body coil for additional scan have developed. The interference of signals made the appearance of MR image badly. As a result it serious affects the classifications for MR images. In this study we would find the most adaptable method to improve inhomogeneity for MRI and get improvement for brain tissue classification by using independent component analysis (ICA) and support vector machine (SVM) method under multiple-channel phase-array coil. We chose three inhomogeneity correction methods: discrete wavelet transform (DWT), local entropy minimize with B-spline (LEM-BS) and local entropy minimize with cubic spline (LEM-CS). In our experiment, these three methods were used as the pre-processing method before applying ICA + SVM to correct the inhomogeneity of MR image. Web site images are first applied in the experiment and the Tanimoto index was used to measure the performance for the three methods. Real phantom MR images were also applied in this experiment. The results show that the LEM-BS is the best choice. Instead of using the average filter for LEM-BS, we use Gaussian filter to get better classification result. The LEM-BS method can be applied not only in single slice but also in multiple slices and sometimes it shows better result.

Electroencephalogram (EEG) comprises valuable details related to the different physiological state of the brain. In this paper, a framework is offered for detecting the epileptic seizures from EEG data recorded from normal subjects and epileptic patients. This framework is based on a discrete wavelet transform (DWT) analysis of EEG signals using linear and nonlinear classifiers. The performance of the different combinations of two-class epilepsy detection is studied using Support Vector Machine (SVM) and neural network analysis (NNA) classifiers for the derived statistical features from DWT. In this new approach first parse EEG signals to sub-bands in different categories with the help of discrete wavelet transform (DWT) and then we derive statistical features such as Mean, Median, Standard Deviation, Kurtosis, Entropy, Skewness for each sub-band. These features, extracted from details and approximation coefficients of DWT sub-bands, are used as input to Principal Component Analysis (PCA). The classification is based on reducing the feature dimension using PCA and deriving the Support Vector Machine (SVM) and neural network analysis (NNA). In classification of normal and epileptic, results obtained exhibited an accuracy of 100% by applying NNA and 99% by SVM it has been found that the computation time of NNA classifier is lesser than SVM to provide 100% accuracy.

Abnormality and presence of neurological brain disorder such as epileptic seizure is diagnosed by analyzing electroencephalography signals accurately. The acquired brain signals are analyzed in time–frequency domains by using wavelet for accurate diagnosis. The online standard EEG database signal is preprocessed to remove power noise and most important eye blink artifact using independent component analysis. Daubechies wavelet is implemented, and decomposition of frequency is carried out up to eight levels. The exact sub-band of frequencies are extracted from band of frequencies which are called as delta band, theta band, alpha band, beta band and gamma band from lower to higher. Suitable features such as Lacunarity, Fluctuation Index, Energy and Entropy, Kolmogorov Entropy, Kurtosis and Skewness are extracted and classified using K-Nearest Neighbor, Support Vector Machine and Probabilistic Neural Network. Performance analysis is carried out by measuring specificity, sensitivity, accuracy and true predictive value. Implementation of independent component analysis to the affected channels in preprocessing block to remove eye blink artifact, decomposition of brain signal using Daubechies wavelet up to eight levels provides accurate diagnosis of epileptic seizure. The proposed seizure detection system provides high metric of performance parameters.

Enhanced detection of artifacts in EEG data using higher-order statistics and independent component analysis

Electroencephalograph (EEG) is brain activity measurement technique that it is widely used by placing electrode on the scalp. The EEG recording signals are vulnerably contaminated by noise. Thus, it can lead to misperception in analyzing EEG signals. One of the noise that has major influence on the recorded EEG signals is ocular artifacts (OAs). In this paper, a new method to remove OAs automatically is proposed by combining between Complete Ensemble Empirical Mode Decomposition (CEEMD) and Independent Component Analysis (ICA). CEEMD-ICA-OD removes OAs based on the concentration of noise that is calculated by calculating entropy for every CEEMD decomposition. Every selected decomposition of CEEMD will be used as input to the ICA method. Then, ICA will decompose selected CEEMD decomposition into independent components (ICs). Then, Chauvenet Criterion (CC) as Outlier Detection (OD) method is used to detect OAs pattern. Every detected OAs pattern will be eliminated by zeroing process on OAs pattern so that the EEG signal information remains secure. The proposed method were compared with Empirical Mode Decomposition (EMD), Ensemble EMD (EEMD) and Complete EEMD (CEEMD) with zeroing ICA combination for every comparison method. The performance measure of the proposed method and the benchmark method is calculated by using Root Mean Square Error (RMSE) and Signal to Noise Ratio (SNR) to show the performance quality in eliminating OAs. The experiment results show that the proposed method is successfully removing OAs better than the other methods with RMSE value and SNR value are 0.2491 and -0.3449, respectively.

Epilepsy is the most common neurological disorder characterized by recurrent seizures due to the abnormality of brain neuronal discharge. One of the tools to help neurologists diagnose epilepsy relies on EEG signals extensively because it can represent the brain's neuron activity. Due to the EEG signal being considered a complex biological signal, the signal complexity-based feature extraction method is suitable for characterizing epileptic EEG signals. One method for extracting EEG signals is entropy which is able to quantify signal complexity. Several previously proposed methods were wavelet entropy and IMF entropy. This study proposed multi-distance signal level difference (MSLD) entropy as a feature extraction method to extract the EEG signal complexity. The decomposed signal from MSLD was then extracted using entropy measurement. In the classification process, a Support Vector Machine (SVM) was applied to classify the normal, pre-ictal, and ictal signals of EEG signals. The highest accuracy of 96% was obtained using cubic SVM as the classifier. The proposed method provides an alternative method for analyzing EEG or other biomedical signals based on signal complexity.

<span>The purpose of this research is to observe the effectiveness of independent component analysis (ICA) method for denoising raw EEG signals based on word imagination, which will be used for word classification on unspoken speech. The electroencephalogram (EEG) signals are signals that represent the electrical activities of the human brain when someone is doing activities, such as sleeping, thinking or other physical activities. EEG data based on the word imagination used for the research is accompanied by artifacts, that come from muscle movements, heartbeat, eye blink, voltage and so on. In previous studies, the ICA method has been widely used and effective for relieving physiological artifacts. Artifact to signal ratio (ASR) is used to measure the effectiveness of ICA in this study. If the ratio is getting larger, the ICA method is considered effective for clearing noise and artifacts from the EEG data. Based on the experiment, the obtained ASR values from 11 subjects on 14 electrodes amounted are within the range of 0,910 to 1,080. Thus, it can be concluded that ICA is effective for removing artifacts from EEG signals based on word imagination.</span>

One of the most frequently used techniques for removing background noise from electroencephalogram (EEG) data is adaptive noise cancellation (ANC). Nonetheless, there exist two primary disadvantages associated with the adaptive noise reduction of EEG signals: the adaptive filter, which is supposed to be an approximation of contaminated noise, lacks the reference signal. The mean squared error (MSE) criterion is frequently employed to achieve this goal in adaptive filters. The MSE criterion, which only considers second-order errors, cannot be used since neither the EEG signal nor the EOG artifact are Gaussian. In this work, we employ an ANC system, deriving an estimate of EOG noise with a discrete wavelet transform (DWT) and input this signal into the reference of the ANC system. The entropy-based error metric is used to reduce the error signal instead of the MSE. Results from computer simulations demonstrate that the suggested system outperforms competing methods with respect to root-mean-square-error, signal-to-noise ratio, and coherence measurements.

Independent component analysis (ICA) has been shown to be a powerful tool for artifactual suppression from electroencephalogram (EEG) recordings. However, the real-time application of this method for artifact rejection has not been considered so far. This article presents a method based on an unsupervised, self-normalizing, adaptive learning algorithm for on-line blind source separation. Simulation results are provided to show the validity and effectiveness of the technique with different distributions. The results from real-data demonstrate that the proposed scheme removes perfectly eye blink and eye movement artifacts from the EEG signals and is suitable for use during on-line EEG monitoring such as EEG-based brain computer interface