Scalp Electroencephalogram Research Articles

Automatic detection of epileptic seizures using Riemannian geometry from scalp EEG recordings.

This paper proposes a new framework for epileptic seizure detection using non-invasive scalp electroencephalogram (sEEG) signals. The major innovation of the current study is using the Riemannian geometry for transforming the covariance matrices estimated from the EEG channels into a feature vector. The spatial covariance matrices are considered as features in order to extract the spatial information of the sEEG signals without applying any spatial filtering. Since these matrices are symmetric and positive definite (SPD), they belong to a special manifold called the Riemannian manifold. Furthermore, a kernel based on Riemannian geometry is proposed. This kernel maps the SPD matrices onto the Riemannian tangent space. The SPD matrices, obtained from all channels of the segmented sEEG signals, have high dimensions and extra information. For these reasons, the sequential forward feature selection method is applied to select the best features and reduce the computational burden in the classification step. The selected features are fed into a support vector machine (SVM) with an RBF kernel to classify the feature vectors into seizure and non-seizure classes. The performance of the proposed method is evaluated using two long-term scalp EEG (CHB-MIT benchmark and private) databases. Experimental results on all 23 subjects of the CHB-MIT database reveal an accuracy of 99.87%, a sensitivity of 99.91%, and a specificity of 99.82%. In addition, the introduced algorithm is tested on the private sEEG signals recorded from 20 patients, having 1380 seizures. The proposed approach achieves an accuracy, a sensitivity, and a specificity of 98.14%, 98.16%, and 98.12%, respectively. The experimental results on both sEEG databases demonstrate the effectiveness of the proposed method for automated epileptic seizure detection, especially for the private database which has noisier signals in comparison to the CHB-MIT database. Graphical Abstract Block diagram of the proposed epileptic seizure detection algorithm.

Application of a convolutional neural network for fully-automated detection of spike ripples in the scalp electroencephalogram

BackgroundA reliable biomarker to identify cortical tissue responsible for generating epileptic seizures is required to guide prognosis and treatment in epilepsy. Combined spike ripple events are a promising biomarker for epileptogenic tissue that currently require expert review for accurate identification. This expert review is time consuming and subjective, limiting reproducibility and high-throughput applications. New methodTo address this limitation, we develop a fully-automated method for spike ripple detection. The method consists of a convolutional neural network trained to compute the probability that a spectrogram image contains a spike ripple. ResultsWe validate the proposed spike ripple detector on expert-labeled data and show that this detector accurately separates subjects with low and high seizure risks. Comparison with Existing MethodThe proposed method performs as well as existing methods that require manual validation of candidate spike ripple events. The introduction of a fully automated method reduces subjectivity and increases rigor and reproducibility of this epilepsy biomarker. ConclusionWe introduce and validate a fully-automated spike ripple detector to support utilization of this epilepsy biomarker in clinical and translational work.

Epileptic Seizure Recognition Using Reduced Deep Convolutional Stack Autoencoder and Improved Kernel RVFLN From EEG Signals.

In this paper, reduced deep convolutional stack autoencoder (RDCSAE) and improved kernel random vector functional link network (IKRVFLN) are combined to recognize the epileptic seizure using both the multichannel scalp and single-channel electroencephalogram (EEG) signals. The novel RDCSAE structure is designed to extract the most discriminative unsupervised features from EEG signals and fed into the proposed supervised IKRVFLN classifier to train efficiently by reducing the mean-square error cost function for recognizing the epileptic seizure activity with promising accuracy. The proposed RDCSAE-IKRVFLN algorithm is tested over the benchmark Boston Children's Hospital multichannel scalp EEG (sEEG) and Boon University, Germany single-channel EEG databases. The less computational complexity, higher learning speed, better model generalization, accurate epileptic seizure recognition, remarkable classification accuracy, negligible false positive rate per hour (FPR/h) and short event recognition time are the main advantages of the proposed RDCSAE-IKRVFLN method over reduced deep convolutional neural network (RDCNN), RDCSAE and RDCSAE-KRVFLN methods. The proposed RDCSAE-IKRVFLN method is implemented in a high-speed reconfigurable field-programmable gate array (FPGA) hardware environment to design a computer-aided-diagnosis (CAD) system for automatic epileptic seizure diagnosis. The simplicity, feasibility, and practicability of the proposed method validate the stable and reliable performances of epilepsy detection and recognition.

Factors affecting interictal unilateral and bilateral discharges and ictal diffusion patterns of scalp electroencephalogram in temporal lobe epilepsy.

The interictal discharges of temporal lobe epilepsy (TLE) can be unilateral or bilateral. In addition, the ictal electroencephalogram (EEG) showed the discharges also tend to spread to the contralateral brain in TLE. The factors influencing unilateral and bilateral interictal discharges in TLE as well as ictal diffusion patterns in scalp EEG during onset of seizure were evaluated in the present study. This was a retrospective analysis of 129 patients with TLE. Cases were classified into unilateral and bilateral discharge groups based on interictal discharge patterns in the EEG. Differences between the two groups in age, gender, disease duration, seizure frequency, magnetic resonance imaging (MRI) findings, origin of TLE, antiepileptic drug (AED) administration, and ictal diffusion patterns during seizures were statistically analyzed. In addition, the differences in ictal diffusion patterns between left and right TLE were statistically analyzed. Statistically significant differences were not observed in gender, disease duration, seizure frequency, MRI findings, administration of AEDs, and ictal diffusion patterns between interictal unilateral and bilateral discharge groups but with statistically significant differences in age and side of origin of the TLE. In addition, whether the EEG-recorded diffusion pattern was confined to the same hemisphere or spread to both hemispheres was investigated and shown statistically significant differences between the left and right temporal lobes. Age and side of origin of TLE affects the TLE interictal discharge patterns. Older patients are more prone to bilateral discharges. Bilateral discharges are more common in right TLE, and the onset of EEG more likely to bilateral diffusion in right TLE.

Deep learning for robust detection of interictal epileptiform discharges

Objective. Automatic detection of interictal epileptiform discharges (IEDs, short as ‘spikes’) from an epileptic brain can help predict seizure recurrence and support the diagnosis of epilepsy. Developing fast, reliable and robust detection methods for IEDs based on scalp or intracranial electroencephalogram (iEEG) may facilitate online seizure monitoring and closed-loop neurostimulation. Approach. We developed a new deep learning approach, which employs a long short-term memory network architecture (‘IEDnet’) and an auxiliary classifier generative adversarial network (AC-GAN), to train on both expert-annotated and augmented spike events from iEEG recordings of epilepsy patients. We validated our IEDnet with two real-world iEEG datasets, and compared IEDnet with the support vector machine (SVM) and random forest (RF) classifiers on their detection performances. Main results. IEDnet achieved excellent cross-validated detection performances in terms of both sensitivity and specificity, and outperformed SVM and RF. Synthetic spike samples augmented by AC-GAN further improved the detection performance. In addition, the performance of IEDnet was robust with respect to the sampling frequency and noise. Furthermore, we demonstrated the cross-institutional generalization ability of IEDnet while testing between two datasets. Significance. IEDnet achieves excellent detection performances in identifying interictal spikes. AC-GAN can produce augmented iEEG samples to improve supervised deep learning.

Multi-Center Validation Study of Automated Classification of Pathological Slowing in Adult Scalp Electroencephalograms Via Frequency Features.

Pathological slowing in the electroencephalogram (EEG) is widely investigated for the diagnosis of neurological disorders. Currently, the gold standard for slowing detection is the visual inspection of the EEG by experts, which is time-consuming and subjective. To address those issues, we propose three automated approaches to detect slowing in EEG: Threshold-based Detection System (TDS), Shallow Learning-based Detection System (SLDS), and Deep Learning-based Detection System (DLDS). These systems are evaluated on channel-, segment-, and EEG-level. The three systems perform prediction via detecting slowing at individual channels, and those detections are arranged in histograms for detection of slowing at the segment- and EEG-level. We evaluate the systems through Leave-One-Subject-Out (LOSO) cross-validation (CV) and Leave-One-Institution-Out (LOIO) CV on four datasets from the US, Singapore, and India. The DLDS achieved the best overall results: LOIO CV mean balanced accuracy (BAC) of 71.9%, 75.5%, and 82.0% at channel-, segment- and EEG-level, and LOSO CV mean BAC of 73.6%, 77.2%, and 81.8% at channel-, segment-, and EEG-level. The channel- and segment-level performance is comparable to the intra-rater agreement (IRA) of an expert of 72.4% and 82%. The DLDS can process a 30 min EEG in 4 s and can be deployed to assist clinicians in interpreting EEGs.

Seizure outcome after epilepsy surgery for patients with normal MRI: A Single center experience

ObjectiveTo analyze the surgical outcome in non-lesional intractable focal epilepsies in our center and to find possible predictors for better outcome. MethodsThis is a retrospective study for 40 adult patients with intractable focal epilepsy following at KFSHRC-Riyadh, who underwent presurgical evaluation followed by resective surgery and continued follow up for a minimum of 2 years. The surgery outcome was evaluated based on the type of surgical procedure and histopathology results. ResultsOut of all 40 patients studied, seizure freedom was achieved in 19 (47.5 %) and 17 (42.5 %) patients at the first and second year respectively in all non-lesional cases. Seizure freedom in non-lesional temporal lobe surgery was achieved in 10 (45 %) of patients at 2 years, 5 (38 %) in non-lesional frontal lobe patients at 2 years and 8 (44 %), 7 (38 %) for all extratemporal at 1 and 2 years respectively. Good prognosis was seen in patients with localized positron emission tomography (PET), had no aura and had a clear ictal onset either on scalp electroencephalogram (EEG) or subdural invasive electroencephalogram. SignificanceThe best surgical outcome is achievable in patients with non-lesional focal epilepsy. This study highlights the prognostic value of the PET scan and ictal scalp/subdural invasive EEG.

EEG Cortical Activities and Networks Altered by Watching 2D/3D Virtual Reality Videos

Abstract. Virtual reality (VR), which can represent real-life events and situations, is being increasingly applied to many fields, such as education, entertainment, and medical rehabilitation. Correspondingly, the neural information processing of VR has attracted attention. However, the underlying neural mechanisms of VR environments have not yet been fully revealed. The purpose of this study was to examine the possible differences in brain activities and networks between the less immersive 2D and the fully immersive 3D VR environments. 3D VR videos and the same 2D scenes were presented to the participants and the scalp electroencephalogram (EEG) was recorded, respectively. Power spectral density (PSD) and the functional connectivity of these EEG signals were analyzed. The results showed that 3D VR videos significantly enhanced the PSD of θ rhythm (4–7 Hz) in the frontal lobe; decreased the PSD of α rhythm (8–13 Hz) in the parietal and the occipital lobes; increased the PSD of β rhythm (14–30 Hz) in the frontal, the parietal, the temporal, and the occipital lobes, relative to 2D VR watching. Furthermore, 3D versus 2D VR-induced alterations in the patterns of brain networks were similar to the patterns of PSD. Specifically, for the θ rhythm, 3D VR significantly enhanced the frontal and the temporal brain functional connectivity; for the α rhythm, 3D VR increased the parietal and the occipital networks; for the β rhythm, 3D VR remarkably increased the frontal, the occipital, the frontal-temporal and the frontal-occipital brain functional connectivity, relative to 2D VR. These significant differences between 3D and 2D VR video-watching suggest that the neural information processing of cortical activities and networks is correlated to the degree of immersion. The present results, collected with previous researches, implicate that some visual-related information processes, such as visual attention, visual perception, and visual immersion are more robust in 3D VR environments.

Epileptic Classification With Deep-Transfer-Learning-Based Feature Fusion Algorithm

Epilepsy ictal detection based on scalp electroencephalograms (EEGs) has been comprehensively studied in the past decades. But few attentions have been paid to the preictal classification. In this article, a comprehensive study on epileptic state classification based on deep transfer learning (TL) is presented. The main contributions include: 1) the subband mean amplitude spectrum (MAS) map that characterizes the typical rhythms of brain activities is extracted for EEG representation; 2) five representative deep neural networks (DNNs) pretrained on ImageNet are applied for EEG feature TL; and 3) a 7-layer hierarchical neural network (HNN) that consists of three fully connected (Fc) and three dropout layers followed by a Softmax layer is developed to perform the epileptic state probability learning and classification. Experiments on the benchmark CHB-MIT and iNeuro EEG databases that contain several different types of seizures show that the proposed algorithm achieves the highest overall accuracies of 96.97% and 87.87% on the 5-state epileptic classification, respectively, that outperforms many existing state-of-the-art methods presented in this article.

Automatic recognition of preictal and interictal EEG signals using 1D-capsule networks

Epilepsy is the most common neurological disorder affecting people of all ages. Seizure prediction can be achieved by separating the preictal state in which the changes in the brain activities begin to occur from the interictal state. Therefore, in this study, a novel method for preictal/interictal recognition, the most important step in seizure prediction from scalp electroencephalogram signals, is proposed. In the proposed method, one-dimensional capsule networks, a novel neural network model, is used. The best classification accuracy for preictal/interictal recognition was achieved with 97.74% in F3-C3 channel pairs. Compared to other methods, our 1D-CapsNet model achieved the best performance. Moreover, the results indicated that the interval that ended 30 min before the onset of seizures contained important information about preictal/interictal recognition. We believe that the proposed method will bring a new perspective to the seizure prediction studies of capsule networks.

Potential Efficacy of Multimodal Mixed Reality in Epilepsy Surgery.

Mixed reality (MR) technology, which can fuse things in real and virtual space in real time, has been used mainly for simulation in neurosurgical training. To develop MR technology into multimodal MR for intraoperative guidance during epilepsy surgery. A 33-yr-old male patient suffered from intractable general tonic seizures due to left temporal meningoencephalocele. Preoperative scalp electroencephalograms localized interictal epileptic activity on the left temporal lobe. Iomazenil single photon emission tomography revealed temporal lobe lateralization. Magnetic resonance imaging (MRI) demonstrated left basal temporal meningoencephalocele extending into the pterygopalatine fossa through a bone defect at the base of the greater sphenoid wing. A 3-dimensional model was created for MR based on multimodal data including computed tomography, MRI tractography, and digital subtraction angiography, which enabled 3-dimensional visualization of abnormal subcortical fiber connections between the meningoencephalocele and the epileptic focus. By using intraoperative multimodal MR, we were able to safely remove the meningoencephalocele and perform epileptic focus resection. The patient was seizure-free postoperatively, and no adverse effects were noted. Intraoperative multimodal MR was a feasible and effective technique, and it can be applied for a wide range of epilepsy surgeries.

Use of common average reference and large-Laplacian spatial-filters enhances EEG signal-to-noise ratios in intrinsic sensorimotor activity

BackgroundOscillations in the resting-state scalp electroencephalogram (EEG) represent various intrinsic brain activities. One of the characteristic EEG oscillations is the sensorimotor rhythm (SMR)—with its arch-shaped waveform in alpha- and betabands—that reflect sensorimotor activity. The representation of sensorimotor activity by the SMR depends on the signal-to-noise ratio achieved by EEG spatial filters. New methodWe employed simultaneous recording of EEG and functional magnetic resonance imaging, and 10-min resting-state brain activities were recorded in 19 healthy volunteers. To compare the EEG spatial-filtering methods commonly used for extracting sensorimotor cortical activities, we assessed nine different spatial-filters: a default reference of EEG amplifier system, a common average reference (CAR), small-, middle- and large-Laplacian filters, and four types of bipolar manners (C3-Cz, C3-F3, C3-P3, and C3-T7). We identified the brain region that correlated with the EEG-SMR power obtained after each spatial-filtering method was applied. Subsequently, we calculated the proportion of the significant voxels in the sensorimotor cortex as well as the sensorimotor occupancy in all significant regions to examine the sensitivity and specificity of each spatial-filter. ResultsThe CAR and large-Laplacian spatial-filters were superior at improving the signal-to-noise ratios for extracting sensorimotor activity from the EEG-SMR signal. Comparison with existing methodsOur results are consistent with the spatial-filter selection to extract task-dependent activation for better control of EEG-SMR-based interventions. Our approach has the potential to identify the optimal spatial-filter for EEG-SMR. ConclusionsEvaluating spatial-filters for extracting spontaneous sensorimotor activity from the EEG is a useful procedure for constructing more effective EEG-SMR-based interventions.

Non-Gaussianity Detection of EEG Signals Based on a Multivariate Scale Mixture Model for Diagnosis of Epileptic Seizures.

The detection of epileptic seizures from scalp electroencephalogram (EEG) signals can facilitate early diagnosis and treatment. Previous studies suggested that the Gaussianity of EEG distributions changes depending on the presence or absence of seizures; however, no general EEG signal models can explain such changes in distributions within a unified scheme. This article describes the formulation of a stochastic EEG model based on a multivariate scale mixture distribution that can represent changes in non-Gaussianity caused by stochastic fluctuations in EEG. In addition, we propose an EEG analysis method by combining the model with a filter bank and introduce a feature representing the non-Gaussianity latent in each EEG frequency band. We applied the proposed method to multichannel EEG data from twenty patients with focal epilepsy. The results showed a significant increase in the proposed feature during epileptic seizures, particularly in the high-frequency band. The feature calculated in the high-frequency band allowed highly accurate classification of seizure and non-seizure segments [area under the receiver operating characteristic curve (AUC) = 0.881] using only a simple threshold. This article proposed a multivariate scale mixture distribution-based stochastic EEG model capable of representing non-Gaussianity associated with epileptic seizures. Experiments using simulated and real EEG data demonstrated the validity of the model and its applicability to epileptic seizure detection. The stochastic fluctuations of EEG quantified by the proposed model can help detect epileptic seizures with high accuracy.

Automated Adult Epilepsy Diagnostic Tool Based on Interictal Scalp Electroencephalogram Characteristics: A Six-Center Study.

The diagnosis of epilepsy often relies on a reading of routine scalp electroencephalograms (EEGs). Since seizures are highly unlikely to be detected in a routine scalp EEG, the primary diagnosis depends heavily on the visual evaluation of Interictal Epileptiform Discharges (IEDs). This process is tedious, expert-centered, and delays the treatment plan. Consequently, the development of an automated, fast, and reliable epileptic EEG diagnostic system is essential. In this study, we propose a system to classify EEG as epileptic or normal based on multiple modalities extracted from the interictal EEG. The ensemble system consists of three components: a Convolutional Neural Network (CNN)-based IED detector, a Template Matching (TM)-based IED detector, and a spectral feature-based classifier. We evaluate the system on datasets from six centers from the USA, Singapore, and India. The system yields a mean Leave-One-Institution-Out (LOIO) cross-validation (CV) area under curve (AUC) of 0.826 (balanced accuracy (BAC) of 76.1%) and Leave-One-Subject-Out (LOSO) CV AUC of 0.812 (BAC of 74.8%). The LOIO results are found to be similar to the interrater agreement (IRA) reported in the literature for epileptic EEG classification. Moreover, as the proposed system can process routine EEGs in a few seconds, it may aid the clinicians in diagnosing epilepsy efficiently.

ScoreNet: A Neural Network-Based Post-Processing Model for Identifying Epileptic Seizure Onset and Offset in EEGs.

We design an algorithm to automatically detect epileptic seizure onsets and offsets from scalp electroencephalograms (EEGs). The proposed scheme consists of two sequential steps: detecting seizure episodes from long EEG recordings, and determining seizure onsets and offsets of the detected episodes. We introduce a neural network-based model called ScoreNet to carry out the second step by better predicting the seizure probability of pre-detected seizure epochs to determine seizure onsets and offsets. A cost function called log-dice loss with a similar meaning to the F1 score is proposed to handle the natural data imbalance inherent in EEG signals signifying seizure events. ScoreNet is then verified on the CHB-MIT Scalp EEG database in combination with several classifiers including random forest, convolutional neural network (CNN), and logistic regression. As a result, ScoreNet improves seizure detection performance over lone epoch-based seizure classification methods; F1 scores increase significantly from 16-37% to 53-70%, and false positive rates per hour decrease from 0.53-5.24 to 0.05-0.61. This method provides clinically acceptable latencies of detecting seizure onset and offset of less than 10 seconds. In addition, an effective latency index is proposed as a metric for detection latency whose scoring considers undetected events to provide better insight into onset and offset detection than conventional time-based metrics.

Automatic Electroencephalogram Artifact Removal Using Deep Convolutional Neural Networks

Scalp electroencephalogram (EEG) is a non-invasive measure of brain activity. It is widely used in several applications including cognitive tasks, sleep stage detection, and seizure prediction. When recorded over several hours, this signal is usually corrupted by noisy disturbances such as experimental errors, environmental interferences, and physiological artifacts. These may generate confounding factors and, therefore, lead to false results. Models able to minimise EEG artifacts are then necessary for improving further analysis and application. In this work, we developed an EEG artifact removal model based on deep convolutional neural networks. The proposed approach was applied on long-term EEGs, acquired from epileptic patients, available in the EPILEPSIAE database. The main goal of our work is to develop a model able to automatically and quickly remove artifacts from EEGs. To develop it, we used EEG segments, manually preprocessed by experts and named target EEG segments. Our approach was evaluated comparing denoised segments with the target segments. Furthermore, we compared our approach with other artifact removal models. Results show that the developed model was able to attenuate the influence of artifacts, present in long-term EEG signals, in a similar way to that performed by experts. Additionally, results evidence that our approach performs better than other artifact removal models, combining a minor reconstruction error with a fast processing. Being a fully automatic and fast model that does not require reference artifact templates, turns it suitable, for example, for continuous preprocessing of long-term electroencephalogram for sleep staging or seizure prediction.

Seizure Onset Detection Using Empirical Mode Decomposition and Common Spatial Pattern.

Automatic seizure onset detection plays an important role in epilepsy diagnosis. In this paper, a novel seizure onset detection method is proposed by combining empirical mode decomposition (EMD) of long-term scalp electroencephalogram (EEG) with common spatial pattern (CSP). First, wavelet transform (WT) and EMD are employed on EEG recordings respectively for filtering pre-processing and time-frequency decomposition. Then CSP is applied to reduce the dimension of multi-channel time-frequency representation, and the variance is extracted as the only feature. Afterwards, a support vector machine (SVM) group consisting of ten SVMs is served as a robust classifier. Finally, the post-processing is adopted to acquire a higher recognition rate and reduce the false detection rate. The results obtained from CHB-MIT database of 977 h scalp EEG recordings reveal that the proposed system can achieve a segment-based sensitivity of 97.34% with a specificity of 97.50% and an event-based sensitivity of 98.47% with a false detection rate of 0.63/h. This proposed detection system was also validated on a clinical scalp EEG database from the Second Hospital of Shandong University, and the system yielded a sensitivity of 93.67% and a specificity of 96.06%. At the event-based level, a sensitivity of 99.39% and a false detection rate of 0.64/h were obtained. Furthermore, this work showed that the CSP spatial filter was helpful to identify EEG channels involved in seizure onsets. These satisfactory results indicate that the proposed system may provide a reference for seizure onset detection in clinical applications.

Dimension Reduction of Polynomial Regression Models for the Estimation of Granger Causality in High-Dimensional Time Series

The causality analysis of multivariate time series and formation of complex networks relies on the estimation of the direct cause-effect from one observed variable to another accounting for the presence of other observed variables. Such effects are quantified by the conditional Granger causality index (CGCI), derived by linear vector autoregressive models (VAR). Dimension reduction approaches have been developed to restrict VAR models, such as the modified backward-in-time selection method (mBTS) giving the restricted CGCI (rCGCI). The rCGCI is limited to linear systems. In this study, we extend the mBTS dimension reduction scheme to polynomial VAR, so as to model nonlinear dynamics and cause-effect relationships and derive the Granger causality measure termed restricted polynomial CGCI (rpCGCI). The complications in the adaptation of mBTS to the restricted polynomial VAR and the construction of the Granger causality index are addressed, involving also a randomization significance test in the steps of mBTS. The rpCGCI has the advantage against other nonlinear Granger causality measures that it can be applied to short time series of high dimension (many observed variables). The simulation study on different types of multivariate stochastic processes and different lengths of generated time series showed the superiority of the proposed rpCGCI as compared to CGCI, rCGCI, pCGCI (using polynomial VAR) and other nonlinear causality measures. Further, rpCGCI was compared favorably to the other measures on signals of heart rate variability, respiration, and oxygen concentration in the blood, as well as multi-channel scalp electroencephalogram (EEG) recordings of epileptic patients containing epileptiform discharges.

Identifying Refractory Epilepsy Without Structural Abnormalities by Fusing the Common Spatial Patterns of Functional and Effective EEG Networks

Drug refractory epilepsy (RE) is believed to be associated with structural lesions, but some RE patients show no significant structural abnormalities (RE-no-SA) on conventional magnetic resonance imaging scans. Since most of the medically controlled epilepsy (MCE) patients also do not exhibit structural abnormalities, a reliable assessment needs to be developed to differentiate RE-no-SA patients and MCE patients to avoid misdiagnosis and inappropriate treatment. Using resting-state scalp electroencephalogram (EEG) datasets, we extracted the spatial pattern of network (SPN) features from the functional and effective EEG networks of both RE-no-SA patients and MCE patients. Compared to the performance of traditional resting-state EEG network properties, the SPN features exhibited remarkable superiority in classifying these two groups of epilepsy patients, and accuracy values of 90.00% and 80.00% were obtained for the SPN features of the functional and effective EEG networks, respectively. By further fusing the SPN features of functional and effective networks, we demonstrated that the highest accuracy value of 96.67% could be reached, with a sensitivity of 100% and specificity of 92.86%. Overall, these findings not only indicate that the fused functional and effective SPN features are promising as reliable measurements for distinguishing RE-no-SA patients and MCE patients but also may provide a new perspective to explore the complex neurophysiology of refractory epilepsy.

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

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