Graph Theory and Machine Learning Based Epileptic Seizures Analysis from EEG

Automatic identification of epileptic seizures from EEG signals using linear programming boosting

Motor Imagery-based Brain-computer interface (MI-BCI) enables communications between the human brain and external machines, which is considered a valuable frontier technology. Motor imagery (MI) identification is a fundamental problem for these applications. However, because of the non-stationary and nonlinear properties of EEG motor imagery signals, it is very hard to get a high accuracy multi-class method. In this study, a multi-class classification method of MI EEG signals using wavelet packets and a one-dimensional convolutional neural network is presented. Data normalization and wavelet threshold denoising were performed for signal preprocessing. Then EEG signals were decomposed and reconstructed by wavelet packets transform. According to the features between different channels of EEG signal, a multilayer one-dimensional convolutional neural network model was adopted to get a multi-class classification model for MI. This model was validated on the public datasets BCI competition IV 2a. The result shows that the mean accuracy of this model is 93.05%. This study is beneficial to MI-BCI research applications in the future.

Brain computer interface (BCI) aims at providing a new communication way without depending on brain's normal output through nerve and muscle. The electroencephalography (EEG) has been widely used for BCI because it is a non-invasive approach. For the EEG signals of left and right hand motor imagery, the event-related desynchronization(ERD) and event-related synchronization(ERS) are used as classification features in this paper. The raw data are transformed by nonlinear methods and classified by Fisher classifier. Compared with the linear methods, the classification accuracy can get an obvious increase to 86.25%. Two different nonlinear transform were arised and one of them is under the consideration of the relativity of two channels of EEG signals. With these nonlinear transform, different misclassifications can get balance.

The root cause of the seizure is a sudden abnormal excessive electrical discharge in the brain and it is an Epileptic Seizure (ES) when such abnormal electrical activity arises particularly for epilepsy. Recognizing ES is crucial for effective treatment as it often repeats and can lead to serious outcomes. Since epilepsy is a neurological issue, detecting ES by analyzing brain signals is the preferred method, with Electroencephalogram (EEG) being the most reliable approach for this purpose. Different Machine Learning (ML) and Deep Learning (DL) methods are extensively used in ES detection from EEG signals. Existing methods first extract features from EEG signals using different methods and then classify ES using appropriate ML/DL methods. This study investigates ML-based ES detection where feature extraction from decomposed EEG signals using various methods and integrating the extracted features to classify ES are the main attractions. Empirical Mode Decomposition (EMD) is employed to systematically break down EEG signals into Intrinsic Mode Functions (IMFs), with the earlier IMFs containing more information than the later ones. From the initial six IMFs, three distinct features named Fluctuation index (F), Variance (V) and Ellipse Area (EA) of the second-order difference plot are extracted. Neural Network (NN), the well-known ML method, is employed in this study for ES classification from extracted F, V and EA features individually and integrating these (i.e., F + V + EA). The experimental analysis is conducted on the benchmark CHB-MIT dataset and the integrated feature set shows promising performance over individual feature sets. The proposed NN-based ES detection with integrated decomposed features outperforms prominent existing methods, showing an accuracy of 99.80%

Epilepsy is a neurological disorder of brain and the electroencephalogram (EEG) signals are commonly used to detect the epileptic seizures, the result of abnormal electrical activity in the brain. This paper focuses on the analysis of EEG signal to detect the presence of the epileptic seizure prior to its occurrence. The result could aid the individual in the initiation of delay sensitive diagnostic, therapeutic and alerting procedures. The methodology involves the multi resolution analysis (MRA) of the EEG signals of epileptic patient. MRA is carried out using discrete wavelet transform with daubechies 8 as the mother wavelet. For EEG data, the database of MIT-BIH of seven patients with different cases of epileptic seizure was used. The result with one of the patients showed presence of a unique pattern during the spectral analysis of the signal over different bands. Hence, based on the first patient, similar region is selected with the other patients and the multi-resolution analysis along with the principal component analysis (PCA) for the dimension reduction is carried out. Finally, these are treated with neural network to perform the classification of the signal either the epilepsy is occurring or not. The final results showed 100% accuracy with the detection with the neural network however it uses a large amount of data for analysis. Thus, the same was tested with dimension reduction using PCA which reduced the average accuracy to 89.51%. All the results have been simulated within the Matlab environment.

Wavelet transform decomposition of electroencephalogram (EEG) signals has been widely used for the analysis and detection of epileptic seizure of patients. However, the classification of EEG signals is still challenging because of high nonstationarity and high dimensionality. The aim of this work is an automatic classification of the EEG recordings by using statistical features extraction and support vector machine. From a real database, two sets of EEG signals are used: EEG recorded from a healthy person and from an epileptic person during epileptic seizures. Three important statistical features are computed at different sub-bands discrete wavelet and wavelet packet decomposition of EEG recordings. In this study, to select the best wavelet for our application, five wavelet basis functions are considered for processing EEG signals. After reducing the dimension of the obtained data by linear discriminant analysis and principal component analysis (PCA), feature vectors are used to model and to train the efficient support vector machine classifier. In order to show the efficiency of this approach, the statistical classification performances are evaluated, and a rate of 100% for the best classification accuracy is obtained and is compared with those obtained in other studies for the same dataset. However, this method is not meant to replace the clinician but can assist him for his diagnosis and reinforce his decision.

Epilepsy seizure is brain neurological abnormality which arises from the sudden deviation of the electrical interaction in the brain. Electroencephalography (EEGs) are obtained from 22 subjects with epileptic seizure and non-epileptic seizure states recorded using 23 channels with a sampling frequency of 256 Hz. The functional connectivity of the brain region can be exacted from the features obtained from EEG signals by measuring phase locking value (PLV). The neuronal connection in the brain can be expressed in terms of phase synchrony. Despite of the fact that brain states should be characterize independently based upon its extracted features. Therefore, a novel functional connectivity index (FCI) feature is proposed, namely Bi-Spectral Phase Concurrence Index (BSPCI). It is used to represent the spectral information with third cumulant order correlation functions of the EEG signal. In this paper, three FCI features were measured namely, the magnitude's mean of the bi-spectral, the normalized entropy of bi-spectrum (NE1) and the normalized entropy of squared bi-spectrum (NE2) from the BSPCI. Rank sum test based on the Wilcoxon approach is used to find the set of statistical difference between quantitative features extracted from EEG signals. The results provide evidence that the FCI will have an impact in separating the difference among seizure states of various epileptic seizure patients. On the part of reducing a large number of the feature vector, feature selection is performed by utilizing the sequential forward selection method. PLV is measured for quantifying the obtained phase synchrony of EEG signal. For the classification of epileptic seizure, Support Vector Machine is utilized which gain a large accuracy for the proposed bi-spectral analysis method when compared with Incremental Gradient Descent (IGD), Logistic Regression (LR) and Multilayer Perceptrons (MLP). The result is compared with IGD, LR, and MLP for obtaining better performance rate and the classification is 98.79% for the proposed work.

While acquiring EEG signal for recording brain activities, we often receive signals from other muscle activities which are added with the brain activity signal thus resulting in a contaminated EEG signal. Muscle activities such as eyeblink (EB) and eye ball movement are referred as Ocular Artifacts (OAs) which highly affect EEG signals. In Brain Computer Interface (BCI) systems, removal of OAs is important for correctly converting the brain thoughts into commands in order to control the external device. Various techniques like Independent component Analysis (ICA), and Principle Component Analysis (PCA) are widely used for the elimination of OAs but these techniques require multi channel EEG signals for processing. In this paper we have proposed the use of dual tree complex wavelet transform (DTCWT) with quantum inspired adaptive wavelet threshold algorithm for the elimination of OAs from single channel EEG signal. We have estimated Relative Root Mean Square Error (RRMSE). Results show better performance in reduction of ocular artifacts when using DTCWT with quantum inspired adaptive threshold.

An EEG signal is a measurement of currents that flow during synaptic excitations of the dendrites of many pyramidal neurons in the cerebral cortex.EEG is used to denote electrical neural activity of the brain. Electroencephalography (EEG) is the recording of electric potentials produced by the local collective partial synchrony of electrical field activity in cortical neuropile, today most commonly measured by an array of electrodes attached to the scalp using water-based gel.

Epilepsy is a life-threatening and challenging neurological disorder which results in a seizure. The seizure makes the functioning of the central nervous system abnormal and the patient ends up with convulsions and a loss of consciousness. Therefore, there is a need for in-depth diagnosis with high accuracy of the disease where electroencephalogram (EEG) signals play a vital role. Spiking neural network (SNN) is used as the classifier to classify EEG-based epileptic seizures. Due to its computational efficiency and biological plausibility, SNN is getting more attention to the classification of time-series data such as EEG signals. Although there are several other classification methods that exist, few of them is able to work with the EEG signals because EEG signals are generally highly variable and noisy. The SpiFoG algorithm is used to train SNN where inputs are encoded in temporal spikes from the feature extracted EEG signals. The dataset used in this research is the Bonn University EEG dataset for the epileptic seizure. From this dataset, three types of epileptic EEG signals are considered. The first is from healthy patients with open eyes, second is from healthy patients with closed eyes, and the third is from patients suffering from a seizure. Our proposed model outperforms state-of-the-art models.KeywordsEpileptic seizureEEGSpiFoGSpiking neurons

In order to measure sleep quality, sleep experts manually classify sleep stages through polysomnography (PSG) signals. However, it is time-consuming and labor-intensive work. Thus, automatic sleep stage classification methods are needed. In this study, we propose an end-to-end automatic sleep staging algorithm using a one-dimensional convolutional neural network (1DCNN) based on an inception network and bidirectional long short-term memory (bi-LSTM). First, a feature map was extracted from input data using the 1D-CNN architecture without preprocessing. Secondly, bi-LSTM learned a stage transition rule using the feature maps. In addition, we used the sleep-EDF public dataset to evaluate our model, and only one channel of EEG signal was used to save computational cost. The accuracy and macro-averaged F1 score of the classification performance were 85.05% and 79.05%, respectively. These results demonstrate state-of-the-art performance compared to previous studies using the same dataset, yielding an effective method for an automatic sleep staging algorithm.

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.

A novel approach based on wavelet analysis and arithmetic coding for automated detection and diagnosis of epileptic seizure in EEG signals using machine learning techniques

Reliable analysis of electroencephalogram (EEG) signals is crucial that could lead the way to correct diagnostic and therapeutic methods for the treatment of patients with neurological abnormalities, especially epilepsy. This paper presents a novel analysis system for detecting epileptic seizure from EEG signals, which uses statistical features based on optimum allocation technique (OAT) with logistic model trees (LMT). The analysis involves applying the OAT to select representative EEG signals that reflect the entire database. Then, some statistical features are extracted from these EEG signals and the obtained feature set is fed into the LMT classification model to detect epileptic seizure. To test the consistency of the proposed method, all experiments are carried out on a benchmark EEG dataset and repeated twenty times with the same parameters in the detection process, and the average values of the performance parameters are reported. The results show very high detection performances for each class, and also confirm the consistency of the proposed method in the repeating process. The proposed method outperforms some state-of-the-art methods of epileptic EEG signal detection using the same EEG dataset.

A broadband method of quantifying phase synchronization for discriminating seizure EEG signals