EEG signals analysis for epileptic seizures detection using polynomial transforms, linear discriminant analysis and support vector machines

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.

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.

Hybrid metaheuristic algorithm enhanced support vector machine for epileptic seizure detection

Recent advancements in the field of neuroimaging have demonstrated that functional near-infrared spectroscopy (fNIRS) has a great potential for deception decoding. fNIRS is relatively new non-invasive optical imaging technique that has the advantages of safety, low cost, portability and easy to use. Light in the range of near infrared is used in fNIRS to measure the hemoglobin concentrations and thus determine the neural activity. In this paper we have compared the abilities of two classification techniques: linear discriminant analysis (LDA) and support vector machine (SVM), to classify the fNIRS data for deception decoding. Five healthy male subjects participated in deception and truth-telling scenarios separately in this study. Signals from the prefrontal cortex of the subjects were collected using continuous wave fNIRS. HbO and HbR signals were used to define the features and then data was classified using LDA and SVM classifiers. The average classification accuracy of SVM is 87.33 % whereas average classification accuracy for LDA is 78.34 %. The higher classification accuracy resulting from SVM is in accordance with the previous literature.

Brain Computer Interface (BCI) technology is used to help patients who do not have control over motor neurons such as ALS or paralyzed patients, to communicate with outer world. This work aims to classify motor imageries using real-time EEG dataset, which was published by Graz University, Austria. The dataset consists of two-channel EEG signals of right-hand movement imagery and left-hand movement imagery of 8 subjects. There are a total of 120 motor imagery trials (60 left and 60 right) EEG signals recorded from each subject. EEG signals are filtered and feature vectors were extracted that consist of 24, 32 and 40 relative band power values (RBPV). In this work, feature vectors classified by three different methods, linear discriminant analysis (LDA), K nearest neighbor (KNN) and support vector machines (SVM). Results show that best performance was achieved by 24 RBPV feature vector and LDA classification method.

The combination of Brain-Computer Interface (BCI) technology, allowing online monitoring and decoding of brain activity, with virtual and mixed reality (MR) systems may help to shape and guide implicit and explicit learning using ecological scenarios. Real-time information of ongoing brain states acquired through BCI might be exploited for controlling data presentation in virtual environments. Brain states discrimination during mixed reality experience is thus critical for adapting specific data features to contingent brain activity. In this study we recorded electroencephalographic (EEG) data while participants experienced MR scenarios implemented through the eXperience Induction Machine (XIM). The XIM is a novel framework modeling the integration of a sensing system that evaluates and measures physiological and psychological states with a number of actuators and effectors that coherently reacts to the user's actions. We then assessed continuous EEG-based discrimination of spatial navigation, reading and calculation performed in MR, using linear discriminant analysis (LDA) and support vector machine (SVM) classifiers. Dynamic single trial classification showed high accuracy of LDA and SVM classifiers in detecting multiple brain states as well as in differentiating between high and low mental workload, using a 5 s time-window shifting every 200 ms. Our results indicate overall better performance of LDA with respect to SVM and suggest applicability of our approach in a BCI-controlled MR scenario. Ultimately, successful prediction of brain states might be used to drive adaptation of data representation in order to boost information processing in MR.

Epilepsy is one of the most common neurological disorders. Epileptic seizures are the manifestation of abnormal hypersynchronous discharges of cortical neurons that impair brain function. Most of the people affected can be treated successfully with drug therapy or neurosurgical procedures. But there is still a large group of epilepsy patients that continues to have frequent seizures. For these patients automated detection of epileptic seizures can be of great clinical importance. Seizure detection can influence daily care or can be used to evaluate treatment effect. Furthermore automated detection can be used to trigger an alarm system during seizures that might be harmful to the patient. This thesis focusses on accelerometry (ACM) based seizure detection. A detailed overview is provided, on the perspectives for long-term epilepsy monitoring and automated seizure detection. The value of accelerometry for seizure detection is shown by means of a clinical evaluation and the first steps are made towards automatic detection of epileptic seizures based on ACM. With accelerometers movements are recorded. A large group of epileptic seizures manifest in specific movement patterns, so called motor seizures. Chapter 2 of this thesis presents an overview of the published literature on available methods for epileptic seizure detection in a long-term monitoring context. Based on this overview recommendations are formulated that should be used in seizure detection research and development. It is shown that for seizure detection in home environments, other sensor modalities besides EEG become more important. The use of alternative sensor modalities (such as ACM) is relatively new and so is the algorithm development for seizure detection based on these measures. It was also found that for both the adaptation of existing techniques and the development of new algorithms, clinical information should be taken more into account. The value of ACM for seizure detection is shown by means of a clinical evaluation in chapter 3. Here 3-D ACM- and EEG/video-recordings of 18 patients with severe epilepsy are visually analyzed. A striking outcome presented in this chapter is the large number of visually detected seizures versus the number of seizures that was expected on forehand and the number of seizures that was observed by the nurses. These results underscore the need for an automatic seizure detection device even more, since in the current situation many seizures are missed and therefore it is possible that patients do not get the right (medical) treatment. It was also observed that 95% of the ACM-patterns during motor seizures are sequences of three elementary patterns: myoclonic, tonic and clonic patterns. These characteristic patterns are a starting point for the development of methods for automated seizure detection based on ACM. It was decided to use a modular approach for the detection methodology and develop algorithms separately for motor activity in general, myoclonic seizures and tonic seizures. Furthermore, clinical information is incorporated in the detection methodology. Therefore in this thesis features were used that are either based on the shape of the patterns of interest as described in clinical practice (chapter 4 and 7), or the features were based on a physiological model with parameters that are related to seizure duration and intensity (chapter 5 and 6). In chapter 4 an algorithm is developed to distinguish periods with and without movement from ACM-data. Hence, when there is no movement there is no motor seizure. The amount of data that needs further analysis for seizure detection is thus reduced. From 15 ACM-signals (measured on five positions on the body), two features are computed, the variance and the jerk. In the resulting 2-D feature space a linear threshold function is used for classification. For training and testing the algorithm ACM data along with video data are used from nocturnal recordings in mentally retarded patients with severe epilepsy. Using this algorithm the amount of data that needs further analysis is reduced considerably. The results also indicate that the algorithm is robust for fluctuations across patients and thus there is no need for training the algorithm for each new patient. For the remaining data it needs to be established whether the detected movement is seizure related or not. To this purpose a model is developed for the accelerometer pattern measured on the arm during a myoclonic seizure (chapter 5). The model consists of a mechanical and an electrophysiological part. This model is used as a matched wavelet filter to detect myoclonic seizures. In chapter 6 the model based wavelet is compared to three other time frequency measures: the short time Fourier transform, the Wigner distribution and the continuous wavelet transform using a Daubechies wavelet. All four time-frequency methods are evaluated in a linear classification setup. Data from mentally retarded patients with severe epilepsy are used for training and evaluation. The results show that both wavelets are useful for detection of myoclonic seizures. On top of that, our model based wavelet has the advantage that it consists of parameters that are related to seizure duration and intensity that are physiological meaningful. Besides myoclonic seizures, the model is also useful for the detection of clonic seizures; physiologically these are repetitive myoclonic seizures. Finally for the detection of tonic seizures, in chapter 7 a set of features is studied that incorporate the mean characteristics of ACM-patterns associated with tonic seizures. Linear discriminant analysis is used for classification in the multi-dimensional feature space. For training and testing the algorithm, again data are used from recordings in mentally retarded patients with severe epilepsy. The results show that our approach is useful for the automated detection of tonic seizures based on 3-D ACM and that it is a promising contribution in a complete multi-sensor seizure detection setup.

Selective attention differentially modulates neural responses to simultaneously presented speech streams. In this study, single-trial EEG classification was performed to identify the attended speech from a two-talker speech mixture. During EEG recordings, normal-hearing listeners paid attention to one speech stream while listening to speech mixtures. The target-to-masker ratios (TMRs) varied from −9 to + 9 dB. Individual speech streams were processed with head-related transfer functions to simulate different spatial locations. Two simulated spatial conditions (0 vs. + /−90 and + 45 vs. −45 degree azimuth) were tested for each TMR. Features related to (1) cross-correlation values between EEG signals and temporal envelope of each speech stream, or (2) correlation values between reconstructed speech from the EEG signals with the acoustic stimuli, were fed to the classifiers. The dimensionality of the feature vector was reduced using Principal Component Analysis. Linear Discriminant Analysis and Support Vector Machine were used to classify the EEG signals. Classifiers were trained and tested with a five-fold cross validation method on data pooled across TMRs and source locations for trial lengths from 50 s to 10 s. Average classification accuracy was 85% with a 50 s trial length and maintained high at 70% with a reduced trial length of 10 s.

Brain-Computer Interface (BCI) or Human-Machine Interface now becoming vital biomedical engineering and technology field which applying EEG technologies to provide assistive device technology (AT) to humans. Hence, this paper presents the results of analyzing EEG signals from various human cognitive states to extract the suitable EEG features that can be employed to control BCI devices which can be used by disabled or paralyzed people. The EEG features in term of power spectral density, spectral centroids, standard deviation and entropy are selected and investigated from two different mental exercises; (i) quick solving math and (ii) relax (do nothing). The selected features then are classified using Linear Discriminant Analysis (LDA), Support Vector Machine (SVM) and K-Nearest Neighbors (k-NN) classifier. Among all these features, the best accuracy have been achieved by the power spectral density. The accuracies of this feature are 95%, 100%, 100% with LDA, SVM and K-NN respectively. Finally, the translation algorithm will be constructed using selected and classified EEG features to control the BCI devices.

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.

This study aims at designing an efficient combinatorial classifier, which fuses linear discriminant analysis (LDA) and kernel support vector machine (SVM) classifiers. The proposed method is called kernel SVM-LDA which benefits from global property of LDA, simultaneous with localized capability of SVM along with mapping ability of RBF kernel to project input data into a more separable high dimensional space. To assess the proposed scheme, Kernel SVM-LDA was applied to some standard datasets derived from UCI database and then compared to standard LDA and kernel SVM classifiers. Kernel SVM-LDA was also employed in cue-based brain computer interface to classify the left and right imagery movements. The results indicate that the introduced method is more superior to that of LDA and kernel SVM because it surpasses the counterparts in terms of robustness, complexity and performance.

Epilepsy is a persistent neurological condition of the brain in which the activity of the brain goes out of normal state. Classification and Analysis of EEG signal is the early approach for epilepsy diagnosis. During this paper, we have a tendency to propose an EEG signal classification approach based on Support Vector Machine (SVM) classifier. In extracting features from the raw EEG data we applied the Hilbert Transform method and used its coefficients. Then, after the PCA dimension reduction a two-class SVM classifier is used for EEG signals automatic classification, one class for healthy subjects and another for subjects with epilepsy. In SVM classifier we need to divide the EEG signals into a training dataset and testing dataset for classification. We have used five sets of EEG signals which are publicly accessible on EEG time series database. The evaluation and comparison of SVM based classifier with two other classification methods such as KNN and LVQ based classifier was done. The average classification accuracy of our study is 95.33%.

By 2021, WHO projects over a billion incapacitated people, with 20% facing daily functional impairments.Brain-Computer Interface (BCI) offers effortless machine control via direct brain-computer interaction, with Motor Imagery (MI) Electroencephalogram (EEG) as the key BCI foundation.The MI EEG signals were collected from the BNCI horizon2020 database for nine participants.The MI EEG data includes four tasks: imaginative movement of the left hand, right hand, feet, and tongue.There are 22 EEG channels with a sampling rate of 250Hz in the data.The MI EEG signals were band-pass filtered with a lower cut-off frequency of 0.5 Hz and an upper cut-off frequency of 100 Hz.Based on the energy count threshold approach (ECTA) nine channels were identified as dominating channels from the filtered MI EEG signals.Energy values for each channel were extracted in the ECTA method for the 3-sec window.And if the energy value of a channel for a particular window is greater than 60% of the maximum channel's energy, then the energy count value will be incremented by one.Finally whichever channels had larger energy counts were identified as a dominant channel.On these nine dominant MI EEG signals discrete wavelet transform using Daubechies 4 mother wavelet a four-level decomposition is applied, and Mu and beta rhythms were extracted.For a 3-sec window from the nine dominant channels, the energy and entropy feature values from the Mu and Beta rhythms were extracted.From the extracted features, 80% of the data is used for training Linear Discriminant Analysis (LDA) and Support Vector Machine (SVM) and 20% for testing classifier models.Both the models performed well on the test data and the results obtained had a highest accuracy 91.72.7 for subject-9 and these results were compared with the existing methods.The obtained results of MI EEG signals classification using BCI holds potential for revolutionizing assistive technology, stroke rehabilitation, virtual reality gaming,

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%

In brain–computer interfaces (BCIs), it is crucial to process brain signals to improve the accuracy of the classification of motor movements. Machine learning (ML) algorithms such as artificial neural networks (ANNs), linear discriminant analysis (LDA), decision tree (D.T.), K-nearest neighbor (KNN), naive Bayes (N.B.), and support vector machine (SVM) have made significant progress in classification issues. This paper aims to present a signal processing analysis of electroencephalographic (EEG) signals among different feature extraction techniques to train selected classification algorithms to classify signals related to motor movements. The motor movements considered are related to the left hand, right hand, both fists, feet, and relaxation, making this a multiclass problem. In this study, nine ML algorithms were trained with a dataset created by the feature extraction of EEG signals.The EEG signals of 30 Physionet subjects were used to create a dataset related to movement. We used electrodes C3, C1, CZ, C2, and C4 according to the standard 10-10 placement. Then, we extracted the epochs of the EEG signals and applied tone, amplitude levels, and statistical techniques to obtain the set of features. LabVIEW™2015 version custom applications were used for reading the EEG signals; for channel selection, noise filtering, band selection, and feature extraction operations; and for creating the dataset. MATLAB 2021a was used for training, testing, and evaluating the performance metrics of the ML algorithms. In this study, the model of Medium-ANN achieved the best performance, with an AUC average of 0.9998, Cohen’s Kappa coefficient of 0.9552, a Matthews correlation coefficient of 0.9819, and a loss of 0.0147. These findings suggest the applicability of our approach to different scenarios, such as implementing robotic prostheses, where the use of superficial features is an acceptable option when resources are limited, as in embedded systems or edge computing devices.