Types Of EEG Signals Research Articles

A novel moving window-based power spectrum features for single-channel EEG classification using machine learning

Electroencephalogram (EEG) signal classification is a crucial and very difficult task. Meanwhile, extracting features that are representative and able to discriminate different types of EEG signals is a complex task. Such features are usually fed to machine learning algorithms to classify the EEG signals based on the extracted features. This paper proposed a highly accurate and real-time features extraction method that can be used to help physicians in detecting different types of seizures and states in EEG signals characterized by a set of features extracted from the power spectrum of the EEG window. This is achieved by applying the following four steps. First, the EEG signals dataset contains different classes of EEG signals: Normal Eye Closed, Normal Eye Opened, Focal Seizure, Non-Focal Seizure, and Ictal Seizure activities. Second, each EEG signal has a length of 4097 samples sampled with a sampling frequency of 173.6 Hz which resulted in 23.6 seconds in length, this signal will be truncated into windows (Sub-signals) with a length of 349 samples (Approximately 2 seconds) with a total number of 12 windows for each signal. Afterward, the Fourier Transform (FT) based power spectrum will be computed for each window, then a set of different features are extracted from each window's FT power spectrum, and these features are classified using different Machine Learning (ML) algorithms. The results showed that the proposed methodology yields around 98% accuracy for the five different classification scenarios using different ML algorithms. The suggested method is hence robust, fast, real-time, accurate, and simple.

Novel Approach for EEG Signal Analysis in a Multifractal Paradigm of Motions. Epileptic and Eclamptic Seizures as Scale Transitions

Two different operational procedures are proposed for evaluating and predicting the onset of epileptic and eclamptic seizures. The first procedure analyzes the electrical activity of the brain (EEG signals) using nonlinear dynamic methods (the time variations of the standard deviation, the variance, the skewness and the kurtosis; the evolution in time of the spatial–temporal entropy; the variations of the Lyapunov coefficients, etc.). The second operational procedure reconstructs any type of EEG signal through harmonic mappings from the usual space to the hyperbolic one using the time homographic invariance of a multifractal-type Schrödinger equation in the framework of the scale relativity theory (i.e., in a multifractal paradigm of motions). More precisely, the explicit differential descriptions of the brain activity in the form of 2 × 2 matrices with real elements disclose, through the in-phase coherences at various scale resolutions (i.e., as scale transitions), the multitude of brain neuronal dynamics, especially sequences of epileptic and eclamptic seizures. These two operational procedures are not mutually exclusive, but rather become complementary, offering valuable information concerning epileptic and eclamptic seizures. In such context, the prediction of epileptic and eclamptic seizures becomes fundamental for patients not responding to medical treatment and also presenting an increased rate of seizure recurrence.

Integrated CWT-CNN for Epilepsy Detection Using Multiclass EEG Dataset

Electroencephalography is a common clinical procedure to record brain signals generated by human activity. EEGs are useful in Brain controlled interfaces and other intelligent Neuroscience applications, but manual analysis of these brainwaves is complicated and time-consuming even for the experts of neuroscience. Various EEG analysis and classification techniques have been proposed to address this problem however, the conventional classification methods require identification and learning of specific EEG characteristics beforehand. Deep learning models can learn features from data without having in depth knowledge of data and prior feature identification. One of the great implementations of deep learning is Convolutional Neural Network (CNN) which has outperformed traditional neural networks in pattern recognition and image classification. Continuous Wavelet Transform (CWT) is an efficient signal analysis technique that presents the magnitude of EEG signals as time-related Frequency components. Existing deep learning architectures suffer from poor performance when classifying EEG signals in the Time-frequency domain. To improve classification accuracy, we propose an integrated CWT and CNN technique which classifies five types of EEG signals using. We compared the results of proposed integrated CWT and CNN method with existing deep learning models e.g., GoogleNet, VGG16, AlexNet. Furthermore, the accuracy and loss of the proposed integrated CWT and CNN method have been cross validated using Kfold cross validation. The average accuracy and loss of Kfold cross-validation for proposed integrated CWT and CNN method are, 76.12% and 56.02% respectively. This model produces results on a publicly available dataset: Epilepsy dataset by UCI (Machine Learning Repository).

Dissimilarity-based time–frequency distributions as features for epileptic EEG signal classification

This work aims at exploring a general framework embedding techniques from classifiers, Time–Frequency Distributions (TFD) and dissimilarity measures for epileptic seizures detection. The proposed approach consists firstly in computing dissimilarities between TFD of electroencephalogram (EEG) signals and secondly in using them to define a decision rule. Compared to the existing approaches, the proposed one uses entire TFD of EEG signals and does not require arbitrary feature extraction. Several dissimilarity measures and TFDs have been compared to select the most appropriate for EEG signals. Classifiers, such as Artificial Neural Network (ANN), Support Vector Machine (SVM), Linear Discriminate Analysis (LDA) and k-Nearest Neighbours (k-NN), have been combined with the proposed approach. In order to evaluate the proposed approach, 13 different classification problems (including 2, 3 and 5-class) pertaining to five types of EEG signals have been used. The comparison between results obtained with the proposed approach and results reported in the literature with the same database of epileptic EEG signals demonstrates the effectiveness of this approach for seizure detection. Experimental results show that this approach has achieved highest accuracy in the most studied classification problems. A high value of 98% is achieved for the 5-class problem. Further, in most classification problems with 2 and 3-class, it also yields a satisfactory accuracy of approximately 100%. The robustness of the proposed approach is evaluated with the addition of noise to the EEG signals at various signal-to-noise ratios (SNRs). The experimental results show that this approach has a good classification accuracy at low SNRs.

GEVD Based on Multichannel Wiener Filter for Removal of EEG Artifacts

The electroencephalography (EEG) signals are contaminated by ocular artifacts usually called as ElectroOculoGraphy(EOG) artifacts. This occurs due to an eye movement and repeatedly blinking eyes, it is a major barrier to overcome when analyzing ElectroEncephaloGram (EEG) data. In this paper, Generalized Eigen Value Decomposition (GEVD) algorithm based on Multichannel Wiener filter (MWF) was proposed. In the GEVD algorithm, the covariance matrix of the artifact is identified and substituted by low rank approximation. For both real and hybrid EEG data is demonstrated using this algorithm and also compared with other existing methods for removal of artifacts. This paper determines generic, robust and fast algorithm for artifact removal of various types of EEG signals. Signal to Error Ratio (SER) and Artifact to Residue Ratio (ARR) both are expressed in dBs. The better performance of artifact removal is expressed with high SER which measures clean EEG distortion and ARR measures the artifact estimation.

Detection of epileptic seizure employing a novel set of features extracted from multifractal spectrum of electroencephalogram signals

Here, a technique for automated detection of epilepsy is proposed, based on a novel set of features derived from the multifractal spectrum of electroencephalogram (EEG) signals. In fractal geometry, multifractal detrended fluctuation analysis (MDFA) is a technique to examine the self-similarity of a non-linear, chaotic and noisy time series. EEG signals which are representatives of complex human brain dynamics can be effectively characterised by MDFA. Here, EEG signals representing healthy, interictal and seizure activities are acquired from an available dataset. The acquired signals are at first analysed using MDFA. Based on the multifractal analysis, 14 novel features are proposed in this study, to distinguish between different types of EEG signals. The statistical significance of the selected features is evaluated using Kruskal–Wallis test and is finally served as input feature vector to a support vector machines classifier for the classification of EEG signals. Four classification problems are presented in this work and it is observed that 100% classification accuracy is obtained for three problems which validate the efficacy of the proposed model for computer-aided diagnosis of epilepsy.

150. DETECTION OF EPILEPSY USING ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM

This study presents the use of Adaptive Neuro-Fuzzy Inference System (ANFIS) for classification of the EEG signals. The data consists of two types of EEG signals, i.e. epileptic patients during epilepsy and healthy patients when their eyes are open. The proposed algorithm has several steps. First, in order to remove the artefacts (filter the signals) we use band-pass Finite Impulse Response (FIR) filtering with the Hamming window. Feature extraction is made in the second step, using Discrete Wavelet Transform (DWT) and statistical analysis. In this way we reduce the dimensionality of the input data, lately used as input parameters in the ANFIS network. ANFIS model learns how to classify the EEG signal, through the standard hybrid learning algorithm. We use special form of ANFIS model, which depending on the number of inputs, splits the model into appropriate number of substructures (sub-ANFIS models). ANFIS model was evaluated in terms of training performance and classification accuracies. From the simulation results it was concluded that the proposed algorithm has good potentials in classifying the EEG signals.

EEG Classification Based on Sparse Representation and Deep Learning

For brain computer interfaces (BCIs) research, the classification of motor imagery brain signals is a major and challenging step. Based on the traditional sparse representation classification, a classification algorithm of electroencephalogram (EEG) based on sparse representation and convolution neural network is proposed by this paper. For the EEG signal, firstly, the features of the signal are obtained through the common spatial pattern (CSP) algorithm, and then the redundant dictionary with sparse representation is constructed based on these features. The sparse representation of the EEG signal is completed and the sparse features can be obtained. Finally, the sparse features are transformed into two dimensional signals, and the convolution neural network is used to complete the classification of EEG signals. Using the dataset downloaded from the website of BCI competition III (dataset IVa), for two types of EEG signals, the experiments show that the recognition accuracy of the method is over 80, and the recognition accuracy is better than that of the traditional SRC algorithm.

Visualization and Sonification of Long-Term Epilepsy Electroencephalogram Monitoring

Long-term electroencephalogram (EEG) monitoring is effective for epilepsy diagnosis. However, it also takes a lot of time for clinicians to correctly interpret the long-time recordings. Real-time computer-based EEG monitoring and classification systems have attracted recently the attention of researchers to help clinicians locate online possible epileptic-form EEG signals. In this paper, we first present an accurate and fast EEG classification algorithm that can recognize three types of EEG signals: normal, spike, and seizure. 16-channel bipolar EEG recordings of epilepsy patients are preprocessed, segmented, and ensemble empirical mode decomposed (EEMD) into intrinsic mode functions (IMFs). Features are extracted and linear discriminant analysis (LDA) is applied to train two classifiers: one is for seizure and non-seizure discrimination, and the other is for normal and spike discrimination. In order to furthermore help the clinicians, the results of LDA are visualized and sonified. The changes of the discriminant in the LDA on continuous EEG segments are backtracked to each feature, and thus to each EEG channel. Accordingly, contours of the changes in EEG channels are depicted. At the same time, sinusoidal waves in 440 or 880 Hz are played when EEG segments are classified into spike or seizure respectively. In the experiment, EEG recordings of six subjects (two normal and four seizure patients) are examined. The experiment result shows that the accuracy of the proposed epileptic EEG classification algorithm is relatively high. In addition, the visualization and sonification algorithms of epileptic-form EEG may greatly help clinicians localize the focus of seizure and nurses take care of seizure patients, immediately.

Generalized Hurst exponent estimates differentiate EEG signals of healthy and epileptic patients

The aim of our current study is to check whether multifractal patterns of the electroencephalographic (EEG) signals of normal and epileptic patients are statistically similar or different. In this regard, the generalized Hurst exponent (GHE) method is used for robust estimation of the multifractals in each type of EEG signals, and three powerful statistical tests are performed to check existence of differences between estimated GHEs from healthy control subjects and epileptic patients. The obtained results show that multifractals exist in both types of EEG signals. Particularly, it was found that the degree of fractal is more pronounced in short variations of normal EEG signals than in short variations of EEG signals with seizure free intervals. In contrary, it is more pronounced in long variations of EEG signals with seizure free intervals than in normal EEG signals. Importantly, both parametric and nonparametric statistical tests show strong evidence that estimated GHEs of normal EEG signals are statistically and significantly different from those with seizure free intervals. Therefore, GHEs can be efficiently used to distinguish between healthy and patients suffering from epilepsy.

Classification of EEG Signals Using Quantum Neural Network and Cubic Spline

Abstract The main aim of this paper is to propose Cubic Spline-Quantum Neural Network (CS-QNN) model for analysis and classification of Electroencephalogram (EEG) signals. Experimental data used here were taken from seven different electrodes. The work has been done in three stages, normalization of the signals, extracting the features by Cubic Spline Technique (CST) and classification using Quantum Neural Network (QNN). The simulation results showed that five types of EEG signals were classified with an average accuracy for seven electrodes that is 94.3% when training 70% of the features while with an average accuracy of 92.84% when training 50% of the features.

Combined strategy neural network using AR parameters for epileptic EEG signals classification

This paper describes the use of combined neural network for the classification of electroencephalogram (EEG) signals. Five types of EEG signals are defined for this work: Sets A and B for normal signals, Sets C and D for interictal signals, and Set E for ictal signal (we can found that in Bonn university). The EEG signals were decomposed using the autoregressive coefficients; this phase is named the feature extraction, and for that two methods are used: Yule-Walker and Burg, the combination of all methods is described. There are several classifiers and different neural networks which are available for the classification of anyway signal; in our work the multilayer preceptron neural network (MLPNN) is utilised for that. MLPNN is described by his architecture and characteristics like a hidden layer in which his size is changed for improving the results of classification. In outputs, three classes are defined (ADE, ACE, BDE, BCE), the accuracies are higher and they are not lower than 90% in which the best is calculated at 97.76%. Our method is very effective for the diagnosis of epileptic EEG signals.

Learning from brain control: clinical application of brain–computer interfaces

Abstract Brain-computer interfaces (BCI) use neuroelectric and metabolic brain activity to activate peripheral devices and computers without mediation of the motor system. In order to activate the BCI patients have to learn a certain amount of brain control. Self-regulation of brain activity was found to follow the principles of skill learning and instrumental conditioning. This review focuses on the clinical application of brain-computer interfaces in paralyzed patients with locked-in syndrome and completely locked-in syndrome (CLIS). It was shown that electroencephalogram (EEG)-based brain-computer interfaces allow selection of letters and words in a computer menu with different types of EEG signals. However, in patients with CLIS without any muscular control, particularly of eye movements, classical EEG-based brain-computer interfaces were not successful. Even after implantation of electrodes in the human brain, CLIS patients were unable to communicate. We developed a theoretical model explaining this fundamental deficit in instrumental learning of brain control and voluntary communication: patients in complete paralysis extinguish goal-directed responseoriented thinking and intentions. Therefore, a reflexive classical conditioning procedure was developed and metabolic brain signals measured with near infrared spectroscopy were used in CLIS patients to answer simple questions with a “yes” or “no”-brain response. The data collected so far are promising and show that for the first time CLIS patients communicate with such a BCI system using metabolic brain signals and simple reflexive learning tasks. Finally, brain machine interfaces and rehabilitation in chronic stroke are described demonstrating in chronic stroke patients without any residual upper limb movement a surprising recovery of motor function on the motor level as well as on the brain level. After extensive combined BCI training with behaviorally oriented physiotherapy, significant improvement in motor function was shown in this previously intractable paralysis. In conclusion, clinical application of brain machine interfaces in well-defined and circumscribed neurological disorders have demonstrated surprisingly positive effects. The application of BCIs to psychiatric and clinical-psychological problems, however, at present did not result in substantial improvement of complex behavioral disorders.

Learning from brain control: clinical application of brain–computer interfaces

Brain–computer interfaces (BCI) use neuroelectric and metabolic brain activity to activate peripheral devices and computers without mediation of the motor system. In order to activate the BCI patients have to learn a certain amount of brain control. Self-regulation of brain activity was found to follow the principles of skill learning and instrumental conditioning. This review focuses on the clinical application of brain–computer interfaces in paralyzed patients with locked-in syndrome and completely locked-in syndrome (CLIS). It was shown that electroencephalogram (EEG)-based brain–computer interfaces allow selection of letters and words in a computer menu with different types of EEG signals. However, in patients with CLIS without any muscular control, particularly of eye movements, classical EEG-based brain–computer interfaces were not successful. Even after implantation of electrodes in the human brain, CLIS patients were unable to communicate. We developed a theoretical model explaining this fundamental deficit in instrumental learning of brain control and voluntary communication: patients in complete paralysis extinguish goal-directed response-oriented thinking and intentions. Therefore, a reflexive classical conditioning procedure was developed and metabolic brain signals measured with near infrared spectroscopy were used in CLIS patients to answer simple questions with a “yes” or “no”-brain response. The data collected so far are promising and show that for the first time CLIS patients communicate with such a BCI system using metabolic brain signals and simple reflexive learning tasks. Finally, brain machine interfaces and rehabilitation in chronic stroke are described demonstrating in chronic stroke patients without any residual upper limb movement a surprising recovery of motor function on the motor level as well as on the brain level. After extensive combined BCI training with behaviorally oriented physiotherapy, significant improvement in motor function was shown in this previously intractable paralysis. In conclusion, clinical application of brain machine interfaces in well-defined and circumscribed neurological disorders have demonstrated surprisingly positive effects. The application of BCIs to psychiatric and clinical–psychological problems, however, at present did not result in substantial improvement of complex behavioral disorders.

Modeling Epileptic EEG Time Series by State Space Model and Kalman Filtering Algorithm

The human brain is one of the most complex physiological systems. Therefore, electroencephalogram (EEG) signal modeling is important to achieve a better understanding of the physical mechanisms generating these signals. The aim of this study is to investigate the application of Kalman filter and the state space model for estimation of electroencephalogram signals in a specific pathological state. For this purpose, two types of EEG signals (normal and partial epilepsy) were analyzed. The estimation performance of the proposed method on EEG signals is evaluated using the root mean square (RMS) measurement. The result of the present study shows that this model is appropriate for the analysis of EEG recordings. In fact, this model is capable of predicting changes in EEG time series with phenomena such as epileptic spikes and seizures.

Detection of Epileptic Seizure Event and Onset Using EEG

This study proposes a method of automatic detection of epileptic seizure event and onset using wavelet based features and certain statistical features without wavelet decomposition. Normal and epileptic EEG signals were classified using linear classifier. For seizure event detection, Bonn University EEG database has been used. Three types of EEG signals (EEG signal recorded from healthy volunteer with eye open, epilepsy patients in the epileptogenic zone during a seizure-free interval, and epilepsy patients during epileptic seizures) were classified. Important features such as energy, entropy, standard deviation, maximum, minimum, and mean at different subbands were computed and classification was done using linear classifier. The performance of classifier was determined in terms of specificity, sensitivity, and accuracy. The overall accuracy was 84.2%. In the case of seizure onset detection, the database used is CHB-MIT scalp EEG database. Along with wavelet based features, interquartile range (IQR) and mean absolute deviation (MAD) without wavelet decomposition were extracted. Latency was used to study the performance of seizure onset detection. Classifier gave a sensitivity of 98.5% with an average latency of 1.76 seconds.

An improved approximate entropy based epilepsy seizure detection using multi-wavelet and artificial neural networks

Epilepsy seizure is the result of the transient and unexpected electrical disturbance of the brain signal. The detection of epilepsy is only possible by analysing the normal and abnormal changes of brain electrical signal. The detection of epilepsy, which includes EEG recordings for the spikes and seizures, is very time consuming, especially in the case of long recordings. In this paper, an Artificial Intelligence (AI) based epilepsy detection technique is proposed. The technique is the combination of Multi-Wavelet Transform (MWT) and Artificial Neural Network (ANN). MWT is a technique based on wavelet theory, which is used to extract the features of EEG signal. The irregularity of EEG signal is measured by using the propose Improved Approximate Entropy (IApE). ANN is an AI that used to detect the type of EEG signal. The proposed technique is implemented, tested and the sensitivity, specificity, accuracy, precision response of IApE and ApE are compared.

A linear feature space for simultaneous learning of spatio-spectral filters in BCI

It is shown how two of the most common types of feature mapping used for classification of single trial Electroencephalography (EEG), i.e. spatial and frequency filtering, can be equivalently performed as linear operations in the space of frequency-specific detector covariance tensors. Thus by first mapping the data to this space, a simple linear classifier can directly learn optimal spatial + frequency filters. Significantly, if the classifier's loss function is convex, learning these filters is a convex minimisation problem. It is also shown how to pre-process the data such that the resulting decision function is robust to the biases inherent in EEG data. Further, based upon ideas from Max Margin Matrix Factorisation, it is shown how the trace norm can be used to select solutions which have low rank. Low rank solutions are preferred as they reflect prior information about the types of EEG signals we expect to see, i.e. that the classifiable information is contained in only a few spatio/spectral pairs. They are also easier to interpret. This feature-space transformation is compared with the Common-Spatial-Patterns on simulated and real Imagined Movement Brain Computer Interface (BCI) data and shown to give state-of-the-art performance.

Statistics over features: EEG signals analysis

This paper presented the usage of statistics over the set of the features representing the electroencephalogram (EEG) signals. Since classification is more accurate when the pattern is simplified through representation by important features, feature extraction and selection play an important role in classifying systems such as neural networks. Multilayer perceptron neural network (MLPNN) architectures were formulated and used as basis for detection of electroencephalographic changes. Three types of EEG signals (EEG signals recorded from healthy volunteers with eyes open, epilepsy patients in the epileptogenic zone during a seizure-free interval, and epilepsy patients during epileptic seizures) were classified. The selected Lyapunov exponents, wavelet coefficients and the power levels of power spectral density (PSD) values obtained by eigenvector methods of the EEG signals were used as inputs of the MLPNN trained with Levenberg–Marquardt algorithm. The classification results confirmed that the proposed MLPNN has potential in detecting the electroencephalographic changes.

Least squares support vector machine employing model-based methods coefficients for analysis of EEG signals

The aim of the study is classification of the electroencephalogram (EEG) signals by combination of the model-based methods and the least squares support vector machines (LS-SVMs). The LS-SVMs were implemented for classification of two types of EEG signals (set A – EEG signals recorded from healthy volunteers with eyes open and set E – EEG signals recorded from epilepsy patients during epileptic seizures). In order to extract the features representing the EEG signals, the spectral analysis of the EEG signals was performed by using the three model-based methods (Burg autoregressive – AR, moving average – MA, least squares modified Yule–Walker autoregressive moving average – ARMA methods). The present research demonstrated that the Burg AR coefficients are the features which well represent the EEG signals and the LS-SVM trained on these features achieved high classification accuracies.