Diagnosis Of Epileptic Seizures Research Articles

A comprehensive comparison of handcrafted features and convolutional autoencoders for epileptic seizures detection in EEG signals

Epilepsy, a brain disease generally associated with seizures, has tremendous effects on people’s quality of life. Diagnosis of epileptic seizures is commonly performed on electroencephalography (EEG) signals, and by using computer-aided diagnosis systems (CADS), neurologists can diagnose epileptic seizure stages more accurately. In these systems, a mandatory stage is feature extraction, performed by handcrafting features or learning them, ordinarily by a deep neural net. While researches in this field commonly show the value of a group of limited features, yet an accurate comparison between different suggested features is essential. In this article, first, a comparison between the importance of 50 different handcrafted features for seizure detection is presented. Additionally, the computational complexity of features is investigated as well. Then the best features based on Fisher scores are picked to classify signals on a benchmark dataset for evaluation. Additionally, a convolutional autoencoder with five layers is applied to learn features in order to have a complete comparison among feature extraction approaches. Finally, a hybrid method is employed, which combines handcrafted features and encoding of autoencoder to reach high performance in seizure detection in EEG signals.

An Accurate Automatic Epileptic Seizure Diagnosis With Logistic Regression Using Electro encephalography Signals

An Accurate Automatic Epileptic Seizure Diagnosis With Logistic Regression Using Electro encephalography Signals

Adaptive boost LS-SVM classification approach for time-series signal classification in epileptic seizure diagnosis applications

Epileptic seizures are characterised by abnormal neuronal discharge, causing notable disturbances in electrical activities of the human brain. Traditional methods based on manual approaches applied in seizure detection in electroencephalograms (EEG) have drawbacks (e.g., time constraint, lack of effective feature identification relative to disease symptoms and susceptibility to human errors) that can lead to inadequate treatment options. Designing an automated expert system to detect epileptic seizures can proactively support a neurologist’s effort to improve authenticity, speed and accuracy of detecting signs of a seizure. We propose a novel two-phase EEG classification technique to detect seizures from EEG by employing covariance matrix coupled with Adaptive Boosting Least Square-Support Vector Machine (i.e., AdaBoost LS-SVM) framework. In first phase, the covariance matrix is employed as a dimensionality reduction tool with feature extraction applied to analyse epileptic patients’ EEG records. Initially, each single EEG channel is partitioned into respective k segment with m clusters. Subsequently, covariance matrix is adopted with eigenvalues of each cluster extracted and tested through statistical metrics to identify the most representative, optimally classified features. In the second phase, a robust classifier (i.e., AB-LS-SVM) is proposed to resolve issues of unbalanced data, to detect epileptic events, yielding a high classification accuracy compared to its competing counterparts. The results demonstrates that AB-LS-SVM (optimised by a covariance matrix) is able to achieve satisfactory results (>99% accuracy) for eleven prominent features in EEG signals. The results are compared with state-of-art algorithms (i.e., k-means, SVM, k-nearest neighbour, Random Forest) on identical databases, demonstrating the capability of AB-LS-SVM method as a promising diagnostic tool and its practicality for implementation in seizure detection. The study avers that the proposed approach can aid clinicians in diagnosis or interventions to treat epileptic disease, including a potential use in expert systems where EEG needs to be classified through pattern recognition.

Developing multi-component dictionary-based sparse representation for automatic detection of epileptic EEG spikes

Epilepsy is a common brain disease which affect approximately one percent of people worldwide and its onset can be diagnosed by detecting electroencephalography (EEG) spike artificially. In this research, we propose a modified sparse representation algorithm for automated epileptic spikes detection, motivated by the finding that EEG tends to be sparse. Using Gaussian function and its derivations, a redundant and multi-component dictionary is firstly constructed and followed by achieving the sparse approximation of EEG based on orthogonal matching pursuit (OMP) algorithm. At last, the morphological structure of the reconstructed EEG is represented quantitatively so that the spikes are detected and located. Experimental results show that the proposed scheme is qualified to detect epileptic spikes automatically employing four EEG subsets containing 90 epochs of EEG data with a satisfactory accuracy of not less than 97.33%. Comparing with other techniques, our approach reaches higher accuracy and is more adequate for epileptic spikes detection with lower computational complexity. As a whole, our scheme provides a new tool for the diagnosis of epileptic seizures.

Effective epileptic seizure detection based on the event-driven processing and machine learning for mobile healthcare

Mobile healthcare is a promising approach. It is realized by using the biomedical implants that are connected to the cloud. A framework for the precise and effective diagnosis of epileptic seizures is designed in this context. To achieve real-time compression and effective signal processing and transmission, it uses an intelligent event-driven electroencephalogram (EEG) signal acquisition. Experimental results show that grace of the event-driven nature an overall 3.3 fold compression and transmission bandwidth usage reduction is achieved by the devised method compared to the conventional counterparts. It promises a notable decrease in the post analysis and classification processing activity. The system performance is studied by using a standard three class EEG epileptic seizure dataset. The highest classification accuracy of 97.5% is secured for a mono-class. The best average classification accuracy of 96.4% is attained for three-classes. Comparison of the system with classical equivalents is made. Results demonstrate more than threefold and sevenfold of outperformance respectively in terms of compression gain and processing efficiency while confirming a comparable classification precision.

Stacking ensemble based deep neural networks modeling for effective epileptic seizure detection

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.

Assessment of the Predictive Value of Outpatient Smartphone Videos for Diagnosis of Epileptic Seizures

Misdiagnosis of epilepsy is common. Video electroencephalogram provides a definitive diagnosis but is impractical for many patients referred for evaluation of epilepsy. To evaluate the accuracy of outpatient smartphone videos in epilepsy. This prospective, masked, diagnostic accuracy study (the OSmartViE study) took place between August 31, 2015, and August 31, 2018, at 8 academic epilepsy centers in the United States and included a convenience sample of 44 nonconsecutive outpatients who volunteered a smartphone video during evaluation and subsequently underwent video electroencephalogram monitoring. Three epileptologists uploaded videos for physicians from the 8 epilepsy centers to review. Measures of performance (accuracy, sensitivity, specificity, positive predictive value, and negative predictive value) for smartphone video-based diagnosis by experts and trainees (the index test) were compared with those for history and physical examination and video electroencephalogram monitoring (the reference standard). Forty-four eligible epilepsy clinic outpatients (31 women [70.5%]; mean [range] age, 45.1 [20-82] years) submitted smartphone videos (530 total physician reviews). Final video electroencephalogram diagnoses included 11 epileptic seizures, 30 psychogenic nonepileptic attacks, and 3 physiologic nonepileptic events. Expert interpretation of a smartphone video was accurate in predicting a video electroencephalogram monitoring diagnosis of epileptic seizures 89.1% (95% CI, 84.2%-92.9%) of the time, with a specificity of 93.3% (95% CI, 88.3%-96.6%). Resident responses were less accurate for all metrics involving epileptic seizures and psychogenic nonepileptic attacks, despite greater confidence. Motor signs during events increased accuracy. One-fourth of the smartphone videos were correctly diagnosed by 100% of the reviewing physicians, composed solely of psychogenic attacks. When histories and physical examination results were combined with smartphone videos, correct diagnoses rose from 78.6% to 95.2%. The odds of receiving a correct diagnosis were 5.45 times greater using smartphone video alongside patient history and physical examination results than with history and physical examination alone (95% CI, 1.01-54.3; P = .02). Outpatient smartphone video review by experts has predictive and additive value for diagnosing epileptic seizures. Smartphone videos may reliably aid psychogenic nonepileptic attacks diagnosis for some people.

Inflammatory signs and epileptic crisis in patients admitted in an emergency unit.

To evaluate inflammatory signs presented in medical records of patients with a main diagnosis of epileptic seizures, admitted in an emergency unit. Cross-sectional and retrospective study. The sample was composed of 191 medical records, from children, adolescents, adults, and elders, with a clinical diagnosis of epileptic seizures, admitted between June 2016 and June 2017 at the emergency unit of a hospital in Porto Alegre/RS. The prevalent inflammatory signs were tachypnea (33.5%) and/or fever (27.2%) associated with leukocytosis (P=0.030). Children/adolescents had seizures less frequently (P=0.010) and these were due to fever (P=0.000). Adults presented seizures more frequently (P=0.006), which were related to medication/intoxication (P=0.000). In elders, seizures occurred due to metabolic or circulatory abnormalities (P=0.000), less often due to fever (P=0.005). Seizures are related to fever and tachypnea, being caused by different etiologies according to age, being more frequent in adults. Fever is related to leukocytosis, regardless of age.

EEG dataset classification using CNN method

This paper proposes a simple Convolutional Neural Network (CNN) program to classify epileptic seizure. The diagnosis of epileptic seizure involves the identification and different characteristic of the Electroencephalography (EEG) signal. As such, it needs a method for identifying and classifying epileptic seizure. Deep learning is part of a neural network that has the ability and pattern to identify and classify epileptic seizure. CNN has been demonstrated high performance on image classification and pattern detection. In this paper, we combine the continuous wavelet transform (CWT) and CNN to classify epileptic seizure. This experiment uses the wavelet transform to convert signal data of EEG to time-frequency domain images. The output of the wavelet transform is an image that will classify into five attributes. In this experiment, we develop a simple program that will compare with other CNN approach (AlexNet and GoogleNet). The results of this experiment are two kinds of data, accuracy, and loss. The resulting accuracy is 72.49%, and the loss is 0.576. This result has a better learning time than GoogleNet and smaller loss result than AlexNet.

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

Epilepsy, a common neurological disorder, is generally detected by electroencephalogram (EEG) signals. Visual inspection and interpretation of EEGs is a slow, time consuming process that is vulnerable to error and subjective variability. Consequently, several efforts to develop automatic epileptic seizure detection and classification methods have been made. The present study proposes a novel computer aided diagnostic technique (CAD) based on the discrete wavelet transform (DWT) and arithmetic coding to differentiate epileptic seizure signals from normal (seizure-free) signals. The proposed CAD technique comprises three steps. The first step decomposes EEG signals into approximations and detail coefficients using DWT while discarding non-significant coefficients in view of threshold criteria; thus, limiting the number of significant wavelet coefficients. The second step converts significant wavelet coefficients to bit streams using arithmetic coding to compute the compression ratio. In the final step, the compression feature set is standardized, whereupon machine-learning classifiers detect seizure activity from seizure-free signals. We employed the widely used benchmark database from Bonn University to compare and validate the technique with results from prior approaches. The proposed method achieved a perfect classification performance (100% accuracy) for the detection of epileptic seizure activity from EEG data, using both linear and non-liner machine-learning classifiers. This CAD technique can thus be considered robust with an extraordinary detection capability that discriminates epileptic seizure activity from seizure-free and normal EEG activity with simple linear classifiers. The method has the potential for efficient application as an adjunct for the clinical diagnosis of epilepsy.

A strategy combining intrinsic time-scale decomposition and a feedforward neural network for automatic seizure detection

Epilepsy is a common neurological disorder which can occur in people of all ages globally. For the clinical treatment of epileptic patients, the detection of epileptic seizures is of great significance. Objective: Electroencephalography (EEG) is an essential component in the diagnosis of epileptic seizures, from which brain surgeons can detect important pathological information about patient epileptiform discharges. This paper focuses on adaptive seizure detection from EEG recordings. We propose a new feature extraction model based on an adaptive decomposition method, named intrinsic time-scale decomposition (ITD), which is suitable for analyzing non-linear and non-stationary data. Approach: Firstly, using the ITD technique, every EEG recording is decomposed into several proper rotation components (PRCs). Secondly, the instantaneous amplitudes and frequencies of these PRCs can be calculated and then we extract their statistical indices. Furthermore, we combine all these statistical indices of the corresponding five PRCs as the feature vector of each EEG signal. Finally, these feature vectors are fed into a feedforward neural network (FNN) classifier for EEG classification. The whole process of feature extraction proposed in this paper only involves one parameter and the role of the ITD method is based on a piecewise linear function, which makes the computation of the model simple and fast. More useful information for classification can be obtained since we take advantage of both instantaneous amplitude and instantaneous frequency for feature extraction. Main results: We consider the 17 classification problems which contain normal versus epileptic, non-seizure versus seizure and normal versus interictal versus ictal using a FNN classifier which only contains one hidden layer. Experimental results show that the proposed method can catch the discriminative features of EEG signals and obtain comparable results when compared with state-of-the-art detection methods. Significance: Therefore, the proposed system has a great potential in real-time seizure detection and provides physicians with a real-time diagnostic aid in their practice.

Differential diagnosis of epileptic seizure and syncope using machine learning algorithms (P2.5-031)

Differential diagnosis of epileptic seizure and syncope using machine learning algorithms (P2.5-031)

Performance Evaluation of Discrete Wavelet Transform, and Wavelet Packet Decomposition for Automated Focal and Generalized Epileptic Seizure Detection

In the past decades, wavelet transforms are widely employed for characterizing the electroencephalogram (EEG) signals for automatic diagnosis of epileptic seizure. But few vital issues like the classification of epileptic seizure types from normal EEG signals has not yet been benefited with wavelet transforms. Hence, in this paper, the two major types of wavelet transform, namely discrete wavelet transform (DWT) and wavelet packet decomposition (WPD) are employed for the automatic diagnosis of the epileptic seizure and its types. The publicly available KITS EEG database consisting of three groups namely, normal, focal epilepsy and generalized epilepsy are utilized in this work. Four experimental cases namely (i) normal-generalized epilepsy, (ii) normal-focal epilepsy, (iii) normal-focal-generalized and (iv) normal-epilepsy are used to investigate the proposed approach. Further, this paper attempts to identify the best wavelet function from the commonly used seven wavelet families and the level of decomposition required to analyse the EEG signals. The nine statistical features are extracted from the wavelet coefficients and fed into the support vector machine (SVM) classifier. From the experimental result it was found out that the DWT with rbio1.1 attained the highest classification accuracy for all the experimental cases.

Automatic Diagnosis of Epileptic Seizure in Electroencephalography Signals Using Nonlinear Dynamics Features

Epilepsy is one of the most common neurological disorders generally characterized by sudden and recurrent unprovoked seizures, which is commonly diagnosed through visual inspection of electroencephalography (EEG) signals. However, since EEG signals are highly complex, nonlinear and nonstationary in nature, both visual inspection and the existing computed aid detection approaches fail to capture the intrinsic dynamics of seizure events, leading to unsatisfactory detection performance. Therefore, an accurate computer-aided automatic epileptic seizure diagnosis system is valuable and urgently needed in clinical application. In this study, we propose an automatic classification method using nonlinear dynamics features and nonlinear classifiers for determining whether the EEG signals are in interictal or ictal activities, respectively. The main novelty of this paper is that investigating the performance of discernable entropy-based nonlinear dynamics features used for automatic diagnosis of an epileptic seizure. First, the discrete wavelet transform is applied to obtain de-noised and sub-band signals. Second, the nonlinear dynamics features are extracted based on information entropy theory. Subsequently, one-way analysis of variance and forward sequential feature selection technique are employed to identify the most explainable features for seizure detection. Then the selected features are fed to six different classifiers, namely, least square-support vector machine (LS-SVM), K-nearest neighbors, logistic regression, linear discriminant analysis, Naive Bayes classifier, and random forest, for automatic diagnosis of the epileptic seizure. Finally, a public EEG dataset from Bonn University, Germany, is utilized to investigate the performance of the proposed method. The experimental results show that the LS-SVM yields the optimal performance with an accuracy of 99.50%, a sensitivity of 100.00%, and a specificity of 99.40%. Our proposed method shows great potential for real-time diagnosis of the epileptic seizures, which can be deployed in hospitals to assist clinicians in accurate diagnosis of seizures.

A Remote System for the Diagnosis and Emergency Care of Epileptic Seizures

Challenges in the development of a system for the remote diagnosis of epileptic seizures occurring outside medical institutions are discussed, as are electrical stimulation of the vagus nerve to terminate epileptic seizures, automatic formation of a database of measures of epileptic seizures from long-term observations of patients, and production of a data source for physicians for planning therapeutic measures.

Sturge-Weber syndrome in children. Clinical features, diagnosis and approaches to therapy on the example of three clinical cases

Sturge-Weber syndrome belongs to the group of phacomotoses and is characterized by a combined lesion of the skin, eyes, nervous system and internal organs. The clinical course of Sturge-Weber syndrome is quite diverse. Of particular interest in the practice of pediatric neurology and neurosurgery is the fact that 72-90% of patients present with epilepsy. Of particular difficulty is the differential diagnosis of epileptic seizures and stroke-like episodes. The article presents clinical cases of patients with Sturge-Weber syndrome of various ages with epileptic seizures and transient ischemic attacks.

A forecasting tool for prediction of epileptic seizures using a machine learning approach

SummaryECG and EEG signals are very helpful in the early diagnosis of epileptic seizures. The research focuses on analysis of ECG and EEG signals applying a deep learning technique to study early prediction of epileptic seizure. Signal processing methods like Empirical Mode Decomposition, spectral analysis, and statistical methods were used. The algorithms were implemented in MATLAB, and the EEG and ECG data were collected from Physiobank and EPILEPSIAE databases. In the window‐based analysis of low‐frequency spectral area of EEG signals, 78.5% of the cases displayed a significant change as the windows progressed and the onset of seizure was approached. The spectral area of IMF components indicated a possible seizure prediction in 68.9% of the analyzed cases. Considering signals from individual EEG electrodes, the least percentage of seizure prediction was indicated by signals from T4 and F4 electrodes (52.3% and 40.7%, respectively, for spectral peaks and 23.8% and 29.6%, respectively, for spectral area). The results of regression analysis show that prediction of seizures can be possible around 20‐30 minutes prior to the actual occurrence of seizures.

Topological Data Analysis of Single-Trial Electroencephalographic Signals.

Epilepsy is a neurological disorder that can negatively affect the visual, audial and motor functions of the human brain. Statistical analysis of neurophysiological recordings, such as electroencephalogram (EEG), facilitates the understanding and diagnosis of epileptic seizures. Standard statistical methods, however, do not account for topological features embedded in EEG signals. In the current study, we propose a persistent homology (PH) procedure to analyze single-trial EEG signals. The procedure denoises signals with a weighted Fourier series (WFS), and tests for topological difference between the denoised signals with a permutation test based on their PH features persistence landscapes (PL). Simulation studies show that the test effectively identifies topological difference and invariance between two signals. In an application to a single-trial multichannel seizure EEG dataset, our proposed PH procedure was able to identify the left temporal region to consistently show topological invariance, suggesting that the PH features of the Fourier decomposition during seizure is similar to the process before seizure. This finding is important because it could not be identified from a mere visual inspection of the EEG data and was in fact missed by earlier analyses of the same dataset.

Psychiatric comorbidity and traumatic brain injury attribution in patients with psychogenic nonepileptic or epileptic seizures: A multicenter study of US veterans.

To determine the frequency and severity of psychiatric disorders and attribution of seizures to traumatic brain injury (TBI) in veterans with verified psychogenic nonepileptic seizures (PNES) versus epileptic seizures (ES). We studied 333 consecutive admissions to the monitoring units of three Veterans Administration epilepsy centers. All patients underwent continuous video-electroencephalographic recording to define definite PNES or ES. Evaluations included the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, 4th edition, PTSD Checklist, Beck Depression Inventory II, and Patient Seizure Etiology Questionnaire. Interviews and questionnaires were completed prior to final seizure type diagnosis and patient debriefing. The primary outcome measure was a comparison of Axis I psychiatric diagnoses in patients diagnosed with PNES versus ES. A total of 81 patients were diagnosed with PNES, and 70 with ES. PTSD was the most frequent Axis I diagnosis in veterans with PNES (64%, vs 13% of those with ES; P<0.001). Posttraumatic stress disorder (PTSD) was common regardless of deployment to a war theater or combat exposure. Mood, substance abuse, and anxiety disorders were also more common in the PNES group. TBI was cited as a likely cause of seizures by 47% of veterans with PNES versus 25% of those with ES (P=0.01). PTSD and attribution of seizures to TBI were found in 30% of veterans with PNES versus 3% of those with ES (P<0.001). In veterans referred for inpatient seizure evaluation, PTSD was strongly associated with a diagnosis of PNES versus ES. The association of PNES with PTSD, attribution of seizures to TBI, or both, may prompt early consideration of PNES.

Practical Management of Epileptic Seizures and Status Epilepticus in Adult Palliative Care Patients.

In terminally ill patients, paroxysmal or episodic changes of consciousness, movements and behavior are frequent. Due to ambiguous appearance, the correct diagnosis of epileptic seizures (ES) and non-epileptic events (NEE) is often difficult. Treatment is frequently complicated by the underlying condition, and an approach indicated in healthier patients may not always be appropriate in the palliative care setting. This article provides recommendations for diagnosis of ES and NEE and treatment options for ES in adult palliative care patients, including aspects of alternative administration routes for antiepileptic drugs such as intranasal, subcutaneous, or rectal application.