Resting-state EEG for the diagnosis of idiopathic epilepsy and psychogenic nonepileptic seizures: A systematic review

Estimating the likelihood of epilepsy from clinically noncontributory electroencephalograms using computational analysis: A retrospective, multisite case-control study

OBJECTIVE

This study was undertaken to validate a set of candidate biomarkers of seizure susceptibility in a retrospective, multisite case-control study, and to determine the robustness of these biomarkers derived from routinely collected electroencephalography (EEG) within a large cohort (both epilepsy and common alternative conditions such as nonepileptic attack disorder).

METHODS

The database consisted of 814 EEG recordings from 648 subjects, collected from eight National Health Service sites across the UK. Clinically noncontributory EEG recordings were identified by an experienced clinical scientist (N = 281; 152 alternative conditions, 129 epilepsy). Eight computational markers (spectral [n = 2], network-based [n = 4], and model-based [n = 2]) were calculated within each recording. Ensemble-based classifiers were developed using a two-tier cross-validation approach. We used standard regression methods to assess whether potential confounding variables (e.g., age, gender, treatment status, comorbidity) impacted model performance.

RESULTS

We found levels of balanced accuracy of 68% across the cohort with clinically noncontributory normal EEGs (sensitivity =61%, specificity =75%, positive predictive value =55%, negative predictive value =79%, diagnostic odds ratio =4.64, area under receiver operated characteristics curve =.72). Group level analysis found no evidence suggesting any of the potential confounding variables significantly impacted the overall performance.

SIGNIFICANCE

These results provide evidence that the set of biomarkers could provide additional value to clinical decision-making, providing the foundation for a decision support tool that could reduce diagnostic delay and misdiagnosis rates. Future work should therefore assess the change in diagnostic yield and time to diagnosis when utilizing these biomarkers in carefully designed prospective studies.

Resting-state brain activity distinguishes patients with generalised epilepsy from others

PURPOSE

Epilepsy is a prevalent neurological disorder characterised by repetitive seizures. It is categorised into three types: generalised epilepsy (GE), focal epilepsy (FE), and combined generalised and focal epilepsy. Correctly subtyping the epilepsy is important to select appropriate treatments. The types are mainly determined (i.e., diagnosed) by their semiologies supported by clinical examinations, such as electroencephalography and magnetoencephalography (MEG). Although these examinations are traditionally based on visual inspections of interictal epileptic discharges (IEDs), which are not always visible, alternative analyses have been anticipated. We examined if resting-state brain activities can distinguish patients with GE, which would help us to diagnose the type of epilepsy.

METHODS

The 5 min resting-state brain activities acquired using MEG were obtained retrospectively from 15 patients with GE. The cortical source of the activities was estimated at each frequency band from delta to high-frequency oscillation (HFO). These estimated activities were compared with reference datasets from 133 healthy individuals and control data from 29 patients with FE.

RESULTS

Patients with GE showed larger theta in the occipital, alpha in the left temporal, HFO in the rostral deep regions, and smaller HFO in the caudal ventral regions. Their area under the curves of the receiver operating characteristic curves was around 0.8-0.9. The distinctive pattern was not found for data from FE.

CONCLUSION

Patients with GE show distinctive resting-state brain activity, which could be a potential biomarker and used complementarily to classical analysis based on the visual inspection of IEDs.

Limited clinical validity of univariate resting-state EEG markers for classifying seizure disorders

Differentiating between epilepsy and psychogenic non-epileptic seizures presents a considerable challenge in clinical practice, resulting in frequent misdiagnosis, unnecessary treatment and long diagnostic delays. Quantitative markers extracted from resting-state EEG may reveal subtle neurophysiological differences that are diagnostically relevant. Two observational, retrospective diagnostic accuracy studies were performed to test the clinical validity of univariate resting-state EEG markers for the differential diagnosis of epilepsy and psychogenic non-epileptic seizures. Clinical EEG data were collected for 179 quasi-consecutive patients (age > 18) with a suspected diagnosis of epilepsy or psychogenic non-epileptic seizures who were medication-naïve at the time of EEG; 148 age- and gender-matched patients subsequently received a diagnosis from specialist clinicians and were included in the analyses. Study 1 is a hypothesis-driven study testing the ability of theta power and peak alpha frequency to classify people with epilepsy and people with psychogenic non-epileptic seizures, with an advanced machine learning pipeline. The next study (Study 2) is data-driven; a high number of quantitative EEG features are extracted and a similar machine learning approach as Study 1 assesses whether previously unexplored univariate EEG measures show promise as diagnostic markers. The results of Study 1 suggest that EEG markers that were previously identified as promising diagnostic indicators (i.e. theta power and peak alpha frequency) have limited clinical validity for the classification of epilepsy and psychogenic non-epileptic seizures (mean accuracy: 48%). The results of Study 2 indicate that identifying univariate markers that show good correlation with a categorical diagnostic label is challenging (mean accuracy: 45-60%). This is due to a considerable overlap in neurophysiological features between the diagnostic classes considered in this study, and to the presence of more dominant EEG dynamics such as alterations due to temporal proximity to epileptiform discharges. Markers that were identified in the context of previous epilepsy research using visually normal resting-state EEG were found to have limited clinical validity for the classification task of distinguishing between people with epilepsy and people with psychogenic non-epileptic seizures. A search for alternative diagnostic markers uncovered the challenges involved and generated recommendations for further research.

Nonlinear and chaos features over EMD/VMD decomposition methods for ictal EEG signals detection

The detection and identification of epileptic seizures attracted considerable relevance for the neurophysiologists. In order to accomplish the detection of epileptic seizures or equivalently ictal EEG states, this paper proposes the use of nonlinear and chaos features not computed over the raw EEG signals as it was commonly experienced, but instead over intrinsic mode functions (IMFs) extracted subsequently to the application of newly time-frequency signal decomposition methods on the basis of empirical mode decomposition (EMD) and variational mode decomposition (VMD) methods. The first step within the proposed methodology is to excerpt the various components of the IMFs by EMD and VMD decomposition methods on time EEG segments. The Hjorth parameters, the Hurst exponent, the Recurrence Quantification Analysis (RQA), the detrended fluctuation analysis (DFA), the Largest Lyapunov Exponent (LLE), The Higuchi and Katz fractal dimensions (HFD and KFD), seven nonlinear and chaos features computed over the IMFs were investigated and their classification performances evaluated using the k-nearest neighbor (KNN) and the multilayer perceptron neural network (MLPNN) classifiers. Furthermore, the combination of the best nonlinear features has also been examined in terms of sensitivity, specificity and overall classification accuracy. The publicly available Bonn EEG dataset has been has been employed to validate the efficiency of the proposed method for detecting ictal EEG signals from normal or interictal EEG segments. Among the several experiments involved in the current study, the ultimate results establish that the overall classification accuracy can achieve 100%, 99.45%, 99.8%, 99.8%, 98.6% and 99.1% for six different epileptic seizure detection case problems studied, confirming the ability of the proposed methodology in helping the clinic practitioners in the epilepsy detection care units to classify seizure events with a great confidence.

Entropy-Based Machine Learning Model for Fast Diagnosis and Monitoring of Parkinson's Disease

This study presents the concept of a computationally efficient machine learning (ML) model for diagnosing and monitoring Parkinson's disease (PD) using rest-state EEG signals (rs-EEG) from 20 PD subjects and 20 normal control (NC) subjects at a sampling rate of 128 Hz. Based on the comparative analysis of the effectiveness of entropy calculation methods, fuzzy entropy showed the best results in diagnosing and monitoring PD using rs-EEG, with classification accuracy (ARKF) of ~99.9%. The most important frequency range of rs-EEG for PD-based diagnostics lies in the range of 0-4 Hz, and the most informative signals were mainly received from the right hemisphere of the head. It was also found that ARKF significantly decreased as the length of rs-EEG segments decreased from 1000 to 150 samples. Using a procedure for selecting the most informative features, it was possible to reduce the computational costs of classification by 11 times, while maintaining an ARKF ~99.9%. The proposed method can be used in the healthcare internet of things (H-IoT), where low-performance edge devices can implement ML sensors to enhance human resilience to PD.

Comparison of Resting-State EEG Network Analyses With and Without Parallel MRI in Genetic Generalized Epilepsy

Genetic generalized epilepsy (GGE) is conceptualized as a brain disorder involving distributed bilateral networks. To study these networks, simultaneous EEG-fMRI measurements can be used. However, inside-MRI EEG suffers from strong MR-related artifacts; it is not established whether EEG-based metrics in EEG-fMRI resting-state measurements are suitable for the analysis of group differences at source-level. We evaluated the impact of the inside-MR measurement condition on statistical group comparisons of EEG on source-level power and functional connectivity in patients with GGE versus healthy controls. We studied the cross-modal spatial relation of statistical group differences in seed-based FC derived from EEG and parallel fMRI. We found a significant increase in power and a frequency-specific change in functional connectivity for the inside MR-scanner compared to the outside MR-scanner condition. For power, we found reduced group difference between GGE and controls both in terms of statistical significance as well as effect size. Group differences for ImCoh remained similar both in terms of statistical significance as well as effect size. We found increased seed-based FC for GGE patients from the thalamus to the precuneus cortex region in fMRI, and in the theta band of simultaneous EEG. Our findings suggest that the analysis of EEG functional connectivity based on ImCoh is suitable for MR-EEG, and that relative group difference in a comparison of patients with GGE against controls are preserved. Spatial correspondence of seed-based FC group differences between the two modalities was found for the thalamus.

Machine-learning for the prediction of one-year seizure recurrence based on routine electroencephalography

Predicting seizure recurrence risk is critical to the diagnosis and management of epilepsy. Routine electroencephalography (EEG) is a cornerstone of the estimation of seizure recurrence risk. However, EEG interpretation relies on the visual identification of interictal epileptiform discharges (IEDs) by neurologists, with limited sensitivity. Automated processing of EEG could increase its diagnostic yield and accessibility. The main objective was to develop a prediction model based on automated EEG processing to predict one-year seizure recurrence in patients undergoing routine EEG. We retrospectively selected a consecutive cohort of 517 patients undergoing routine EEG at our institution (training set) and a separate, temporally shifted cohort of 261 patients (testing set). We developed an automated processing pipeline to extract linear and non-linear features from the EEGs. We trained machine learning algorithms on multichannel EEG segments to predict one-year seizure recurrence. We evaluated the impact of IEDs and clinical confounders on performances and validated the performances on the testing set. The receiver operating characteristic area-under-the-curve for seizure recurrence after EEG in the testing set was 0.63 (95% CI 0.55-0.71). Predictions were still significantly above chance in EEGs with no IEDs. Our findings suggest that there are changes other than IEDs in the EEG signal embodying seizure propensity.

Computer-assisted analysis of routine electroencephalogram to identify hidden biomarkers of epilepsy: protocol for a systematic review

INTRODUCTION

The diagnosis of epilepsy frequently relies on the visual interpretation of the electroencephalogram (EEG) by a neurologist. The hallmark of epilepsy on EEG is the interictal epileptiform discharge (IED). This marker lacks sensitivity: it is only captured in a small percentage of 30 min routine EEGs in patients with epilepsy. In the past three decades, there has been growing interest in the use of computational methods to analyse the EEG without relying on the detection of IEDs, but none have made it to the clinical practice. We aim to review the diagnostic accuracy of quantitative methods applied to ambulatory EEG analysis to guide the diagnosis and management of epilepsy.

METHODS AND ANALYSIS

The protocol complies with the recommendations for systematic reviews of diagnostic test accuracy by Cochrane. We will search MEDLINE, EMBASE, EBM reviews, IEEE Explore along with grey literature for articles, conference papers and conference abstracts published after 1961. We will include observational studies that present a computational method to analyse the EEG for the diagnosis of epilepsy in adults or children without relying on the identification of IEDs or seizures. The reference standard is the diagnosis of epilepsy by a physician. We will report the estimated pooled sensitivity and specificity, and receiver operating characteristic area under the curve (ROC AUC) for each marker. If possible, we will perform a meta-analysis of the sensitivity and specificity and ROC AUC for each individual marker. We will assess the risk of bias using an adapted QUADAS-2 tool. We will also describe the algorithms used for signal processing, feature extraction and predictive modelling, and comment on the reproducibility of the different studies.

ETHICS AND DISSEMINATION

Ethical approval was not required. Findings will be disseminated through peer-reviewed publication and presented at conferences related to this field.

PROSPERO REGISTRATION NUMBER

CRD42022292261.

A Classification Model of EEG Signals Based on RNN-LSTM for Diagnosing Focal and Generalized Epilepsy

Epilepsy is a chronic neurological disorder caused by abnormal neuronal activity that is diagnosed visually by analyzing electroencephalography (EEG) signals.

BACKGROUND

Surgical operations are the only option for epilepsy treatment when patients are refractory to treatment, which highlights the role of classifying focal and generalized epilepsy syndrome. Therefore, developing a model to be used for diagnosing focal and generalized epilepsy automatically is important.

METHODS

A classification model based on longitudinal bipolar montage (LB), discrete wavelet transform (DWT), feature extraction techniques, and statistical analysis in feature selection for RNN combined with long short-term memory (LSTM) is proposed in this work for identifying epilepsy. Initially, normal and epileptic LB channels were decomposed into three levels, and 15 various features were extracted. The selected features were extracted from each segment of the signals and fed into LSTM for the classification approach.

RESULTS

The proposed algorithm achieved a 96.1% accuracy, a 96.8% sensitivity, and a 97.4% specificity in distinguishing normal subjects from subjects with epilepsy. This optimal model was used to analyze the channels of subjects with focal and generalized epilepsy for diagnosing purposes, relying on statistical parameters.

CONCLUSIONS

The proposed approach is promising, as it can be used to detect epilepsy with satisfactory classification performance and diagnose focal and generalized epilepsy.

Entropy Measures of Electroencephalograms towards the Diagnosis of Psychogenic Non-Epileptic Seizures

Psychogenic non-epileptic seizures (PNES) may resemble epileptic seizures but are not caused by epileptic activity. However, the analysis of electroencephalogram (EEG) signals with entropy algorithms could help identify patterns that differentiate PNES and epilepsy. Furthermore, the use of machine learning could reduce the current diagnosis costs by automating classification. The current study extracted the approximate sample, spectral, singular value decomposition, and Renyi entropies from interictal EEGs and electrocardiograms (ECG)s of 48 PNES and 29 epilepsy subjects in the broad, delta, theta, alpha, beta, and gamma frequency bands. Each feature-band pair was classified by a support vector machine (SVM), k-nearest neighbour (kNN), random forest (RF), and gradient boosting machine (GBM). In most cases, the broad band returned higher accuracy, gamma returned the lowest, and combining the six bands together improved classifier performance. The Renyi entropy was the best feature and returned high accuracy in every band. The highest balanced accuracy, 95.03%, was obtained by the kNN with Renyi entropy and combining all bands except broad. This analysis showed that entropy measures can differentiate between interictal PNES and epilepsy with high accuracy, and improved performances indicate that combining bands is an effective improvement for diagnosing PNES from EEGs and ECGs.

Combined electrophysiological and morphological phenotypes in patients with genetic generalized epilepsy and their healthy siblings

OBJECTIVE

Genetic generalized epilepsy is characterized by aberrant neuronal dynamics and subtle structural alterations. We evaluated whether a combination of magnetic and electrical neuronal signals and cortical thickness would provide complementary information about network pathology in GGE. We also investigated if these imaging phenotypes were present in healthy siblings of the patients to test for genetic influence.

METHODS

In this cross-sectional study, we analyzed five minutes of resting-state data acquired using electroencephalography (EEG) and magnetoencephalography (MEG) in patients, their siblings, and controls, matched for age and sex. We computed source-reconstructed power and connectivity in six frequency bands (1-40 Hz) and cortical thickness (derived from magnetic resonance imaging (MRI)). Group differences were assessed using permutation analysis of linear models for each modality separately and jointly for all modalities using a non-parametric combination.

RESULTS

Patients with GGE (n = 23) had higher power than controls (n = 35) in all frequencies, with a more posterior focus in MEG than EEG. Connectivity was also increased, particularly in frontotemporal and central regions in theta (strongest in EEG) and low beta frequencies (strongest in MEG), which was eminent in the joint EEG/MEG analysis. EEG showed weaker connectivity differences in higher frequencies, possibly related to drug effects. The inclusion of cortical thickness reinforced group differences in connectivity and power. Siblings (n = 18) had functional and structural patterns intermediate between those of patients and controls.

SIGNIFICANCE

EEG detected increased connectivity and power in GGE similar to MEG, but with different spectral sensitivity, highlighting the importance of theta and beta oscillations. Cortical thickness reductions in GGE corresponded to functional imaging patterns. Our multimodal approach extends the understanding of the resting-state in GGE and points to genetic underpinnings of the imaging markers studied, providing new insights into the causes and consequences of epilepsy.

Prediction Value of Epilepsy Secondary to Inferior Cavity Hemorrhage Based on Scalp EEG Wave Pattern in Deep Learning

Objective

To search the predictive value of epilepsy secondary to acute subarachnoid hemorrhage (aSAH) based on EEG wave pattern in deep learning.

Methods

A total of 156 cases of secondary epilepsy with lower cavity hemorrhage in our hospital were selected and divided into the late epilepsy group and the early epilepsy group according to seizure time, and the nonseizure group and the seizure group according to seizure condition. General data of patients were collected, the EEG types of each group were analyzed, and the disease recurrence rate, treatment effect, and symptom onset time were compared.

Results

Rapid and slow and rapid blood flow velocity were the main abnormal manifestations of epilepsy secondary to inferior cavity hemorrhage, accounting for 33.3% and 18.6%, respectively. Compared with the seizure group, the proportion of type ii and type iii in the nonseizure group was higher, and the proportion of type ii and type iii in the early epilepsy group was higher than in the late epilepsy group (P < 0.05). The diagnostic accuracy, missed diagnosis rate, misdiagnosis rate, specificity, and sensitivity of the EEG wave pattern were 94.9%, 3.2%, 1.9%, 91.7%, and 96.2%, respectively. Compared with the early epilepsy group, the recurrence rate of type iii and type iv in the late epilepsy group was higher (P < 0.05). The effective rates of the attack group and the nonattack group were 72.7% and 97.0%, respectively. Compared with the attack group, the effective rate of the nonattack group was higher (P < 0.05). The effective rates of the early epilepsy group and the late epilepsy group were 91.7% and 85.0%, respectively. Compared with the late epilepsy group, the effective rate of the early epilepsy group was higher (P < 0.05). Compared with the early epilepsy group, the late epilepsy group had longer tonic-clonic seizures, atonic seizures, and absent seizures, and the difference between the groups was statistically significant (P < 0.05).

Conclusion

In aSAH secondary epilepsy disease prediction, based on indepth study of the scalp EEG wave type prediction, they play an important role, including aSAH high-risk secondary epilepsy wave types for V, III, and IV types, as well as early and late epilepsy associated with disease stage. Through the diagnosis method to predict the severity of disease, this builds a good foundation for clinical treatment. It is beneficial to improve the effective rate of treatment.

Weighted Brain Network Analysis on Different Stages of Clinical Cognitive Decline

This study addresses brain network analysis over different clinical severity stages of cognitive dysfunction using electroencephalography (EEG). We exploit EEG data of subjective cognitive impairment (SCI) patients, mild cognitive impairment (MCI) patients and Alzheimer’s disease (AD) patients. We propose a new framework to study the topological networks with a spatiotemporal entropy measure for estimating the connectivity. Our results show that functional connectivity and graph analysis are frequency-band dependent, and alterations start at the MCI stage. In delta, the SCI group exhibited a decrease of clustering coefficient and an increase of path length compared to MCI and AD. In alpha, the opposite behavior appeared, suggesting a rapid and high efficiency in information transmission across the SCI network. Modularity analysis showed that electrodes of the same brain region were distributed over several modules, and some obtained modules in SCI were extended from anterior to posterior regions. These results demonstrate that the SCI network was more resilient to neuronal damage compared to that of MCI and even more compared to that of AD. Finally, we confirm that MCI is a transitional stage between SCI and AD, with a predominance of high-strength intrinsic connectivity, which may reflect the compensatory response to the neuronal damage occurring early in the disease process.

Suppressing of Power Line Artifact From Electroencephalogram Measurements Using Sparsity in Frequency Domain

Electroencephalogram (EEG) plays an important role in brain disease diagnosis and research of brain-computer interface (BCI). However, the measurements of EEG are often exposed to strong interference of power line artifact (PLA). Digital notch filters (DNFs) can be applied to remove the PLA effectively, but it also results in severe signal distortions in the time domain. To address this problem, spectrum correction (SC) based methods can be utilized. These methods estimate harmonic parameters of the PLA such that compensation signals are produced to remove the noise. In order to ensure high accuracy during harmonic parameter estimations, a novel approach is proposed in this paper. This novel approach is based on the combination of sparse representation (SR) and SC. It can deeply mine the information of PLA in the frequency domain. Firstly, a ratio-based spectrum correction (RBSC) using rectangular window is employed to make rough estimation of the harmonic parameters of PLA. Secondly, the two spectral line closest to the estimated frequency are calculated. Thirdly, the two spectral lines with high amplitudes can be utilized as input of RBSC to make finer estimations of the harmonic parameters. Finally, a compensation signal, based on the extracted harmonic parameters, is generated to suppress PLA. Numerical simulations and actual EEG signals with PLA were used to evaluate the effectiveness of the improved approach. It is verified that this approach can effectively suppress the PLA without distorting the time-domain waveform of the EEG signal.

Signal complexity indicators of health status in clinical EEG

Brain signal variability changes across the lifespan in both health and disease, likely reflecting changes in information processing capacity related to development, aging and neurological disorders. While signal complexity, and multiscale entropy (MSE) in particular, has been proposed as a biomarker for neurological disorders, most observations of altered signal complexity have come from studies comparing patients with few to no comorbidities against healthy controls. In this study, we examined whether MSE of brain signals was distinguishable across patient groups in a large and heterogeneous set of clinical-EEG data. Using a multivariate analysis, we found unique timescale-dependent differences in MSE across various neurological disorders. We also found MSE to differentiate individuals with non-brain comorbidities, suggesting that MSE is sensitive to brain signal changes brought about by metabolic and other non-brain disorders. Such changes were not detectable in the spectral power density of brain signals. Our findings suggest that brain signal complexity may offer complementary information to spectral power about an individual's health status and is a promising avenue for clinical biomarker development.