Epileptic EEG Signals Research Articles

Mean amplitude spectrum based epileptic state classification for seizure prediction using convolutional neural networks

An increasing number of algorithms have been proposed for epileptic seizure prediction in recent years. But most of them are based on a partition of the electroencephalograph (EEG) signal of an epileptic patient into preictal, ictal (seizure), and interictal states. In this paper, we propose to further divide the preictal interval into multiple subintervals. Besides discriminating the seizure state from the preictal and interictal states, we also distinguish the preictal subintervals from each other. An epileptic state classification algorithm for epileptic EEG signals is then developed. The amplitude spectrums of EEG signals from 18 channels are firstly calculated and divided into 19 frequency subbands. The mean amplitude spectrum (MAS) on each of the 19 frequency subbands is then computed for each channel to form a MAS map of size 18 $$\times$$ 19. Finally, the MAS map is fed to a convolutional neural network (CNN) for feature extraction and a support vector machine (SVM) is employed for the epileptic state classification. Experiments show that, for a three-subinterval partition of the preictal state, a classification accuracy of 86.25% has been achieved on the CHB-MIT database by the MAS-based epileptic state classification algorithm using CNN and SVM.

EMPIRICAL MODE DECOMPOSITION-BASED PROCESSING FOR AUTOMATED DETECTION OF EPILEPSY

Epilepsy is a chronic illness of the brain characterized by recurring seizure attacks. Electroencephalogram (EEG) can record the electrical activity of the brain and is extensively used to analyze and diagnose epileptic seizures. However, the EEG signals are highly non-linear and chaotic and are difficult to analyze due to their small magnitude. Hence, empirical mode decomposition (EMD), a non-linear technique, has been widely adopted to capture the subtle changes present in the EEG signals. Hence, it is an added advantage to develop an automated computer-aided diagnostic (CAD) system to detect the different brain activities from the EEG signals using machine learning approaches. In this paper, we focus on the previous works which have used the EMD technique in the automated detection of normal or epileptic EEG signals.

Common and Special Knowledge-Driven TSK Fuzzy System and Its Modeling and Application for Epileptic EEG Signals Recognition

Takagi-Sugeno-Kang (TSK) fuzzy systems are well known for their good balances between approximation accuracy and interpretability. Among a wide variety of existing TSK fuzzy systems, most of them are driven by special knowledge since the learned parameters of each fuzzy rule are totally different. However, common knowledge is equally important and useful in practice and hence a TSK fuzzy system embedded with common knowledge should be more intuitive and interpretable when tackling with real-world problems. In this paper, we propose a common and special knowledge-driven TSK fuzzy system (CSK-TSK-FS), in which the parameters corresponding to each feature in then-parts of fuzzy rules always keep invariant and these parameters are viewed as common knowledge. As for its modeling, except the gradient descent techniques and other existing training algorithms, we can obtain a trained CSK-TSK-FS from a trained GMM or a trained FLNN because the proposed fuzzy system CSK-TSK-FS is mathematically equivalent to a special GMM and a FLNN. CSK-TSK-FS has three characteristics: (1) with the classical centroid defuzzification strategy, the involved common knowledge can be separated from fuzzy rules such that the interpretability of CSK-TSK-FS can be enhanced; (2) it can be trained quickly by the proposed LLM-based training algorithm; (3) the equivalence relationships among CSK-TSK-FS, GMM and FLNN allow them to share some commonality in training such that the proposed LLM-based training algorithm provides a novel fast training tool for training GMM and FLNN. Experimental results on UCI, KEEL and epileptic EEG datasets demonstrate the promising classification of CSK-TSK-FS.

Epileptic EEG Signals Recognition Using a Deep View-Reduction TSK Fuzzy System With High Interpretability

Takagi-Sugeno-Kang (TSK) fuzzy systems are well known for their good balance between approximation accuracy and interpretability. In this paper, we propose a deep view-reduction TSK fuzzy system termed as DVR-TSK-FS in which two powerful mechanisms associating with a deep structure are developed: 1) during the multi-view learning in each component, a sample-distribution-dependent parameter is defined to control the learning of the weight of each view. The parameter is not fixed by users, it is set according to the feature space in advance such that the learnt weight of each view indeed reflects the amount of pattern information involved in each view; 2) during the iteration of DRV-TSK-FS in each component, weak views are automatically reduced by comparing the learnt weight with a fixed threshold which is also automatically set according to the number of objects and the dimension of the feature space. 3) All components are linked in a stacked way based on the stacked generalization principle such that the outputs of all previous components are augmented into the current one which can help open the manifold structure of the original feature space. DRV-TSK-FS is testified on a multi-view EEG dataset for epileptic EEG signals recognition.

Automated System for Epileptic EEG Detection Using Iterative Filtering

The nonstationary characteristics present in electroencephalogram (EEG) signal require a crucial analysis that can reveal a method for diagnosis of neurological abnormalities, especially epilepsy. This article presents a new technique for automated classification of epileptic EEG signals based on iterative filtering (IF) of EEG signals. The superiority of IF over empirical mode decomposition for the classification of seizure EEG signals is presented. In this article, EEG epochs are decomposed into their intrinsic mode functions (IMFs) using IF. Amplitude envelope (AE) function is extracted from these modes, using the discrete separation energy algorithm. The features are extracted from these IMFs and AE functions. The feature set includes K-nearest neighbor entropy estimator, log energy entropy, Shannon entropy, and Poincar $\acute{\text{e}}$ plot parameters. These features are tested for their discriminative strength, on the basis of their $p$ -values, for classification of EEG signals into seizure, seizure-free, and normal classes. This proposed methodology has obtained a high classification accuracy using random forest classifier and takes far less time, which can be suitable for real-time implementation.

Automated epileptic seizures detection using multi-features and multilayer perceptron neural network

Detection of epileptic seizure activities from long-term multi-channel electroencephalogram (EEG) signals plays a significant role in the timely treatment of the patients with epilepsy. Visual identification of epileptic seizure in long-term EEG is cumbersome and tedious for neurologists, which might also lead to human error. Therefore, an automated tool for accurate detection of seizures in a long-term multi-channel EEG is essential for the clinical diagnosis. This study proposes an algorithm using multi-features and multilayer perceptron neural network (MLPNN) classifier. After appropriate approval from the ethical committee, recordings of EEG data were collected from the Institute of Neurosciences, Ramaiah Memorial College and Hospital, Bengaluru. Initially, preprocessing was performed to remove the power-line noise and motion artifacts. Four features, namely power spectral density (Yule–Walker), entropy (Shannon and Renyi), and Teager energy, were extracted. The Wilcoxon rank-sum test and descriptive analysis ensure the suitability of the proposed features for pattern classification. Single and multi-features were fed to the MLPNN classifier to evaluate the performance of the study. The simulation results showed sensitivity, specificity, and false detection rate of 97.1%, 97.8%, and 1 h−1, respectively, using multi-features. Further, the results indicate the proposed study is suitable for real-time seizure recognition from multi-channel EEG recording. The graphical user interface was developed in MATLAB to provide an automated biomarker for normal and epileptic EEG signals.

Predicción de crisis epilépticas utilizando el coeficiente de correlación producto-momento de Pearson a partir de un clasificador lineal de la distribución Gaussiana generalizada

To predict an epileptic event means the ability to determine in advance the time of the seizure with the highest possible accuracy. A correct prediction benchmark for epilepsy events in clinical applications is a typical problem in biomedical signal processing that helps to an appropriate diagnosis and treatment of this disease. In this work, we use Pearson's product-moment correlation coefficient from generalized Gaussian distribution parameters coupled with a linear-based classifier to predict between seizure and non-seizure events in epileptic EEG signals. The performance in 36 epileptic events from 9 patients showing good performance with 100% of effectiveness for sensitivity and specificity greater than 83% for seizures events in all brain rhythms. Pearson's test suggests that all brain rhythms are highly correlated in non-seizure events but no during the seizure events. This suggests that our model can be scaled with the Pearson's product-moment correlation coefficient for the detection of epileptic seizures.

Study of Weak Periodic Signals in the EEG Signals and Their Relationship With Postsynaptic Potentials.

In this paper, weak periodic signals detected in the EEG signals were analyzed for each region of the brain, and its relationships with postsynaptic potentials were investigated. For this, EEG signals were collected from 16 different channels according to the international standard channel 10/20 system in the scalp of two patients, one epileptic and one non-epileptic. In our recent work, we have detected weak periodic signals in the EEG signals using a Duffing oscillator system. In this paper, we have used a compact method by combining the Duffing oscillator system with scale index, which is based on a wavelet analysis in order to make the determination of the weak periodic signals more straightforward in practice. A number of weak periodic signals were detected in the range of frequencies 4.7-16 Hz in all channels, and however, these were less observed in epileptic EEG signals than non-epileptic. This methodology illustrates that weak periodic signals can be used to distinguish epileptic EEG signals from non-epileptic EEG signals. We have revealed some findings that the weak periodic signals are the postsynaptic potentials of the brain. Therefore, we can measure some frequencies values of the postsynaptic potentials within the EEG signals with this methodology.

Evaluation of Window Size in Classification of Epileptic Short-Term EEG Signals Using a Brain Computer Interface Software

Evaluation of Window Size in Classification of Epileptic Short-Term EEG Signals Using a Brain Computer Interface Software

Evaluation of window size in classification of epileptic short-term EEG signals using a Brain Computer Interface software

The complexity of epilepsy created a fertile ground for further research in automated methods, attempting to help the epileptologists’ task. Over the past years, great breakthroughs have emerged in computer-aided analysis and the advent of Brain Computer Interface (BCI) systems has greatly facilitated the automated seizure analysis. In this study, an evaluation of the window size in automated seizure detection is proposed. The EEG signals from the University of Bonn was employed and segmented into 24 epochs of different window lengths with 50% overlap each time. Statistical and spectral features were extracted in the OpenViBE scenario that were used to train four different classifiers. Results in terms of accuracy were above 80% for the Decision Trees classifier. Also, results indicated that different window sizes provide small variations in classification accuracy.

A stable feature extraction method in classification epileptic EEG signals.

Epilepsy is one of the most common neurological disorders. Electroencephalogram (EEG) signals are generally employed in diagnosing epilepsy. Therefore, extracting relevant features from EEG signals is one of the major tasks in an accurate diagnosis. In this study, the local ternary patterns, which is an image processing method, was improved in order to extract robust features from epileptic EEG signals. The EEG signals that were recorded by the Department of Etymology in the Bonn University were employed in the evaluation and validation of the proposed approach. Low and up features, which were extracted by the proposed one-dimensional ternary patterns, were classified by some machine learning methods such that support vector machine, functional trees, random forest (RF), Bayes networks (BayesNet), and artificial neural network, while the highest accuracies were obtained by RF. Achieved accuracies were found successful according to the current literature.

Denoising Sparse Autoencoder-Based Ictal EEG Classification.

Automatic seizure detection technology can automatically mark the EEG by using the epileptic detection algorithm, which is helpful to the diagnosis and treatment of epileptic diseases. This paper presents an EEG classification framework based on the denoising sparse autoencoder. The denoising sparse autoencoder (DSAE) is an improved unsupervised deep neural network over sparse autoencoder and denoising autoencoder, which can learn the closest representation of the data. The sparsity constraint applied in the hidden layer of the network makes the expression of data as sparse as possible so as to obtain a more efficient representation of EEG signals. In addition, corrupting operation used in input data help to enhance the robustness of the system and make it suitable for the analysis of non-stationary epileptic EEG signals. In this paper, we first imported the pre-processed training data to the DSAE network and trained the network. A logistic regression classifier was connected to the top of the DSAE. Then, put the test data into the system for classification. Finally, the output results of the overall network were post-processed to obtain the final epilepsy detection results. In the two-class (nonseizure and seizure EEGs) problem, the system has achieved effective results with the average sensitivity of 100%, specificity of 100%, and recognition of 100%, showing that the proposed framework can be efficient for the classification of epileptic EEGs.

Automatic seizure detection based on kernel robust probabilistic collaborative representation.

Visual inspection of electroencephalogram (EEG) recordings for epilepsy diagnosis is very time-consuming. Therefore, much research is devoted to developing a computer-assisted diagnostic system to relieve the workload of neurologists. In this study, a kernel version of the robust probabilistic collaborative representation-based classifier (R-ProCRC) is proposed for the detection of epileptic EEG signals. The kernel R-ProCRC jointly maximizes the likelihood that a test EEG sample belongs to each of the two classes (seizure and non-seizure), and uses the kernel function method to map the EEG samples into the higher dimensional space to relieve the problem that they are linearly non-separable in the original space. The wavelet transform with five scales is first employed to process the raw EEG signals. Next, the test EEG samples are collaboratively represented on the training sets by the kernel R-ProCRC and they are categorized by checking which class has the maximum likelihood. Finally, post-processing is deployed to reduce misjudgment and acquire more stable results. This method is evaluated on two EEG databases and yields an accuracy of 99.3% for interictal and ictal EEGs on the Bonn database. In addition, the average sensitivity of 97.48% and specificity of 96.81% are achieved from the Freiburg database. Graphical abstract Visual inspection of EEG recordings for epilepsy diagnosis is very time-consuming. Therefore, many researchers are devoted to developing a computer-assisted diagnostic system to relieve the workload of neurologists. In this paper, a kernel version of the robust probabilistic collaborative representation based classifier (R-ProCRC) is proposed for the detection of epileptic EEG signals. The kernel R-ProCRC jointly maximizes the likelihood that a test EEG sample belongs to each of the two classes, i.e., seizure and non-seizure, and uses the kernel function method to map the EEG samples into the higher dimensional space to relieve the problem that they are linearly non-separable in the original space. The main procedures of the proposed method are exhibited in the two figures as following, Fig. 1 The main procedures of the proposed method. (a) The schematic diagram of EEG classification based on the Freiburg database. (b) The detailed procedures of the kernel R-ProCRC This method has been evaluated on two different types of EEG databases and shows superior performance.

Epileptic Seizure Detection Based on Time-Frequency Images of EEG Signals Using Gaussian Mixture Model and Gray Level Co-Occurrence Matrix Features.

The electroencephalogram (EEG) signal analysis is a valuable tool in the evaluation of neurological disorders, which is commonly used for the diagnosis of epileptic seizures. This paper presents a novel automatic EEG signal classification method for epileptic seizure detection. The proposed method first employs a continuous wavelet transform (CWT) method for obtaining the time-frequency images (TFI) of EEG signals. The processed EEG signals are then decomposed into five sub-band frequency components of clinical interest since these sub-band frequency components indicate much better discriminative characteristics. Both Gaussian Mixture Model (GMM) features and Gray Level Co-occurrence Matrix (GLCM) descriptors are then extracted from these sub-band TFI. Additionally, in order to improve classification accuracy, a compact feature selection method by combining the ReliefF and the support vector machine-based recursive feature elimination (RFE-SVM) algorithm is adopted to select the most discriminative feature subset, which is an input to the SVM with the radial basis function (RBF) for classifying epileptic seizure EEG signals. The experimental results from a publicly available benchmark database demonstrate that the proposed approach provides better classification accuracy than the recently proposed methods in the literature, indicating the effectiveness of the proposed method in the detection of epileptic seizures.

Transductive Joint-Knowledge-Transfer TSK FS for Recognition of Epileptic EEG Signals.

Intelligent recognition of electroencephalogram (EEG) signals is an important means to detect seizure. Traditional methods for recognizing epileptic EEG signals are usually based on two assumptions: 1) adequate training examples are available for model training and 2) the training set and the test set are sampled from data sets with the same distribution. Since seizures occur sporadically, training examples of seizures could be limited. Besides, the training and test sets are usually not sampled from the same distribution for generic non-patient-specific recognition of EEG signals. Hence, the two assumptions in traditional recognition methods could hardly be satisfied in practice, which results in degradation of model performance. Transfer learning is a feasible approach to tackle this issue attributed to its ability to effectively learn the knowledge from the related scenes (source domains) for model training in the current scene (target domain). Among the existing transfer learning methods for epileptic EEG recognition, transductive transfer learning fuzzy systems (TTL-FSs) exhibit distinctive advantages-the interpretability that is important for medical diagnosis and the transfer learning ability that is absent from traditional fuzzy systems. Nevertheless, the transfer learning ability of TTL-FSs is restricted to a certain extent since only the discrepancy in marginal distribution between the training data and test data is considered. In this paper, the enhanced transductive transfer learning Takagi-Sugeno-Kang fuzzy system construction method is proposed to overcome the challenge by introducing two novel transfer learning mechanisms: 1) joint knowledge is adopted to reduce the discrepancy between the two domains and 2) an iterative transfer learning procedure is introduced to enhance transfer learning ability. Extensive experiments have been carried out to evaluate the effectiveness of the proposed method in recognizing epileptic EEG signals on the Bonn and CHB-MIT EEG data sets. The results show that the method is superior to or at least competitive with some of the existing state-of-art methods under the scenario of transfer learning.

An adaptive method for feature selection and extraction for classification of epileptic EEG signal in significant states

Electroencephalography (EEG) is a measurement tool to measure the electrical activity of brain observed due to chemical variation in brain. The EEG analysis has important role in feature extraction and classification methods for detecting and predicting various brain diseases. Epilepsy is a major disease characterized by seizures and observed due to sudden abnormal electrical discharges in the brain. Mostly the research is carried out to classify the normal EEG signal from epileptic EEG signal. The objective of the paper is to identify pre-seizure state and seizure state of EEG signal using time and frequency features. Classifying these states of EEG signal using fuzzy classifier helps in predicting seizures. The methodology uses hypothetical testing to refine feature selection and pattern adapted wavelet transform to improve classification. The appropriate feature selection reduces computational complexity of the classifier. The artifacts are removed from the EEG signal using Independent Component Analysis (ICA). The CHB-MIT EEG scalp dataset from Children’s Hospital, Boston is used for experimentation. The result shows classification accuracy of 96.48 %, True Positive Rate 96.52% and False Positive Rate 0.352 for seizure detection and 96.02% accuracy for Pre-seizure state detection with 13–110 s earlier than the onset seizures.

Detection of Epileptic EEG Signal Using Improved Local Pattern Transformation Methods

This paper aims to introduce efficient computer-aided techniques for automated epileptic diagnosis using the features based on improved local pattern transformation methods (LPT). To analyze electroencephalographic (EEG) signal, three techniques, namely one-dimensional local neighbor descriptive count, one-dimensional local gradient count and one-dimensional local binary count, are proposed in this work. Further, a signature point-based improved LPT approach is introduced for effectual classification of EEG signals. The features are computed at the signature points of the EEG signals, which are detected by using the difference of Gaussian pyramid. The features extracted from the signature points of the EEG signals are fed into artificial neural network (ANN) classifier for the discrimination of EEG signals. In this paper, seventeen different classification cases based on six different experimental cases are evaluated using the University of Bonn EEG database. Experimental results show that high classification accuracy for all the cases is achieved using the proposed approach and it also compares favorably to other state-of-the-art methods.

Generalized Hidden-Mapping Transductive Transfer Learning for Recognition of Epileptic Electroencephalogram Signals.

Electroencephalogram (EEG) signal identification based on intelligent models is an important means in epilepsy detection. In the recognition of epileptic EEG signals, traditional intelligent methods usually assume that the training dataset and testing dataset have the same distribution, and the data available for training are adequate. However, these two conditions cannot always be met in practice, which reduces the ability of the intelligent recognition model obtained in detecting epileptic EEG signals. To overcome this issue, an effective strategy is to introduce transfer learning in the construction of the intelligent models, where knowledge is learned from the related scenes (source domains) to enhance the performance of model trained in the current scene (target domain). Although transfer learning has been used in EEG signal identification, many existing transfer learning techniques are designed only for a specific intelligent model, which limit their applicability to other classical intelligent models. To extend the scope of application, the generalized hidden-mapping transductive learning method is proposed to realize transfer learning for several classical intelligent models, including feedforward neural networks, fuzzy systems, and kernelized linear models. These intelligent models can be trained effectively by the proposed method even though the data available are insufficient for model training, and the generalization abilities of the trained model is also enhanced by transductive learning. A number of experiments are carried out to demonstrate the effectiveness of the proposed method in epileptic EEG recognition. The results show that the method is highly competitive or superior to some existing state-of-the-art methods.

Feature extraction using Pythagorean means for classification of epileptic EEG signals

Electroencephalogram (EEG) is a widely used tool for the study and diagnosis of epilepsy. The patients subjected to epilepsy require long term monitoring of EEG. Automatic seizure detection will eliminate chances of missing any seizure, make detection easy and reduce burden on physicians. In this work, different combination of Pythagorean means (time domain features) namely arithmetic mean (AM), geometric mean (GM) and harmonic mean (HM) of energy per epoch are used as features to classify EEG data into normal, seizure free and seizure classes by using a linear classifier. The classification accuracy of 100% is achieved in two and three class problem with a single feature/epoch and in five class problem with two features/epoch. The novelty of this work is use of new and simple features (in epileptic EEG signal classification), reduced complexity and high performance.

A Study of Combined Lossy Compression and Seizure Detection on Epileptic EEG Signals

Electroencephalogram (EEG) has been widely used in diagnosing and detecting epileptic seizure. Large epileptic EEG databases have been built, the use of EEG compression is therefore becoming necessary. Epilepsy causes a change on EEG characteristics, especially on frequency, hence exploiting these features may improve the performance of EEG lossy compression techniques that are mostly working on frequency domain. In this paper, we propose a lossy compression method for epileptic EEG data, by exploiting the characteristics of EEG under epilepsy. Moreover, the recovered EEG signals processed by the proposed method are used by an EEG-based seizure detection system to evaluate the possibility of applying in real world as well as the impact of lossy compression on seizure detection. The results show that the proposed method gives a higher result, and applying the proposed method to EEG-based seizure detection system is feasible and has the advantage compared to using lossless ones.