Seizure-free EEG Signals Research Articles

Nonictal electroencephalographic measures for the diagnosis of functional seizures.

Functional seizures (FS) look like epileptic seizures but are characterized by a lack of epileptic activity in the brain. Approximately one in five referrals to epilepsy clinics are diagnosed with this condition. FS are diagnosed by recording a seizure using video-electroencephalography (EEG), from which an expert inspects the semiology and the EEG. However, this method can be expensive and inaccessible and can present significant patient burden. No single biomarker has been found to diagnose FS. However, the current limitations in FS diagnosis could be improved with machine learning to classify signal features extracted from EEG, thus providing a potentially very useful aid to clinicians. The current study has investigated the use of seizure-free EEG signals with machine learning to identify subjects with FS from those with epilepsy. The dataset included interictal and preictal EEG recordings from 48 subjects with FS (mean age = 34.76 ± 10.55 years, 14 males) and 29 subjects with epilepsy (mean age = 38.95 ± 13.93 years, 18 males) from which various statistical, temporal, and spectral features from the five EEG frequency bands were extracted then analyzed with threshold accuracy, five machine learning classifiers, and two feature importance approaches. The highest classification accuracy reported from thresholding was 60.67%. However, the temporal features were the best performing, with the highest balanced accuracy reported by the machine learning models: 95.71% with all frequency bands combined and a support vector machine classifier. Machine learning was much more effective than using individual features and could be a powerful aid in FS diagnosis. Furthermore, combining the frequency bands improved the accuracy of the classifiers in most cases, and the lowest performing EEG bands were consistently delta and gamma.

Automatic classification of seizure and seizure-free EEG signals based on phase space reconstruction features.

Epilepsy is a type of brain disorder triggered by an abrupt electrical imbalance of neuronal networks. An electroencephalogram (EEG) is a diagnostic tool to capture the underlying brain mechanisms and detect seizure onset in epileptic patients. To detect seizures, neurologists need to manually monitor EEG recordings for long periods, which is challenging and susceptible to errors depending on expertise and experience. Therefore, automatic identification of seizure and seizure-free EEG signals becomes essential. This study introduces a method based on the features extracted from the phase space reconstruction for classifying seizure and seizure-free EEG signals. The computed features are derived from the elliptical area and interquartile range of the Euclidean distance by varying percentage values of data points ranging from 50 to 100%. We consider two public datasets and evaluate these features in each EEG epoch that includes the healthy, interictal, preictal, and ictal stages of epileptic subjects, utilizing the K-nearest neighbor classifier for classification. Results show that the features have higher values during the seizure than the seizure-free EEG signals and healthy subjects. Furthermore, the proposed features can effectively discriminate seizure EEG signals from the seizure-free and normal subjects with 100% accuracy, sensitivity, and specificity in both datasets. Likewise, the classification between the preictal stage and seizure EEG signals attains 98% accuracy. Overall, the reconstructed phase space features significantly enhance the accuracy of detecting epileptic EEG signals compared with existing methods. This advancement holds great potential in assisting neurologists in swiftly and accurately diagnosing epileptic seizures from EEG signals.

Epileptic detection in single and multi-lead EEG signals using persistent homology based on bi-directional weighted visibility graphs.

Epilepsy is a widespread neurological disorder, and its recurrence and suddenness are making automatic detection of seizure an urgent necessity. For this purpose, this paper performs topological data analysis (TDA) of electroencephalographic (EEG) signals by the medium of graphs to explore the potential brain activity information they contain. Through our innovative method, we first map the time series of epileptic EEGs into bi-directional weighted visibility graphs (BWVGs), which give more comprehensive reflections of the signals compared to previous existing structures. Traditional graph-theoretic measurements are generally partial and mainly consider differences or correlations in vertices or edges, whereas persistent homology (PH), the essential part of TDA, provides an alternative way of thinking by quantifying the topology structure of the graphs and analyzing the evolution of these topological properties with scale changes. Therefore, we analyze the PH for BWVGs and then obtain the two indicators of persistence and birth-death for homology groups to reflect the topology of the mapping graphs of EEG signals and reveal the discrepancies in brain dynamics. Furthermore, we adopt neural networks (NNs) for the automatic detection of epileptic signals and successfully achieve a classification accuracy of 99.67% when distinguishing among three different sets of EEG signals from seizure, seizure-free, and healthy subjects. In addition, to accommodate multi-leads, we propose a classifier that incorporates graph structure to distinguish seizure and seizure-free EEG signals. The classification accuracies of the two subjects used in the classifier are as high as 99.23% and 94.76%, respectively, indicating that our proposed model is useful for the analysis of EEG signals.

Recognizing seizure using Poincaré plot of EEG signals and graphical features in DWT domain.

Electroencephalography (EEG) signals are considered one of the oldest techniques for detecting disorders in medical signal processing. However, brain complexity and the non-stationary nature of EEG signals represent a challenge when applying this technique. The current paper proposes new geometrical features for classification of seizure (S) and seizure-free (SF) EEG signals with respect to the Poincaré pattern of discrete wavelet transform (DWT) coefficients. DWT decomposes EEG signal to four levels, and thus Poincaré plot is shown for coefficients. Due to patterns of the Poincaré plot, novel geometrical features are computed from EEG signals. The computed features are involved in standard descriptors of 2‑D projection (STD), summation of triangle area using consecutive points (STA), as well as summation of shortest distance from each point relative to the 45-degree line (SSHD), and summation of distance from each point relative to the coordinate center (SDTC). The proposed procedure leads to discriminate features between S and SF EEG signals. Thereafter, a binary particle swarm optimization (BPSO) is developed as an appropriate technique for feature selection. Finally, k-nearest neighbor (KNN) and support vector machine (SVM) classifiers are used for classifying features in S and SF groups. By developing the proposed method, we have archived classification accuracy of 99.3 % with respect to the proposed geometrical features. Accordingly, S and SF EEG signals have been classified. Also, Poincaré plot of SF EEG signals has more regular geometrical shapes as compared to S group. As a final remark, we notice that the Poincaré plot of coefficients in S EEG signals has occupied more space as compared to SF EEG signals (Tab. 3, Fig. 11, Ref. 57). Text in PDF www.elis.sk Keywords: EEG signal, DWT, Poincaré plot, geometrical feature, BPSO, SVM, KNN.

Modeling of seizure and seizure-free EEG signals based on stochastic differential equations

• We proposed a novel signal modeling for EEG signals in order to detect seizure and seizure free intervals. • A major advantage of this model is that it does not require decomposing EEG signal to its spectral constructive rhythms. • Our results demonstrated promising performance of proposed model. • This model is computationally very simple and reliable and can be used in real time applications of epileptic EEG classification. • Furthermore, this type of statistical modeling can be used for various EEG segment classifications. seizures commonly occurs in epileptic patients and decrease their quality of life. Investigating past attacks and predict future seizures can be done by exact classification between healthy and seizure based segments in electroencephalograph (EEG) recordings of these patients. Modeling EEG signal can help to extract discriminative features from it. These features make automatic classification more accurate. In this paper we propose a new modeling for EEG signals based on stochastic differential equations (SDE). In this statistical modeling, EEG signals are modeled with a self-similar fractional Levy stable process due to their inherent self-similarity. These processes are considered as response of SDE to the zero mean white symmetric alpha stable noise and inversely, by applying a derivative operator on these processes this white noise could be obtained again. We use a scale invariant fractional derivative operator for this purpose. Having fitted a probability distribution to the histogram of EEG signal after derivation, parameters of fitted histogram can be applied as features for classification task. We modeled healthy and epileptic segments of EEG signal from Bonn University database, and Neurology and Sleep Centre of New Delhi database. As an application of proposed model, we used features obtained from modeled signals to train an SVM classifier. Experimental result revealed highest classification of 99.8% for Bonn University database and 99.1% for Sleep Centre of New Delhi database, between normal and epileptic EEG signals. In conclusion, the proposed model is simple (does not require any decomposition of EEG signals), accurate and computationally efficient.

Automatic identification of epileptic seizures using volume of phase space representation.

Epilepsy is a neurological disorder that affects people of any age, which can be detected by Electroencephalogram (EEG) signals. This paper proposes a novel method called Volume of Phase Space Representation (VOPSR) to classify seizure and seizure-free EEG signals automatically. Primarily, the recorded EEG signal is disintegrated into several Intrinsic Mode Functions (IMFs) using the Empirical Mode Decomposition (EMD) method and the three-dimensional phase space have been reconstructed for the obtained IMFs. The volume is measured for the obtained 3D-PSR for different IMFs called VOPSR, which is used as a feature set for the classification of Epileptic seizure EEG signals. Support vector machine (SVM) is used as a classifier for the classification of epileptic and epileptic-free EEG signals. The classification performance of the proposed method is evaluated under different kernels such as Linear, Polynomial and Radial Basis Function (RBF) kernels. Finally, the proposed method outperforms noteworthy state-of-the-art classification methods in the context of epileptic EEG signals, achieving 99.13% accuracy (average) with the Linear, Polynomial, and RBF kernels. The proposed technique can be used to detect epilepsy from the EEG signals automatically without human intervention.

Higher-Order Phase-Space Reconstruction for Detection of Epileptic Electroencephalogram

In this paper, the authors propose a new technique for the classification of seizures, non-seizures, and seizure-free EEG signals based on non-linear trajectories of EEG signals. The EEG signals are decomposed using the EMD technique to obtain intrinsic mode functions (IMFs). The phase space of these IMFs is then reconstructed using a novel technique of higher-order dimensions (3D, 4D, 5D, 6D, 7D, 8D, 9D, and 10D). The existing techniques of seizure detection have deployed 2D & 3D phase–space reconstruction only. The Euclidean distance of all higher-order PSR is used as a feature to classify seizures, non-seizures, and seizure-free EEG signals. The performance of the proposed method is analyzed on the Bonn University database in which 7D reconstructed phase space classification accuracy of 99.9% has been achieved both using Random Forest classifier and J48 decision tree.

Classification of EEG signals for epileptic seizures detection and eye states identification using Jacobi polynomial transforms-based measures of complexity and least-square support vector machine

Abstract Background and objectives Epilepsy is the most prevalent neurological disorder in humans which is characterized by recurrent seizures resulting in neurologic, cognitive, psychological and social consequences. In addition, the automatic identification of brain conditions can be helpful in eye-brain-computer interface. That is why during these last few decades, many researchers focused on the development of classification systems for the automatic analysis and detection of epileptic seizures and eye states, which then can be integrated into implantable devices intended to detect the onset of seizures and trigger a focal treatment to block or suppress the seizures progression and to improve the living conditions of patients. In this impetus, this work aims to develop a novel automated classification system which can be performed to detect and identify whether electroencephalogram (EEG) signals belong to epileptic patients in seizure or seizure-free conditions, or to normal individuals with opened or closed eyes. Methods The proposed classification system consisted of models based on Jacobi polynomial transforms (JPTs). Discrete Legendre transform (DLT) and discrete Chebychev transform (DChT) firstly extract the beta (β) and gamma (γ) rhythms of EEG signals. Thereafter, different measures of complexity are computed from the EEG signals and their extracted rhythms, and applied as inputs of the least-square support vector machine (LS-SVM) classifier with radial basis function (RBF) kernel. Results The Kruskal-Wallis statistical test is performed and demonstrated that computed JPTs-based measures of complexity sufficiently discriminate EEG signals since their values are significantly higher for seizure EEG signals as compared to those of seizure-free and normal EEG signals with opened or closed eyes. In addition, these measures are used to construct eleven relevant classification problems using the LS-SVM which gained out area under curve (AUC) between 0.983 and 1 which is still proportional to maximum classification accuracies of 88.75% and 100% for normal with opened eyes versus normal with closed eyes, and normal versus seizure or seizure-free versus seizure classification problems, respectively. Conclusion Overall, it is found that the proposed classification system extends to be less complex for practical applications and can be suitable for automatic epileptic seizures detection and eye states identification.

Dynamic learning framework for epileptic seizure prediction using sparsity based EEG Reconstruction with Optimized CNN classifier

The World Health Organization (WHO) recently stated that epilepsy affects nearly 65 million people of the world population. Early forecast of the oncoming seizures is of paramount importance in saving the life of epileptic patients. This paper demonstrates a phase transition-based seizure prediction approach from multi-channel scalp electroencephalogram (EEG) recordings. The primary focus of this work is to discriminate the seizure and seizure-free EEG signals by learning the dynamics of preictal, interictal and ictal period. We propose an adaptive optimization approach using non-linear conjugate gradient technique in conjunction with Sparsity based EEG Reconstruction (SER) and three-dimensional Optimized Convolutional Neural Network (3D OCNN) classifier, based on Fletcher Reeves (FR) algorithm. Sparsity based artifact removal approach along with a 3D OCNN classifier, classifies the various states of seizures. FR algorithm is deployed with the deep neural network architecture to accelerate the convergence rate and to reduce the complexity of the proposed non-linear model. The Principle Component Analysis (PCA) algorithm replacing the Singular Value Decomposition (SVD) in the K-SVD algorithm, further reduces the time and complexity of the pre-processing stage. We further propose a Phase Transition based Kullback-Leibler divergence (PTB-KL) predictor for obtaining the Optimal Seizure Prediction Horizon (OSPH). The proposed model is evaluated using three diverse databases such as CHB-MIT, NINC and SRM respectively. Empirical results on the three EEG databases of 300 recordings outperforms the state-of-art approaches with an accuracy score of 0.98, sensitivity score of 0.99 and False Prediction Rate (FPR) of 0.07 FP/h. Statistical assessment of the proposed predictor gains an OSPH of about 1.1 h prior to the seizure onset. Experimental results prove that the phase transition-based seizure prediction approach is a promising one for accurate real-time prediction of epilepsy using scalp EEG data.

Classification of seizure and seizure free EEG signals using optimal mother wavelet and relative power

<p>This paper presents an approach for the selection of mother wavelet for classification of EEG epilepsy signals .Wavelet transform is very popular for analyzing signals in time and frequency domain. But as there are various wavelet families exist and not a one fits to all, in this study author have experimented with 51 wavelets from six different families Haar(haar), Daubechies(Db), Symlet(Sym), Coiflets(Coif), Biorthogonal(Bior) and Discrete Meyer(Dmey). Optimal mother wavelet is selected on the basis of highest correlation between input signal and reconstructed signal. With Discrete wavelet transform four levels of decomposition have been used to create the five EEG rhythms delta, theta, alpha, beta and gamma. Five features kurtosis, skew, mean, standard deviation and relative power have been extracted from each decomposed level by using the optimal mother wavelet. Statistical significance of the extracted features has been computed by Mann Whitney U test with significance level p<0.05 and statistical parameters sensitivity, specificity and accuracy for performance evaluation of the classifier have been computed. Results shown that out of six experimented wavelet families, five families with eight wavelets have qualified the correlation test. Out of five extracted feature relative power is more statistically significant for all type of classification and all EEG bands .Classifier used is support vector machine and accuracy of classifier lies in the range of 74% to 100 % for 14 classifications between different subsets.</p>

EEG Classification Using TQWT and Classifiers

The purpose of this study is to detect the epileptic seizures, which can be indicated by the abnormal disturbances in intracranial neurons using the electroencephalogram (EEG) signals. The EEG signals are grouped into three categories viz., Normal EEG signals (Z and O subsets), Seizure-free EEG signals (N and F subsets), and Seizure EEG signals (S subset). Whereas, for classification in this study, EEG signals are divided into three groups namely NF-S, O-FS, and ZO-NF-S. The signal length is fixed to be 4096 samples. The EEG signals will be decomposed by using Tunable-Q Wavelet Transform (TQWT), which produces intrinsic mode functions (IMFs) in decreasing order of frequency. These IMFs are analysed to gather the features of these signals, which help to classify them into various categories, and these features are fed as inputs to three classifiers viz., Random Forest (RF), Decision Table (DT), and Logistic Regression (LR). Logistic Regression classifier has showed higher accuracy, specificity and sensitivity for NF-S and O-F-S groups in comparison to RF and DT classifiers, whereas, Random Forest classifier expressed higher accuracy, specificity and sensitivity for ZO-NF-S groups in comparison to other classifiers. By utilising LR classifier, the suitable parameters of TQWT in NF-S (seizure-free vs. Seizure) are Q=6, r=3, and J=9 and showed maximum accuracy of 98%; and in O-F-S (Normal vs. Seizure-free vs. Seizure), Q=1, r=3, and J=9 attained maximum accuracy of 94.7%. Whereas, in ZONF-S (Normal vs. Seizure-free vs. Seizure), Q=4, r=3, and J=9 expressed maximum accuracy of 99.8% utilising Random Forest classifier.

Analysis of statistical coefficients and autoregressive parameters over intrinsic mode functions (IMFs) for epileptic seizure detection.

Epilepsy is a persistent neurological disorder impacting over 50 million people around the world. It is characterized by repeated seizures defined as brief episodes of involuntary movement that might entail the human body. Electroencephalography (EEG) signals are usually used for the detection of epileptic seizures. This paper introduces a new feature extraction method for the classification of seizure and seizure-free EEG time segments. The proposed method relies on the empirical mode decomposition (EMD), statistics and autoregressive (AR) parameters. The EMD method decomposes an EEG time segment into a finite set of intrinsic mode functions (IMFs) from which statistical coefficients and autoregressive parameters are computed. Nevertheless, the calculated features could be of high dimension as the number of IMFs increases, the Student's t-test and the Mann-Whitney U test were thus employed for features ranking in order to withdraw lower significant features. The obtained features have been used for the classification of seizure and seizure-free EEG signals by the application of a feed-forward multilayer perceptron neural network (MLPNN) classifier. Experimental results carried out on the EEG database provided by the University of Bonn, Germany, demonstrated the effectiveness of the proposed method which performance assessed by the classification accuracy (CA) is compared to other existing performances reported in the literature.

Classification of epileptic electroencephalogram signals using tunable-Q wavelet transform based filter-bank

The epilepsy is a neurological disorder and the seizure events frequently appear in epileptic patients. This disorder can be analysed through electroencephalogram (EEG) signals. In this paper, we propose a novel approach for automated identification of seizure EEG signals. The proposed method in this paper decomposes EEG signal into set of sub-band signals by applying tunable-Q wavelet transform (TQWT) based filter-bank. The sub-bands in TQWT based filter-bank have different value of quality (Q)-factor and have nearly constant bandwidth (BW). The features are computed by applying cross-information potential (CIP) on $$N_s$$ number of sub-band signals, for $$N_s$$ values varying from two to maximum number of sub-band signals obtained from TQWT based filter-bank. The features are computed for various values of $$N_s$$ and fed as input to random forest (RF) classifier. We have observed that, with the increase in the $$N_s$$, the number of computed features increases and hence the classification accuracy (ACC) depends on $$N_s$$. In this work, we have obtained ACC of $$99 \%$$ in the classification of normal, seizure-free, and seizure EEG signals using our proposed method. The developed algorithm is ready to be tested using huge database and can be employed to aid the epileptologists to screen the seizure-free and seizure patients accurately.

A Novel Signal Modeling Approach for Classification of Seizure and Seizure-Free EEG Signals.

This paper presents a signal modeling-based new methodology of automatic seizure detection in EEG signals. The proposed method consists of three stages. First, a multirate filterbank structure is proposed that is constructed using the basis vectors of discrete cosine transform. The proposed filterbank decomposes EEG signals into its respective brain rhythms: delta, theta, alpha, beta, and gamma. Second, these brain rhythms are statistically modeled with the class of self-similar Gaussian random processes, namely, fractional Brownian motion and fractional Gaussian noises. The statistics of these processes are modeled using a single parameter called the Hurst exponent. In the last stage, the value of Hurst exponent and autoregressive moving average parameters are used as features to design a binary support vector machine classifier to classify pre-ictal, inter-ictal (epileptic with seizure free interval), and ictal (seizure) EEG segments. The performance of the classifier is assessed via extensive analysis on two widely used data set and is observed to provide good accuracy on both the data set. Thus, this paper proposes a novel signal model for EEG data that best captures the attributes of these signals and hence, allows to boost the classification accuracy of seizure and seizure-free epochs.

An optimum allocation sampling based feature extraction scheme for distinguishing seizure and seizure-free EEG signals.

Epileptic seizure is the common neurological disorder, which is generally identified by electroencephalogram (EEG) signals. In this paper, a new feature extraction methodology based on optimum allocation sampling (OAS) and Teager energy operator (TEO) is proposed for detection of seizure EEG signals. The OAS scheme selects the finite length homogeneous sequence from non-homogeneous recorded EEG signal. The trend of selected sequence by OAS is still non-linear, which is analyzed by non-linear operator TEO. The TEO convert non-linear but homogenous EEG sequence into amplitude-frequency modulated (AM-FM) components. The statistical measures of AM-FM components used as input features to least squares support vector machine classifier for classification of seizure and seizure-free EEG signals. The proposed methodology is evaluated on a benchmark epileptic seizure EEG database. The experimental results demonstrate that the proposed scheme has capability to effectively distinguish seizure and seizure-free EEG signals.

Detection of epileptic seizure and seizure‐free EEG signals employing generalised S ‐transform

In this contribution, a novel technique for classification of electroencephalogram (EEG) signals has been presented employing generalised Stockwell ( S )-transform technique. S -transform is a technique for analysis of any non-stationary time series in joint time–frequency frame. In this work, epileptic seizure and seizure-free EEG signals have been taken from an available existing database and generalised S -transform is applied individually on different sets of EEG signals. Selective features like standard deviation and energy are evaluated from the joint time–frequency S -transform contour of the transformed signals and are eventually being classified using support vector machines (SVMs) and k-nearest neighbour (kNN) classifier. In this work, two different classification problems are addressed, namely (i) seizure and healthy (ii) seizure and inter-ictal, where both EEG signals of healthy and inter-ictal zone are considered to be in seizure-free class. For different cases investigated in this study, the highest overall classification accuracy of 98.44% is achieved using SVM classifier where as 100% accuracy is obtained using kNN classifier, which are comparable and even better than the results obtained in the existing literatures, analysed on the same dataset.

A novel framework based on biclustering for automatic epileptic seizure detection

Automatic epileptic seizure detection based on electroencephalogram is crucial to epilepsy diagnosis and treatment. However, the large numbers of time series make it quite challenging to establish a high performance automatic detection method. Considering different physiological states of the brain could be characterized by distinct combinations or interactions of similar discontinuous local temporal patterns, a novel framework based on biclustering for automatic epileptic seizure detection is proposed in this paper. First, the CC algorithm is used to identify similar discontinuous local temporal patterns. Then, the bicluster membership matrix using a new similarity measurement is constructed to reduce the dimensionality. At last, the ELM classifier is adopted to discriminate between epileptic seizure and seizure-free EEG signals. With extensive comparative studies and evaluations on the publicly available Bonn epileptic EEG dataset, it indicates that the proposed framework could not only automatically detect or predict an epilepsy seizure with high performances with respect to accuracy, robustness and efficiency, but also implicitly provide valuable knowledge for studying the mechanisms of epilepsy.

A novel approach for time–frequency localization of scaling functions and design of three-band biorthogonal linear phase wavelet filter banks

Design of time–frequency localized filters and functions is a classical subject in the field of signal processing. Gabor's uncertainty principle states that a function cannot be localized in time and frequency domain simultaneously and there exists a nonzero lower bound of 0.25 on the product of its time variance and frequency variance called time–frequency product (TFP). Using arithmetic mean (AM)– geometric mean (GM) inequality, product of variances and sum of variances can be related and it can be shown that sum of variances has lower bound of one. In this paper, we compute the frequency variance of the filter from its discrete Fourier transform (DFT) and propose an equivalent summation based discrete-time uncertainty principle which has the lower bound of one. We evaluate the performance of the proposed discrete-time time–frequency uncertainty measure in multiresolution setting and show that the proposed DFT based concentration measure generate sequences which are even more localized in time and frequency domain than that obtained from the Slepian, Ishii and Furukawa's concentration measures. The proposed design approach provides the flexibility in which the TFP can be made arbitrarily close to the lowest possible lower bound of 0.25 by increasing the length of the filter. In the other proposed approach, the sum of the time variance and frequency variance is used to formulate a positive definite matrix to measure the time–frequency joint localization of a bandlimited function from its samples. We design the time–frequency localized bandlimited low pass scaling and band pass wavelet functions using the eigenvectors of the formulated positive definite matrix. The samples of the time–frequency localized bandlimited function are obtained from the eigenvector of the positive definite matrix corresponding to its minimum eigenvalue. The TFP of the designed bandlimited scaling and wavelet functions are close to the lowest possible lower bound of 0.25 and 2.25 respectively. We propose a design method for time–frequency localized three-band biorthogonal linear phase (BOLP) wavelet perfect reconstruction filter bank (PRFB) wherein the free parameters can be optimized for time–frequency localization of the synthesis basis functions for the specified frequency variance of the analysis scaling function. The performance of the designed filter bank is evaluated in classification of seizure and seizure-free electroencephalogram (EEG) signals. It is found that the proposed filter bank outperforms other existing methods for the classification of seizure and seizure-free EEG signals.

Tunable-Q Wavelet Transform Based Multiscale Entropy Measure for Automated Classification of Epileptic EEG Signals

This paper analyzes the underlying complexity and non-linearity of electroencephalogram (EEG) signals by computing a novel multi-scale entropy measure for the classification of seizure, seizure-free and normal EEG signals. The quality factor (Q) based multi-scale entropy measure is proposed to compute the entropy of the EEG signal in different frequency-bands of interest. The Q -based entropy (QEn) is computed by decomposing the signal with the tunable-Q wavelet transform (TQWT) into the number of sub-bands and estimating K-nearest neighbor (K-NN) entropies from various sub-bands cumulatively. The optimal selection of Q and the redundancy parameter (R) of TQWT showed better robustness for entropy computation in the presence of high- and low-frequency components. The extracted features are fed to the support vector machine (SVM) classifier with the wrapper-based feature selection method. The proposed method has achieved accuracy of 100% in classifying normal (eyes-open and eyes-closed) and seizure EEG signals, 99.5% in classifying seizure-free EEG signals (from the hippocampal formation of the opposite hemisphere of the brain) from seizure EEG signals and 98% in classifying seizure-free EEG signals (from the epileptogenic zone) from seizure EEG signals, respectively, using the SVM classifier. We have also achieved classification accuracies of 99% and 98.6% in classifying seizure versus non-seizure EEG signals and the individual three classes, namely normal, seizure-free and seizure EEG signals, respectively. The performance measure of the proposed multi-scale entropy has been found to be comparable with the existing state of the art epileptic EEG signals classification methods studied using the same database.

Detection of epileptic seizure using Kraskov entropy applied on tunable-Q wavelet transform of EEG signals

Abstract The electroencephalogram (EEG) signals are basically electrophysiological signals that are normally used to access the condition of brain. Epilepsy is one of the brain's disorder. Automated diagnosis of epilepsy can be done by measuring and analyzing the nonlinear and non-stationary trends in EEG signals. This paper introduces a new diagnostic approach for analysis and classification of seizure and seizure-free EEG signals. In time-scale domain, the tunable-Q wavelet transform (TQWT) can reliably represent the sparsity in oscillatory signals. The proposed methodology begins with application of TQWT to efficiently characterize the non-stationary behavior and sparsity of EEG signals. TQWT decomposes the considered signals into a valuable set of band-limited signals termed as sub-bands for better feature extraction. Kraskov entropy is a nonlinear parameter to detect the presence of nonlinear trends in the signals. After decomposition, Kraskov entropy is computed from the specific sub-band as a decisive feature in order to discriminate seizure-free from the epileptic seizure EEG signals. Subsequently, obtained feature vectors are used for classifying the seizure and seizure-free EEG signals using the least square support vector machine (LS-SVM) classifier. While doing analysis, it has been observed that value of proposed Kraskov entropy based feature is significantly higher for seizure EEG signals as compared to that of seizure-free EEG signals. Furthermore, the experimental results of this work has demonstrated significant values of classification accuracy, sensitivity, specificity and Matthew's correlation coefficient. It is noteworthy that proposed framework uses single feature to diagnose the epilepsy accurately. Also the application of the proposed work on EEG data from the University of Bonn, Germany highlights the consistency and in some cases superiority of the proposed method over other popular methods.