No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
An efficient feature selection and explainable classification method for EEG-based epileptic seizure detection
Abstract
Epilepsy is a prevalent neurological disorder that poses life-threatening emergencies. Early electroencephalo-gram (EEG) seizure detection can mitigate the risks and aid in the treatment of patients with epilepsy. EEG based epileptic seizure (ES) detection has significant applications in epilepsy treatment and medical diagnosis. Therefore, this paper presents an innovative framework for efficient ES detection, providing coefficient and distance correlation feature selection algorithms, a Bagged Tree-based classifer (BTBC), and Explainable Artificial Intelligence (XAI). Initially, the Butterworth filter is employed to eliminate various artifacts, and the discrete wavelet transform (DWT) is used to decompose the EEG signals and extract various eigenvalue features of the statistical time domain (STD) as linear and Fractal dimension-based non-linear (FD-NL). The optimal features are then identified through correlation coefficients with-value and distance correlation analysis.These features are subsequently utilized by the Bagged Tree-based classifer (BTBC). The proposed model provides best performance in mitigating overfitting issues and improves the average accuracy by 2% using (CD, E), (AB, CD, E), and (A, B) experimental types as compared to other machine learning (ML) models using well-known Bonn and UCI-EEG benchmark datasets. Finally, SHapley additive exPlanation (SHAP) was used to interpret and explain the decision-making process of the proposed model. The results highlight the framework's capability to accurately classify ES, thereby improving the diagnosis process in patients with brain dysfunctions.
... In [22], the authors performed minor signal processing steps such as filtering, and used the discrete wavelet transform (DWT) to decompose the EEG signals and extract various eigenvalue features of the statistical time domain (STD) as linear and Fractal Dimension-based Nonlinear (FD-NL) features. Following this feature extraction step, the optimal features were identified through correlation coefficients with p-value and distance correlation analysis and classified using a Bagged Tree-Based Classifer (BTBC), followed by SHAP to provide the explanations. ...
... The EEG windows were separated into the following three sub-bands: low frequencies (<12 Hz), beta (12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25), and gamma (>25 Hz). Second-order Butterworth filters were used for band separation. ...
... The features that had a moderate/strong monotonic relation with the XAI explanations were diverse, including STD, MAD IR, RMS, and Range. Most features were estimated for the beta band (12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25). ...
Epilepsy is a neurological disease with one of the highest rates of incidence worldwide. Although EEG is a crucial tool for its diagnosis, the manual detection of epileptic seizures is time consuming. Automated methods are needed to streamline this process; although there are already several works that have achieved this, the process by which it is executed remains a black box that prevents understanding of the ways in which machine learning algorithms make their decisions. A state-of-the-art deep learning model for seizure detection and three EEG databases were chosen for this study. The developed models were trained and evaluated under different conditions (i.e., three distinct levels of overlap among the chosen EEG data windows). The classifiers with the best performance were selected, then Shapley Additive Explanations (SHAPs) and Local Interpretable Model-Agnostic Explanations (LIMEs) were employed to estimate the importance value of each EEG channel and the Spearman’s rank correlation coefficient was computed between the EEG features of epileptic signals and the importance values. The results show that the database and training conditions may affect a classifier’s performance. The most significant accuracy rates were 0.84, 0.73, and 0.64 for the CHB-MIT, Siena, and TUSZ EEG datasets, respectively. In addition, most EEG features displayed negligible or low correlation with the importance values. Finally, it was concluded that a correlation between the EEG features and the importance values (generated by SHAP and LIME) may have been absent even for the high-performance models.
... In the last couple of decades, machine learning has been vigorously applied for phishing detection through different models and methods for various purposes [32][33][34]. For instance, Wang [35] proposed ensemble classifiers for email filtering that eliminated five algorithms: Support Vector Machines, K-Nearest Neighbor, Gaussian Naive Bayes, Bernoulli Naive Bayes, and Random Forest Classifier. ...
The escalating reliance on revolutionary online web services has introduced heightened security risks, with persistent challenges posed by phishing despite extensive security measures. Traditional phishing systems, reliant on machine learning and manual features, struggle with evolving tactics. Recent advances in deep learning offer promising avenues for tackling novel phishing challenges and malicious URLs. This paper introduces a two-phase stack generalized model named AntiPhishStack, designed to detect phishing sites. The model leverages the learning of URLs and character-level TF-IDF features symmetrically, enhancing its ability to combat emerging phishing threats. In Phase I, features are trained on a base machine learning classifier, employing K-fold cross-validation for robust mean prediction. Phase II employs a two-layered stacked-based LSTM network with five adaptive optimizers for dynamic compilation, ensuring premier prediction on these features. Additionally, the symmetrical predictions from both phases are optimized and integrated to train a meta-XGBoost classifier, contributing to a final robust prediction. The significance of this work lies in advancing phishing detection with AntiPhishStack, operating without prior phishing-specific feature knowledge. Experimental validation on two benchmark datasets, comprising benign and phishing or malicious URLs, demonstrates the model’s exceptional performance, achieving a notable 96.04% accuracy compared to existing studies. This research adds value to the ongoing discourse on symmetry and asymmetry in information security and provides a forward-thinking solution for enhancing network security in the face of evolving cyber threats.
The efficient patient-independent and interpretable framework for electroencephalogram (EEG) epileptic seizure detection (ESD) has informative challenges due to the complex pattern of EEG nature. Automated detection of ES is crucial, and Explainable Artificial Intelligence (XAI) is urgently needed to justify algorithmic predictions in clinical settings. Therefore, this study implements an XAI-based computer-aided ES detection system (XAI-CAESDs), comprising three major modules including of feature engineering module, a seizure detection module, and an explainable decision-making process module in a smart healthcare system. To ensure the privacy and security of biomedical EEG data, the blockchain is employed. Initially, the Butterworth filter eliminates various artifacts, and the Dual-Tree Complex Wavelet Transform (DTCWT) decomposes EEG signals, extracting real and imaginary eigenvalue features using frequency domain (FD), time domain (TD), and Fractal Dimension (FD) of linear and non-linear features. The best features are selected by using Correlation Coefficients (CC) and Distance Correlation (DC). The selected features are fed into the Stacking Ensemble Classifiers (SEC) for EEG ES detection. Further, the Shapley Additive Explanations (SHAP) method of XAI is implemented to facilitate the interpretation of predictions made by the proposed approach, enabling medical experts to make accurate and understandable decisions. The proposed ensemble-based stacking classifiers in XAI-CAESDs have demonstrated 2% best average accuracy, Recall, specificity, and F1-score using the University of California, Irvine, Bonn University, and Boston Children's Hospital-MIT EEG data sets. The proposed framework enhances decision-making and the diagnosis process using biomedical EEG signals and ensures data security in smart healthcare systems.
The automatic classification of epilepsy electroencephalogram (EEG) signals plays a crucial role in diagnosing neurological diseases. Although promising results have been achieved by deep learning methods in this task, capturing the minute abnormal characteristics, contextual information, and long dependencies of EEG signals remains a challenge. To address this challenge, a positional multi-length and mutual-attention (PMM) network is proposed for the automatic classification of epilepsy EEG signals. The PMM network incorporates a positional feature encoding process that extracts minute abnormal characteristics from the EEG signal and utilizes a multi-length feature learning process with a hierarchy residual dilated LSTM (RDLSTM) to capture long contextual dependencies. Furthermore, a mutual-attention feature reinforcement process is employed to learn the global and relative feature dependencies and enhance the discriminative abilities of the network. To validate the effectiveness PMM network, we conduct extensive experiments on the public dataset and the experimental results demonstrate the superior performance of the PMM network compared to state-of-the-art methods.
In clinical practice, algorithmic predictions may seriously jeopardise patients’ health and thus are required to be validated by medical experts before a final clinical decision is met. Towards that aim, there is need to incorporate explainable artificial intelligence techniques into medical research. In the specific field of epileptic seizure detection there are several machine learning algorithms but less methods on explaining them in an interpretable way. Therefore, we introduce XAI4EEG: an application-aware approach for an explainable and hybrid deep learning-based detection of seizures in multivariate EEG time series. In XAI4EEG, we combine deep learning models and domain knowledge on seizure detection, namely (a) frequency bands, (b) location of EEG leads and (c) temporal characteristics. XAI4EEG encompasses EEG data preparation, two deep learning models and our proposed explanation module visualizing feature contributions that are obtained by two SHAP explainers, each explaining the predictions of one of the two models. The resulting visual explanations provide an intuitive identification of decision-relevant regions in the spectral, spatial and temporal EEG dimensions. To evaluate XAI4EEG, we conducted a user study, where users were asked to assess the outputs of XAI4EEG, while working under time constraints, in order to emulate the fact that clinical diagnosis is done - more often than not - under time pressure. We found that the visualizations of our explanation module (1) lead to a substantially lower time for validating the predictions and (2) leverage an increase in interpretability, trust and confidence compared to selected SHAP feature contribution plots.
Epileptic seizures are a chronic and persistent neurological illness that mainly affects the human brain. Electroencephalogram (EEG) is considered an effective tool among neurologists to detect various brain disorders, including epilepsy, owing to its advantages, such as its low cost, simplicity, and availability. In order to reduce the severity of epileptic seizures, it is necessary to design effective techniques to identify the disease at an earlier stage. Since the traditional way of diagnosing epileptic seizures is laborious and time-consuming, automated tools using machine learning (ML) and deep learning (DL) models may be useful. This paper presents an intelligent deep canonical sparse autoencoder-based epileptic seizure detection and classification (DCSAE-ESDC) model using EEG signals. The proposed DCSAE-ESDC technique involves two major processes, namely, feature selection and classification. The DCSAE-ESDC technique designs a novel coyote optimization algorithm (COA)-based feature selection technique for the optimal selection of feature subsets. Moreover, the DCSAE-based classifier is derived for the detection and classification of different kinds of epileptic seizures. Finally, the parameter tuning of the DSCAE model takes place via the krill herd algorithm (KHA). The design of the COA-based feature selection and KHA-based parameter tuning shows the novelty of the work. For examining the enhanced classification performance of the DCSAE-ESDC technique, a detailed experimental analysis was conducted using a benchmark epileptic seizure dataset. The comparative results analysis portrayed the better performance of the DCSAE-ESDC technique over existing techniques, with maximum accuracy of 98.67% and 98.73% under binary and multi-classification, respectively.
Epileptic seizure is one of the most chronic neurological diseases that instantaneously disrupts the lifestyle of affected individuals. Towards developing novel and efficient technology for epileptic seizure management, recent diagnostic approaches have focused on developing machine/deep learning models (ML/DL) based Electroencephalogram (EEG) methods. Importantly, EEG's non-invasiveness and ability to offer repeated patterns of epileptic-related electrophysiological information have motivated development of varied ML/DL algorithms for epileptic seizure diagnosis in the recent years. However, EEG’s low amplitude and non-stationary characteristics make it difficult for existing ML/DL models to achieve a consistent and satisfactory diagnosis outcome, especially in clinical settings were environmental could hardly be avoided. Though several recent works have explored the use of EEG-based ML/DL methods and statistical feature for seizure diagnosis, it is unclear what the advantages and limitations of these works are, which might preclude advancement of research and development in the field of epileptic seizure diagnosis and appropriate criteria for selecting ML/DL models and statistical feature extraction methods for EEG-based epileptic seizure diagnosis. Therefore, this paper attempts to bridge this research gap by conducting an extensive systematic review on recent developments of EEG-based ML/DL technologies for epileptic seizure diagnosis. In the review, current development in seizure diagnosis, various statistical feature extraction methods, ML/DL models, their performances, limitations, and core challenges as applied in EEG-based epileptic seizure diagnosis were meticulously reviewed and compared. In addition, proper criteria for selecting of appropriate and efficient feature extraction techniques and ML/DL models for epileptic seizure diagnosis were also discussed. Findings from this study will aid researchers in deciding the most efficient ML/DL models with optimal feature extraction methods to improve the performance of EEG-based epileptic seizure diagnosis.
Epilepsy is a chronic neurological illness that affects millions of people throughout the world. Epilepsy affects around 50 million people globally. It is estimated that if epilepsy is correctly diagnosed and treated, up to 70% of people with the condition will be seizure-free. There is a need to detect epilepsy at the initial stages to reduce symptoms by medications and other strategies. We use Epileptic Seizure Recognition dataset to train the model which is provided by UCI Machine Learning Repository. There are 179 attributes and 11,500 unique values in this dataset. MLP, PCA with RF, QDA, LDA, and PCA with ANN were applied among them; PCA with ANN provided the better metrics. For the metrics, we received the following findings. It is 97.55% Accuracy, 94.24% Precision, 91.48% recall, 83.38% hinge loss, and 2.32% mean squared error.
Electroencephalography (EEG) signals are commonly used to identify and diagnose brain disorders. Each EEG normal waveform consists of the following wave forms: Gamma(γ) wave, Beta (β) wave, Alpha (α), Theta (θ), and Delta (δ). The term Neurological Diseases " NurDis " is used to describe a variety of conditions that affect the nervous Epilepsy, neuro infections (bacterial and viral), brain tumors, cerebrovascular diseases, Alzheimer's disease, and various dementia are all examples of neurological disorders. Encephalitis is one of the illnesses that affects the brain. The EEG signals used in this paper were from the CHB-MIT Scalp EEG database. The discrete wavelet transform (DWT) was utilized to extract characteristics from the filtered EEG data. Finally, classifiers such as K Nearest Neighbor (KNN) and Support vector machine (SVM) were used to categorize the EEG signals into normal and pathological signal classes using all of the computed characteristics. In order to categorize the signal in a normal and anomalous group, the KNN and SVM classifiers are employed. For both classifiers, performance assessments (accuracy, sensitivity and specificity) are determined. KNN classifier accuracy is 71.88%, whereas SVM classifier accuracy is 81.23%. The sensitivity of KNN and SVM are 80.14% and 77.31%, respectively. The KNN classification specificity is 69.62% and the SVM classification specificity is 98%. Both classifiers performance is evaluated using the confusion matrix.
Introduction:
Progress in the clinical application of next-generation-sequencing-based techniques has resulted in a dramatic increase in the recognized genetic heterogeneity of the Rett syndrome spectrum (RSS). Our awareness of the considerable overlap with pediatric-onset epilepsies and epileptic/developmental encephalopathies (EE/DE) genes is also growing, and the presence of variable clinical features inside a general frame of commonalities has drawn renewed attention into deep phenotyping.
Methods:
We decided to review the medical literature on atypical Rett syndrome and "Rett-like" phenotypes, with special emphasis on described cases with pediatric-onset epilepsies and/or EE-DE, evaluating Neul's criteria for Rett syndrome and associated movement disorders and notable stereotypies.
Results:
"Rett-like" features were described in syndromic and non-syndromic monogenic epilepsy- and DE/EE-related genes, in "intellectual disability plus epilepsy"-related genes and in neurodegenerative disorders. Additionally, prominent stereotypies can be observed in monogenic complex neurodevelopmental disorders featuring epilepsy with or without autistic features outside of the RSS.
Conclusions:
Patients share a complex neurodevelopmental and neurological phenotype (developmental delay, movement disorder) with impaired gait, abnormal tone and hand stereotypies. However, the presence and characteristics of regression and loss of language and functional hand use can differ. Finally, the frequency of additional supportive criteria and their distribution also vary widely.
Epilepsy remains one of the most common chronic neurological disorders; hence, there is a need to further investigate various models for automatic detection of seizure activity. An effective detection model can be achieved by minimizing the complexity of the model in terms of trainable parameters while still maintaining high accuracy. One way to achieve this is to select the minimum possible number of features. In this paper, we propose a long short-term memory (LSTM) network for the classification of epileptic EEG signals. Discrete wavelet transform (DWT) is employed to remove noise and extract 20 eigenvalue features. The optimal features were then identified using correlation and P value analysis. The proposed method significantly reduces the number of trainable LSTM parameters required to attain high accuracy. Finally, our model outperforms other proposed frameworks, including popular classifiers such as logistic regression (LR), support vector machine (SVM), K-nearest neighbor (K-NN) and decision tree (DT).
Seizure associated with abnormal brain activities caused by epileptic disorder is widely typical and has many symptoms, such as loss of awareness and unusual behavior as well as confusion. In this paper, a classification of the Epileptic Seizure dataset was done using different classifiers. It was shown that the Random Forest classifier outperformed K- Nearest Neighbor (K-NN), Naïve Bayes, Logistic Regression, Decision Tree (D.T.), Random Tree, J48, Stochastic Gradient Descent (S.G.D.) classifiers with 97.08% Accuracy, ROC = 0.996, and RMSE = 0.1527. Sensitivity analysis for some of these classifiers was performed to study the performance of the classifier to classify the Epileptic Seizure dataset with respect to some changes in their parameters. Then a prediction of the dataset using feature selection based on attributes variance was also performed.
Documents are stored in digital form across several organizations. Printing this amount of data and placing it into folders instead of digitally is against the practical, economical, and ecological perspective. An efficient way of retrieving data from digitally stored documents is also required. This article presents a real-time supervised learning technique for document classification based on Deep Convolutional Neural Network, which aims to reduce the impact of adverse document image issues such as signatures, marks, logo, and handwritten notes. The proposed technique's major steps include data augmentation, feature extraction using pre-trained neural network models, feature fusion, and feature selection. We propose a novel data augmentation technique, which normalizes the imbalanced dataset using the secondary dataset, RVL-CDIP. The DCNN features are extracted using the VGG19 and AlexNet networks. The extracted features are fused, and the fused feature vector is optimized by applying a Pearson Correlation Coefficient based technique to select the optimized features while removing the redundant features. The proposed technique is tested on the Tobacco-3482 dataset, which gives a classification accuracy of 93.1% using a cubic Support Vector Machine classifier, proving the validity of the proposed technique.
Recently, artificial intelligence and machine learning in general have demonstrated remarkable performances in many tasks, from image processing to natural language processing, especially with the advent of deep learning (DL). Along with research progress, they have encroached upon many different fields and disciplines. Some of them require high level of accountability and thus transparency, for example, the medical sector. Explanations for machine decisions and predictions are thus needed to justify their reliability. This requires greater interpretability, which often means we need to understand the mechanism underlying the algorithms. Unfortunately, the blackbox nature of the DL is still unresolved, and many machine decisions are still poorly understood. We provide a review on interpretabilities suggested by different research works and categorize them. The different categories show different dimensions in interpretability research, from approaches that provide "obviously" interpretable information to the studies of complex patterns. By applying the same categorization to interpretability in medical research, it is hoped that: 1) clinicians and practitioners can subsequently approach these methods with caution; 2) insight into interpretability will be born with more considerations for medical practices; and 3) initiatives to push forward data-based, mathematically grounded, and technically grounded medical education are encouraged.
Cardiovascular diseases (CVDs), including asymptomatic myocardial ischemia, angina, myocardial infarction, and ischemic heart failure, are the leading cause of death globally. Early detection and treatment of CVDs significantly contribute to the prevention or delay of cardiovascular death. Electrocardiogram (ECG) records the electrical impulses generated by heart muscles, which reflect regular or irregular beating activity. Computer-aided techniques provide fast and accurate tools to identify CVDs using a patient's ECG signal, which have achieved great success in recent years. Latest computational diagnostic techniques based on ECG signals for estimating CVDs conditions are summarized here. The procedure of ECG signals analysis is discussed in several subsections, including data preprocessing, feature engineering, classification, and application. In particular, the End-to-End models integrate feature extraction and classification into learning algorithms, which not only greatly simplifies the process of data analysis, but also shows excellent accuracy and robustness. Portable devices enable users to monitor their cardiovascular status at any time, bringing new scenarios as well as challenges to the application of ECG algorithms. Computational diagnostic techniques for ECG signal analysis show great potential for helping health care professionals, and their application in daily life benefits both patients and sub-healthy people.
Epileptic seizure is one of the common neurological disorder now a day. But this is curable if it can be detected in the early stage. So, this research become a necessity in the early prediction of epileptic seizure. A complete and reliable system can classify the epileptic seizure patients and the states of epileptic seizure. This research explores a supervised machine learning and deep learning model for the classification of epileptic seizure patients from the Epileptic Seizure dataset of UCI machine learning repository. The dataset has 11,500 instances; every information contains 178 attributes. XGBoost is used for the Machine learning approach and ANN is used for Deep learning approach. The proposed ANN algorithm has the improved accuracy and accurately classified the epileptic seizure class patients. 10-fold cross validation is used for the validation purpose. XGBoost acquires 96.6% test accuracy and ANN acquires 98.26% test accuracy. The proposed Deep Learning approach has out-performed the conventional epileptic seizure classifier algorithms. Additionally, the Deep learning model enhances the performance of epileptic seizure detection.
Electroencephalogram (EEG) is one of the most powerful tools that offer valuable information related to different abnormalities in the human brain. One of these abnormalities is the epileptic seizure. A framework is proposed for detecting epileptic seizures from EEG signals recorded from normal and epileptic patients. The suggested approach is designed to classify the abnormal signal from the normal one automatically. This work aims to improve the accuracy of epileptic seizure detection and reduce computational costs. To address this, the proposed framework uses the 54-DWT mother wavelets analysis of EEG signals using the Genetic algorithm (GA) in combination with other four machine learning (ML) classifiers: Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Artificial Neural Network (ANN), and Naive Bayes (NB). The performance of 14 different combinations of two-class epilepsy detection is investigated using these four ML classifiers. The experimental results show that the four classifiers produce comparable results for the derived statistical features from the 54-DWT mother wavelets; however, the ANN classifier achieved the best accuracy in most datasets combinations, and it outperformed the other examined classifiers.
The robustness and computational load are the key challenges in motor imagery (MI) based on electroencephalography (EEG) signals to decode for the development of practical brain-computer interface (BCI) systems. In this study, we propose a robust and simple automated multivariate empirical wavelet transform (MEWT) algorithm for the decoding of different MI tasks. The main contributions of this study are four-fold. First, the multiscale principal component analysis method is utilized in the preprocessing module to obtain robustness against noise. Second, a novel automated channel selection strategy is proposed and then is further verified with comprehensive comparisons among three different strategies for decoding channel combination selection. Third, a sub-band alignment method by utilizing MEWT is adopted to obtain joint instantaneous amplitude and frequency components for the first time in MI applications. Four, a robust correlation-based feature selection strategy is applied to largely reduce the system complexity and computational load. Extensive experiments for subject-specific and subject independent cases are conducted with the three-benchmark datasets from BCI competition III to evaluate the performances of the proposed method by employing typical machine-learning classifiers. For subject-specific case, experimental results show that an average sensitivity, specificity and classification accuracy of 98% was achieved by employing multilayer perceptron neural networks, logistic model tree and least-square support vector machine (LS-SVM) classifiers, respectively for three datasets, resulting in an improvement of upto 23.50% in classification accuracy as compared with other existing method. While an average sensitivity, specificity and classification accuracy of 93%, 92.1% and 91.4% was achieved for subject independent case by employing LS-SVM classifier for all datasets with an increase of up to 18.14% relative to other existing methods. Results also show that our proposed algorithm provides a classification accuracy of 100% for subjects with small training size in subject-specific case, and for subject independent case by employing a single source subject. Such satisfactory results demonstrate the great potential of the proposed MEWT algorithm for practical MI EEG signals classification.
The classification recognition rate of motor imagery is a key factor to improve the performance of brain–computer interface (BCI). Thus, we propose a feature extraction method based on discrete wavelet transform (DWT), empirical mode decomposition (EMD), and approximate entropy. Firstly, the electroencephalogram (EEG) signal is decomposed into a series of narrow band signals with DWT, then the sub-band signal is decomposed with EMD to get a set of stationary time series, which are called intrinsic mode functions (IMFs). Secondly, the appropriate IMFs for signal reconstruction are selected. Thus, the approximate entropy of the reconstructed signal can be obtained as the corresponding feature vector. Finally, support vector machine (SVM) is used to perform the classification. The proposed method solves the problem of wide frequency band coverage during EMD and further improves the classification accuracy of EEG signal motion imaging
Currently, epilepsy disease (ED) is considered to be one of the gradual decreases in brain function over a period of several months or years. Seizure status is the primary common cause of epilepsy disease. The main goal of this study was to discover seizure and epilepsy status using the predication algorithm on the test results achieved from patient medical reports. This study proposed an automatic epilepsy diagnostic method based on a self-organization map method (SOM) using a radial basis function neural networks (RBF) approach. The hybrid technique sought to enhance epilepsy diagnosis precision and to decrease the misdiagnosis of seizure disease. The SOM algorithm was employed to differentiate the unknown patterns of the seizure and epilepsy dataset. The experiments were performed on various RBF neural network algorithms with integrated SOM algorithms to predict and classify the standard epilepsy disease dataset. The hybrid method was tested on the UCI epilepsy dataset. The overall detection accuracy with 10-fold cross validation using SOM-RBF method achieved 97.47%. The results were compared with other modern classification techniques for seizure prediction and detection in terms of evaluation factor.
The aim of this study is to establish a hybrid model for epileptic seizure detection with genetic algorithm (GA) and particle swarm optimization (PSO) to determine the optimum parameters of support vector machines (SVMs) for classification of EEG data. SVMs are one of the robust machine learning techniques and have been extensively used in many application areas. The kernel parameter’s setting for SVMs in training process effects the classification accuracy. We used GA- and PSO-based approach to optimize the SVM parameters. Compared to the GA algorithm, the PSO-based approach significantly improves the classification accuracy. It is shown that the proposed Hybrid SVM can reach a classification accuracy of up to 99.38% for the EEG datasets. Hence, the proposed Hybrid SVM is an efficient tool for neuroscientists to detect epileptic seizure in EEG.
Epilepsy detection from electrical characteristics of EEG signals obtained from the brain of undergone subject is a challenge task for both research and neurologist due to the non-stationary and chaotic nature of EEG signals. As epileptic EEG signals contain huge fluctuating information about the functional behavior of the brain, it is hard to distinguish the fundamental dynamic, complex network of EEG signals without considering the strength among the nodes as they are connected with each other on the basis of these strengths. The prior research on natural visibility graph did not consider this issue in epileptic seizure, although it is a very important key point to have representative information from the signals. Hence, this study aims to introduce a new idea for epilepsy detection using complex network statistical properties by measuring different strengths of the edges in natural visibility graph theory, which is considered as weight. Thus the proposed method is named as “Weighted Visibility Graph”. In this proposed method, firstly, the epileptic EEG signals are transformed into complex network and then two important statistical properties of a network such as Modularity and Average Weighted Degree used for extracting the imperative characteristics from a network of EEG signals. After that, the extracted features are evaluated by two modern machine-learning classifiers such as, support vector machine (SVM) with different kernel function and k-nearest neighbor (KNN). The experimental results demonstrate that the combined effect of both features is valuable for network metrics to characterize the EEG time series signals in case of weighted complex network generating up to 100% classification accuracy.
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 14 different combinations of two-class epilepsy detection is studied using naïve Bayes (NB) and k-nearest neighbor (k-NN) classifiers for the derived statistical features from DWT. It has been found that the NB classifier performs better and shows an accuracy of 100% for the individual and combined statistical features derived from the DWT values of normal eyes open and epileptic EEG data provided by the University of Bonn, Germany. It has been found that the computation time of NB classifier is lesser than k-NN to provide better accuracy. So, the detection of an epileptic seizure based on DWT statistical features using NB classifiers is more suitable in real time for a reliable, automatic epileptic seizure detection system to enhance the patient's care and the quality of life.
The study reported herein proposes a new method for the diagnosis of epilepsy from electroencephalography (EEG) signals based on complex classifiers. To carry out the study, first the features of EEG data are extracted using a dual-tree complex wavelet transformation (DTCWT) at different levels of granularity to obtain size reduction. In subsequent phases, five features (based on statistical measurements-maximum value, minimum value, arithmetic mean, standard deviation, median value) are obtained by using the feature vectors, and are presented as the input dimension to the complex-valued neural networks (CVANN). The evaluation of the proposed method is conducted using the k-fold cross-validation methodology, reporting on classification accuracy, sensitivity and specificity. The proposed method is tested using a benchmark EEG dataset and high accuracy rates were obtained. The stated results show that the proposed method can be used to design an accurate classification system for epilepsy diagnosis.
Almost one-third of people with epilepsy continue to have seizures despite appropriate antiepileptic drug treatment, placing them at considerable risk of cognitive and psychosocial dysfunction and death. We recommend early referral to an epilepsy center when seizures are difficult to control.
We compare dynamical properties of brain electrical activity from different recording regions and from different physiological and pathological brain states. Using the nonlinear prediction error and an estimate of an effective correlation dimension in combination with the method of iterative amplitude adjusted surrogate data, we analyze sets of electroencephalographic (EEG) time series: surface EEG recordings from healthy volunteers with eyes closed and eyes open, and intracranial EEG recordings from epilepsy patients during the seizure free interval from within and from outside the seizure generating area as well as intracranial EEG recordings of epileptic seizures. As a preanalysis step an inclusion criterion of weak stationarity was applied. Surface EEG recordings with eyes open were compatible with the surrogates' null hypothesis of a Gaussian linear stochastic process. Strongest indications of nonlinear deterministic dynamics were found for seizure activity. Results of the other sets were found to be inbetween these two extremes.
Early detection and proper treatment of epilepsy is essential and meaningful to those who suffer from this disease. The adoption of deep learning (DL) techniques for automated epileptic seizure detection using electroencephalography (EEG) signals has shown great potential in making the most appropriate and fast medical decisions. However, DL algorithms have high computational complexity and suffer low accuracy with imbalanced medical data in multi seizure-classification task. Motivated from the aforementioned challenges, we present a simple and effective hybrid DL approach for epileptic seizure detection in EEG signals. Specifically, first we use a K-means Synthetic minority oversampling technique (SMOTE) to balance the sampling data. Second, we integrate a 1D Convolutional Neural Network (CNN) with a Bidirectional Long Short-Term Memory (BiLSTM) network based on Truncated Backpropagation Through Time (TBPTT) to efficiently extract spatial and temporal sequence information while reducing computational complexity. Finally, the proposed DL architecture uses softmax and sigmoid classifiers at the classification layer to perform multi and binary seizure-classification tasks. In addition, the 10-fold cross-validation technique is performed to show the significance of the proposed DL approach. Experimental results using the publicly available UCI epileptic seizure recognition data set shows better performance in terms of precision, sensitivity, specificity, and F1-score over some baseline DL algorithms and recent state-of-the-art techniques.
Detecting dementia presents a barrier to advancing individualized healthcare. Electroencephalographic (EEG) signals’ nonlinear nature has been characterized using entropies. While a working memory (WM), the EEGs of 5 patients suffering vascular dementia (VD), 15 patients had stroke-related mild cognitive impairment (SMCI), and 15 healthy normal control (NC) participants were evaluated in this study. A four-step framework for the automatic identification of dementia is provided, with the first stage employing the newly developed automatic independent component analysis and wavelet (AICA-WT) method. In the second stage, nonlinear entropy features using fuzzy entropy (FuzzEn), fluctuation-based dispersion entropy (FDispEn), and bubble entropy (BubbEn) were utilized to extract various dynamical properties from multi-channel EEG signals derived from patients with dementia. A statistical examination of the individual performance was conducted using analysis of variance (ANOVA) to determine the degree of EEG complexity across brain regions. Afterwards, the nonlinear local tangent space alignment (LSTA) dimensionality reduction approach was utilized to enhance the automatic diagnosis of dementia patients’. Using k-nearest neighbors (kNN), support vector machine (SVM), and decision tree (DT) classifiers, the impairment of post-stroke patients was finally identified. BubbEn is chosen to develop a new BubbEn-LTSA mapping process for creating the innovative AICA-WT-BubbEn-LTSA dementia recognition framework, which is the basis for an automated VD detection.
In recent years, automated seizure identification with electroencephalogram (EEG) signals has received considerable attention and appears to be an appropriate approach for diagnosis and treatment of the disease. This paper analyze the ability of Hjorth parameters for seizure detection using EEG signals. The tunable-Q wavelet transform (TQWT) is applied to decompose an EEG signal into various sub-bands at different levels. The Hjorth parameters namely activity, mobility and complexity are studied over the decomposed components. The University of Bonn, Germany dataset is studied to validate the proposed method with including seizure, seizure-free and normal categories of EEG signal. Classification findings show that the proposed technique with estimating the Hjorth parameters preserves efficiency and is appropriate for automated identification of epileptic seizures. In this work, very high classification accuracy is achieved in various set on combinations. The proposed technique is compared with state-of-the-art approaches available in the literature.
The detection of epileptic seizures from electroencephalogram (EEG) signals is traditionally performed by clinical experts through visual inspection. It is a long process, is error prone, and requires a highly trained expert. In this research, a new method is presented for seizure classification for EEG signals using a dual-tree complex wavelet transform (DT-CWT) and fast Fourier transform (FFT) coupled with a least square support vector machine (LS-SVM) classifier. In this method, each EEG signal is divided into four segments. Each segment is further split into smaller sub-segments. The DT-CWT is applied to decompose each sub-segment into detailed and approximation coefficients (real and imaginary parts). The obtained coefficients by the DT-CWT at each decomposition level are passed through an FFT to identify the relevant frequency bands. Finally, a set of effective features are extracted from the sub-segments, and are then forwarded to the LS-SVM classifier to classify epileptic EEGs. In this paper, two epileptic EEG databases from Bonn and Bern Universities are used to evaluate the extracted features using the proposed method. The experimental results demonstrate that the method obtained an average accuracy of 97.7% and 96.8% for the Bonn and Bern databases, respectively. The results prove that the proposed DT-CWT and FFT based features extraction is an effective way to extract discriminative information from brain signals. The obtained results are also compared to those by k-means and Naïve Bayes classifiers as well as with the results from the previous methods reported for classifying epileptic seizures and identifying the focal and non-focal EEG signals. The obtained results show that the proposed method outperforms the others and it is effective in detecting epileptic seziures in EEG signals. The technique can be adopted to aid neurologists to better diagnose neurological disorders and for an early seizure warning system.
Epilepsy is a disorder of the brain characterized seizures and requires constant monitoring in serious patients. Electroencephalogram (EEG) signals are frequently used in epilepsy diagnosis and monitoring. Deep learning methods, such as CNN, LSTM, and RNN have been widely applied to EEG signals with varying levels of success. A new paradigm of battery packed wearable gadgets has recently gained popularity that constantly monitors a patient’s signals. These gadgets acquire the data, perform some preprocessing, and are connected to the cloud for epileptic diagnosis and monitoring. Power consumption due to data transmission is a major issue in these devices. Moreover, in a constant monitoring environment, the number of classes to be identified are usually higher and overlapping. In this context, we propose a new framework for EEG based epilepsy detection which requires a low data transmission while achieving similar or better accuracy on a multiclass problem. We propose a new preprocessing mechanism that uses adaptive rate sampling, modified activity selection, filtering and wavelet decomposition to extract a handful of highly discriminatory features which needs to be transmitted instead of the entire EEG waveform. Furthermore, we also propose a novel ensemble of sub-problems based classification paradigm to achieve high accuracy using the reduced data. Our proposed solution shows many-fold increase in computational gains and an accuracy of 100% and 99.38% on 2-class problem when tested respectively on the popular University of Bonn and CHB-MIT datasets. The accuracy of 99.6% on 3-class, 96% on 4-class, and 92% on 5-class problem is secured for the University of Bonn dataset.
Heterogeneous palmprint recognition has attracted considerable research attention in recent years because it has the potential to greatly improve the recognition performance for personal authentication. In this article, we propose a simultaneous heterogeneous palmprint feature learning and encoding method for heterogeneous palmprint recognition. Unlike existing hand-crafted palmprint descriptors that usually extract features from raw pixels and require strong prior knowledge to design them, the proposed method automatically learns the discriminant binary codes from the informative direction convolution difference vectors of palmprint images. Differing from most heterogeneous palmprint descriptors that individually extract palmprint features from each modality, our method jointly learns the discriminant features from heterogeneous palmprint images so that the specific discriminant properties of different modalities can be better exploited. Furthermore, we present a general heterogeneous palmprint discriminative feature learning model to make the proposed method suitable for multiple heterogeneous palmprint recognition. Experimental results on the widely used PolyU multispectral palmprint database clearly demonstrate the effectiveness of the proposed method.
Electroencephalography signals obtained from the brain‘s electrical activity are commonly used for the diagnosis of neurological diseases. These signals indicate the electrical activity in the brain and contain information about the brain. Epilepsy, one of the most important diseases in the brain, manifests itself as a result of abnormal pathological oscillating activity of a group of neurons in the brain. Automated systems that employed the electroencephalography signals are being developed for the assessment and diagnosis of epileptic seizures. The aim of this study is to focus on the effectiveness of stacking ensemble approach based model for predicting whether there is epileptic seizure or not. So, this study enables the readers and researchers to examine the proposed stacking ensemble model. The benchmark clinical dataset provided by Bonn University was used to assess the proposed model. Comparative experiments were conducted by utilizing the proposed model and the base deep neural networks model to show the effectiveness of the proposed model for seizure detection. Experiments show that the proposed model is proven to be competitive to base DNN model. The results indicate that the performance of the epileptic seizure detection by the stacking ensemble based deep neural networks model is high; especially the average accuracy value of 97.17%. Also, its average sensitivity with 93.11% is superior to the base DNN model. Thus, it can be said that the proposed model can be included in an expert system or decision support system. In this context, this system would be precious for the clinical diagnosis and treatment of epilepsy.
Analysis of electroencephalogram (EEG) signal is crucial due to its non-stationary characteristics, which could lead the way to proper detection method for the treatment of patients with neurological abnormalities, especially for epilepsy. The performance of EEG-based epileptic seizure detection relies largely on the quality of selected features from an EEG data that characterize seizure activity. This paper presents a novel analysis method for detecting epileptic seizure from EEG signal using Improved Correlation-based Feature Selection method (ICFS) with Random Forest classifier (RF). The analysis involves, first applying ICFS to select the most prominent features from the time domain, frequency domain, and entropy based features. An ensemble of Random Forest (RF) classifiers is then learned on the selected set of features. The experimental results demonstrate that the proposed method shows better performance compared to the conventional Correlation-based method and also outperforms some other state-of-the-art methods of epileptic seizure detection using the same benchmark EEG dataset.
In present work, a methodology for automatic vigilance level detection of human brain using nonlinear features of Electroencephalogram (EEG) signals is presented. Vigilance level detection methodology consists of three steps, EEG channels selection, feature extraction and classification. EEG signals obtained from 64 channels are sub-divided into four frequency sub-bands i.e. alpha, beta, delta and theta. Channel selection criteria Maximum Energy to Shannon Entropy ratio is applied on each frequency band to select appropriate EEG channels. EEG signals obtained from selected channels are further divided into frequency sub-bands i.e. alpha, beta and alpha–beta bands. Three nonlinear features such as Higuchi fractal dimension, Petrosian fractal dimension and Detrended Fluctuation Analysis are calculated to prepare three feature vectors respective to each frequency sub-bands. Three machine learning techniques are used for vigilance level detection such as Support Vector Machine, Least Square-Support Vector Machine and Artificial Neural Network.
This paper explores the three different methods to explicitly recognize the healthy and epileptic EEG signals: Modified, Improved, and Advanced forms of Generalized Fractal Dimensions (GFD). The newly proposed scheme is based on GFD and the discrete wavelet transform (DWT) for analyzing the EEG signals. First EEG signals are decomposed into approximation and detail coefficients using DWT and then GFD values of the original EEGs, approximation and detail coefficients are computed. Significant differences are observed among the GFD values of the healthy and epileptic EEGs allowing us to classify seizures with high accuracy. It is shown that the classification rate is very less accurate without DWT as a preprocessing step. The proposed idea is illustrated through the graphical and statistical tools. The EEG data is further tested for linearity by using normal probability plot and we proved that epileptic EEG had significant nonlinearity whereas healthy EEG distributed normally and similar to Gaussian linear process. Therefore, we conclude that the GFD and the wavelet decomposition through DWT are the strong indicators of the state of illness of epileptic patients.
The newly inaugurated Research Resource for Complex Physiologic Signals, which was created under the auspices of the National Center for Research Resources of the National Institutes of Health, is intended to stimulate current research and new investigations in the study of cardiovascular and other complex biomedical signals. The resource has 3 interdependent components. PhysioBank is a large and growing archive of well-characterized digital recordings of physiological signals and related data for use by the biomedical research community. It currently includes databases of multiparameter cardiopulmonary, neural, and other biomedical signals from healthy subjects and from patients with a variety of conditions with major public health implications, including life-threatening arrhythmias, congestive heart failure, sleep apnea, neurological disorders, and aging. PhysioToolkit is a library of open-source software for physiological signal processing and analysis, the detection of physiologically significant events using both classic techniques and novel methods based on statistical physics and nonlinear dynamics, the interactive display and characterization of signals, the creation of new databases, the simulation of physiological and other signals, the quantitative evaluation and comparison of analysis methods, and the analysis of nonstationary processes. PhysioNet is an on-line forum for the dissemination and exchange of recorded biomedical signals and open-source software for analyzing them. It provides facilities for the cooperative analysis of data and the evaluation of proposed new algorithms. In addition to providing free electronic access to PhysioBank data and PhysioToolkit software via the World Wide Web (http://www.physionet. org), PhysioNet offers services and training via on-line tutorials to assist users with varying levels of expertise.
EEG based epilepsy diagnosis system using reconstruction phase space and Naïve Bayes classifier
- Obeidat
Obeidat MA, Mansour AM. EEG based epilepsy diagnosis system using reconstruction phase space and Naïve Bayes classifier. WSEAS Trans Circuits Syst
2018;17.
Tunable-Q wavelets transform based filter banks for non-stationary signals analysis and classification
- A Nishad
Nishad A. Tunable-Q wavelets transform based filter banks for non-stationary
signals analysis and classification. 2019.
epileptic seizure recognition data set
- Uml
- Repository
UML. Repository, epileptic seizure recognition data set. 2022, Available online:https://archive.ics.uci.edu/ml/datasets/Epileptic+Seizure+Recognition, (accessed December 7, 2022).
Recognition enhancement of dementia patients' working memory using entropy-based features and local tangent space alignment algorithm. In: Advances in non-invasive biomedical signal sensing and processing with machine learning
- N K Al-Qazzaz
- Shbm Ali
- S A Ahmad
Al-Qazzaz NK, Ali SHBM, Ahmad SA. Recognition enhancement of dementia
patients' working memory using entropy-based features and local tangent space
alignment algorithm. In: Advances in non-invasive biomedical signal sensing and
processing with machine learning. Springer; 2023, p. 345-73.
Efficient approach to detect epileptic seizure using machine learning models for modern healthcare system
- T I Rohan
- Msu Yusuf
- M Islam
- S Roy
Rohan TI, Yusuf MSU, Islam M, Roy S. Efficient approach to detect epileptic
seizure using machine learning models for modern healthcare system. In: 2020
IEEE region 10 symposium (TENSYMP). IEEE; 2020, p. 1783-6.
A hybrid technique for EEG signals evaluation and classification as a step towards to neurological and cerebral disorders diagnosis
- Abdulbaqi