Seizure onset predicts its type

Differentiation of subclinical and clinical electrographic events in long-term electroencephalographic recordings

OBJECTIVE

With the advent of ultra-long-term recordings for monitoring of epilepsies, the interpretation of results of isolated electroencephalographic (EEG) recordings covering only selected brain regions attracts considerable interest. In this context, the question arises of whether detected ictal EEG patterns correspond to clinically manifest seizures or rather to purely electrographic events, that is, subclinical events.

METHODS

EEG patterns from 268 clinical seizures and 252 subclinical electrographic events from 50 patients undergoing video-EEG monitoring were analyzed. Features extracted included predominant frequency band, duration, association with rhythmic muscle artifacts, spatial extent, and propagation patterns. Classification using logistic regression was performed based on data from the whole dataset of 10-20 system EEG recordings and from a subset of two temporal electrode contacts.

RESULTS

Correct separation of clinically manifest and purely electrographic events based on 10-20 system EEG recordings was possible in up to 83.8% of events, depending on the combination of features included. Correct classification based on two-channel recordings was only slightly inferior, achieving 78.6% accuracy; 74.4% and 74.8%, respectively, of events could be correctly classified when using duration alone with either electrode set, although classification accuracies were lower for some subgroups of seizures, particularly focal aware seizures and epileptic arousals.

SIGNIFICANCE

A correct classification of subclinical versus clinical EEG events was possible in 74%-83% of events based on full EEG recordings, and in 74%-78% when considering only a subset of two electrodes, matching the channel number available from new implantable diagnostic devices. This is a promising outcome, suggesting that ultra-long-term low-channel EEG recordings may provide sufficient information for objective seizure diaries. Intraindividual optimization using high numbers of ictal events may further improve separation, provided that supervised learning with external validation is feasible.

Pre-ictal fluctuation of EEG functional connectivity discriminates seizure phenotypes in mesial temporal lobe epilepsy

OBJECTIVE

We explored whether quantifiable differences between clinical seizures (CSs) and subclinical seizures (SCSs) occur in the pre-ictal state.

METHODS

We analyzed pre-ictal stereo-electroencephalography (SEEG) retrospectively across mesial temporal lobe epilepsy patients with recorded CSs and SCSs. Power spectral density and functional connectivity (FC) were quantified within and between the seizure onset zone (SOZ) and the early propagation zone (PZ), respectively. To evaluate the fluctuation of neural connectivity, FC variability was computed. Measures were further verified by a logistic regression model to evaluate their classification potentiality through the area under the receiver-operating-characteristics curve (AUC).

RESULTS

Fifty-four pre-ictal SEEG epochs (27 CSs and 27 SCSs) were selected among 14 patients. Within the SOZ, pre-ictal FC variability of CSs was larger than SCSs in 1-45 Hz during 30 seconds before seizure onset. Pre-ictal FC variability between the SOZ and PZ was larger in SCSs than CSs in 55-80 Hz within 1 minute before onset. Using these two variables, the logistic regression model achieved an AUC of 0.79 when classifying CSs and SCSs.

CONCLUSIONS

Pre-ictal FC variability within/between epileptic zones, not signal power or FC value, distinguished SCSs from CSs.

SIGNIFICANCE

Pre-ictal epileptic network stability possibly marks seizure phenotypes, contributing insights into ictogenesis and potentially helping seizure prediction.

Efficacy of bedside optic nerve sheath diameter measurement in differentiating provoked seizure from unprovoked seizure in the emergency department

BACKGROUND

Emergency departments (EDs) are typically the first medical contact for seizure patients, and early diagnosis and treatment is primarily the responsibility of emergency physicians.

OBJECTIVES

Demonstrate the efficacy of bedside ocular ultrasonography for optic nerve sheath diameter (ONSD) measurement in differentiating provoked seizure from unprovoked seizure in the ED.

DESIGN

Prospective observational study SETTINGS: Tertiary care hospital PATIENTS AND METHODS: Patients presenting to the ED with seizure were divided into two groups according to medical history, physical examination, laboratory results, cranial computed tomography findings and electroencephalography results. Patients with seizures that did not have a specific cause (unprovoked) were compared with patients who had seizures caused by underlying pathology (provoked). The measurement of the ONSD was taken at the bedside within 30 minutes of arrival. The study compared the ONSD values, age, sex, type of seizure, and Glasgow Coma Score between the two groups.

MAIN OUTCOME MEASURE

Efficacy of ONSD to distinguish between provoked and unprovoked seizures.

SAMPLE SIZE

210 patients RESULTS: One hundred and fourteen (54.3%) patients were in the provoked seizure group and 96 (45.7%) were in the unprovoked seizure group. The ONSD measurements were significantly higher in the provoked seizure group compared with the unprovoked seizure group (median 6.1 mm vs. 5.2 mm, P<.001). The cut-off value of ONSD higher than 5.61 was significantly associated with the prediction of the provoked seizure (P<.001). The area under the curve value was 0.882 (95% CI: 0.830-0.922) with a sensitivity of 86.5 and specificity of 78.9%.

CONCLUSIONS

Bedside ONSD measurement by means of ocular ultrasound is an effective method for differentiating provoked seizure from unprovoked seizure.

LIMITATIONS

Statistical significance of age on ONSD and exclusion of pediatric patients.

CONFLICT OF INTEREST

None.

Deep learning-based EEG emotion recognition: Current trends and future perspectives

Automatic electroencephalogram (EEG) emotion recognition is a challenging component of human-computer interaction (HCI). Inspired by the powerful feature learning ability of recently-emerged deep learning techniques, various advanced deep learning models have been employed increasingly to learn high-level feature representations for EEG emotion recognition. This paper aims to provide an up-to-date and comprehensive survey of EEG emotion recognition, especially for various deep learning techniques in this area. We provide the preliminaries and basic knowledge in the literature. We review EEG emotion recognition benchmark data sets briefly. We review deep learning techniques in details, including deep belief networks, convolutional neural networks, and recurrent neural networks. We describe the state-of-the-art applications of deep learning techniques for EEG emotion recognition in detail. We analyze the challenges and opportunities in this field and point out its future directions.

A comparison of uni- and multi-variate methods for identifying brain networks activated by cognitive tasks using intracranial EEG

Cognitive tasks are commonly used to identify brain networks involved in the underlying cognitive process. However, inferring the brain networks from intracranial EEG data presents several challenges related to the sparse spatial sampling of the brain and the high variability of the EEG trace due to concurrent brain processes. In this manuscript, we use a well-known facial emotion recognition task to compare three different ways of analyzing the contrasts between task conditions: permutation cluster tests, machine learning (ML) classifiers, and a searchlight implementation of multivariate pattern analysis (MVPA) for intracranial sparse data recorded from 13 patients undergoing presurgical evaluation for drug-resistant epilepsy. Using all three methods, we aim at highlighting the brain structures with significant contrast between conditions. In the absence of ground truth, we use the scientific literature to validate our results. The comparison of the three methods’ results shows moderate agreement, measured by the Jaccard coefficient, between the permutation cluster tests and the machine learning [0.33 and 0.52 for the left (LH) and right (RH) hemispheres], and 0.44 and 0.37 for the LH and RH between the permutation cluster tests and MVPA. The agreement between ML and MVPA is higher: 0.65 for the LH and 0.62 for the RH. To put these results in context, we performed a brief review of the literature and we discuss how each brain structure’s involvement in the facial emotion recognition task.

Spectral and network characterization of focal seizure types and phases

BACKGROUND AND OBJECTIVE

Currently, epileptic seizure characterization relies on several clinical features that allow their classification into different types. The present work aims to characterize both seizure types and phases based exclusively on electrophysiological characteristics.

METHODS

Based on the analysis of intracranial EEG recordings of 129 seizures from 22 patients obtained from the European Epilepsy Database, network and spectral measures were calculated in five-second temporal windows. Statistically significant differences between each window of the seizure phases (preictal, ictal, and postictal) and the interictal phase were used to identify/classify seizure types and their phases. A support vector machine (SVM) working on a multidimensional feature space of network and spectral measures was implemented for the classification of each seizure type; a traditional statistical approach was also conducted to highlight the underlying patterns to each seizure type or phase.

RESULTS

The percentage of correct classification of seizure types, corrected by chance, provided by the SVM exceeded 70%, considering all measures and the entire seizure (preictal + ictal + postictal). This percentage increased to more than 80% when all the measures during the ictal period for the depth electrodes or during the postictal for subdural electrodes were considered. Regarding the statistical approach, several measures presented a monotonic ascending and descending behavior with respect to seizure severity; these changes were observed during the ictal and postictal periods. Some measures were specific of each seizure type.

CONCLUSIONS

Our results provide a new framework to seizure characterization and reveal the possibility of an exclusively intracranial EEG-based classification. This could be used to build an automatic seizure classification system and provides new evidence of the network-related physiopathology of epilepsies. Thus, the novelty of this work is the possibility of differentiating seizure types based exclusively on the EEG recordings, providing evidence of the underlying patterns or characteristics to each seizure type and/or phase that would allow their optimal classification.

Epileptic seizure onset predicts its duration

BACKGROUND

Epilepsy is characterized by recurrent seizures that have a variety of manifestations. The severity of, and risks for patients associated with, seizures are largely linked to the duration of seizures. Methods that determine seizure duration based on seizure onsets could be used to help mitigate the risks associated with what might be extended seizures by guiding timely interventions.

METHODS

Using long-term intracranial electroencephalography (iEEG) recordings, this article presents a method for predicting whether a seizure is going to be long or short by analyzing the seizure onset. The definition of long and short depends on each patient's seizure distribution. By analyzing 2954 seizures from 10 patients, patient-specific classifiers were built to predict seizure duration given the first few seconds from the onset.

RESULTS

The proposed methodology achieved an average area under the receiver operating characteristic curve (AUC) performance of 0.7 for the 5 of 10 patients with above chance prediction performance (p value from 0.04 to 10^-9 ).

CONCLUSIONS

Our results imply that the duration of seizures can be predicted from the onset in some patients. This could form the basis of methods for predicting status epilepticus or optimizing the amount of electrical stimulation delivered by seizure control devices.

Evidence for long memory in focal seizure duration

Objective

A major source of disability for people with epilepsy involves uncertainty surrounding seizure timing and severity. Although patients often report that long seizure-free intervals are followed by more severe seizures, there is little experimental evidence supporting this observation. Optimal characterization of seizure severity is debated; however, seizure duration is associated with seizure type and can be quantified in electrographic recordings as a limited proxy of clinical seizure severity. Here, using chronic intracranial electroencephalography (cEEG), we investigate the relationship between interseizure interval (ISI) and duration of the subsequent seizure.

Methods

We performed a retrospective analysis of 14 subjects implanted with a responsive neurostimulation device (RNS System) that provides cEEG, including timestamps of electrographic seizures. We determined seizure durations for isolated seizures and for representative seizures from clusters determined through unsupervised methods. For each subject, the median ISI preceding long-duration seizures, defined as the top quintile of seizure durations, was compared with the median ISI preceding seizures with durations in the residual quintiles. In a group analysis, the mean seizure duration and the proportion of long-duration seizures were compared across ISI categories representing different lengths.

Results

For 5 out of 14 subjects (36%), the median ISI preceding long-duration seizures was significantly greater than the median ISI preceding shorter-duration seizures. In the group analysis, when ISI was categorized by length, the proportion of long-duration seizures within the high ISI category was significantly higher than that of the low ISI category (P < 0.001).

Significance

By leveraging cEEG and accounting for seizure clusters, we found that the likelihood of long-duration seizures positively correlates with ISI length, in a subset of individuals. These findings corroborate anecdotal clinical observations and support the existence of capacitor-like long memory processes governing the dynamics of focal seizures.

Spectral organization of focal seizures within the thalamotemporal network

Objective

To investigate dynamic changes in neural activity between the anterior nucleus of the thalamus (ANT) and the seizure onset zone (SOZ) in patients with drug‐resistant temporal lobe epilepsy (TLE) based on anatomic location, seizure subtype, and state of vigilance (SOV).

Methods

Eleven patients undergoing stereoelectroencephalography for seizure localization were recruited prospectively for local field potential (LFP) recording directly from the ANT. The SOZ was identified using line length and epileptogenicity index. Changes in power spectral density (PSD) were compared between the two anatomic sites as seizures (N = 53) transitioned from interictal baseline to the posttermination stage.

Results

At baseline, the thalamic LFPs were significantly lower and distinct from the SOZ with the presence of higher power in the fast ripple band (P < 0.001). Temporal changes in ictal power of neural activity within ANT mimic those of the SOZ, are increased significantly at seizure onset (P < 0.05), and are distinct for seizures that impaired awareness or that secondarily generalized (P < 0.05). The onset of seizure was preceded by a decrease in the mean power spectral density (PSD) in ANT and SOZ (P < 0.05). Neural activity correlated with different states of vigilance at seizure onset within the ANT but not in the SOZ (P = 0.005).

Interpretation

The ANT can be recruited at the onset of mesial temporal lobe seizures, and the recruitment pattern differs with seizure subtypes. Furthermore, changes in neural dynamics precede seizure onset and are widespread to involve temporo‐thalamic regions, thereby providing an opportunity to intervene early with closed‐loop DBS.

Prediction of epilepsy seizure from multi-channel electroencephalogram by effective connectivity analysis using Granger causality and directed transfer function methods

Epilepsy is a chronic disorder, which causes strange perceptions, muscle spasms, sometimes seizures, and loss of awareness, associated with abnormal neuronal activity in the brain. The goal of this study is to investigate how effective connectivity (EC) changes effect on unexpected seizures prediction, as this will authorize the patients to play it safe and avoid risk. We approve the hypothesis that EC variables near seizure change significantly so seizure can be predicted in accordance with this variation. We introduce two time-variant coefficients based on standard deviation of EC on Freiburg EEG dataset by using directed transfer function and Granger causality methods and compare index changes over the course of time in five different frequency bands. Comparison of the multivariate and bivariate analysis of factors is implemented in this investigation. The performance based on the suggested methods shows the seizure occurrence period is approximately 50 min that is expected onset stated in, the maximum value of sensitivity approaching ~ 80%, and 0.33 FP/h is the false prediction rate. The findings revealed that greater accuracy and sensitivity are obtained by the designed system in comparison with the results of other works in the same condition. Even though these results still are not sufficient for clinical applications. Based on the conclusions, it can generally be observed that the greater results by DTF method are in the gamma and beta frequency bands.

Seizure pathways: A model-based investigation

We present the results of a model inversion algorithm for electrocorticography (ECoG) data recorded during epileptic seizures. The states and parameters of neural mass models were tracked during a total of over 3000 seizures from twelve patients with focal epilepsy. These models provide an estimate of the effective connectivity within intracortical circuits over the time course of seizures. Observing the dynamics of effective connectivity provides insight into mechanisms of seizures. Estimation of patients seizure dynamics revealed: 1) a highly stereotyped pattern of evolution for each patient, 2) distinct sub-groups of onset mechanisms amongst patients, and 3) different offset mechanisms for long and short seizures. Stereotypical dynamics suggest that, once initiated, seizures follow a deterministic path through the parameter space of a neural model. Furthermore, distinct sub-populations of patients were identified based on characteristic motifs in the dynamics at seizure onset. There were also distinct patterns between long and short duration seizures that were related to seizure offset. Understanding how these different patterns of seizure evolution arise may provide new insights into brain function and guide treatment for epilepsy, since specific therapies may have preferential effects on the various parameters that could potentially be individualized. Methods that unite computational models with data provide a powerful means to generate testable hypotheses for further experimental research. This work provides a demonstration that the hidden connectivity parameters of a neural mass model can be dynamically inferred from data. Our results underscore the power of theoretical models to inform epilepsy management. It is our hope that this work guides further efforts to apply computational models to clinical data.

A Comparison of Machine Learning Classifiers for Energy-Efficient Implementation of Seizure Detection

The closed-loop application of electrical stimulation via chronically implanted electrodes is a novel approach to stop seizures in patients with focal-onset epilepsy. To this end, an energy efficient seizure detector that can be implemented in an implantable device is of crucial importance. In this study, we first evaluated the performance of two machine learning algorithms (Random Forest classifier and support vector machine (SVM)) by using selected time and frequency domain features with a limited need of computational resources. Performance of the algorithms was further compared to a detection strategy implemented in an existing closed loop neurostimulation device for the treatment of epilepsy. The results show a superior performance of the Random Forest classifier compared to the SVM classifier and the reference approach. Next, we implemented the feature extraction and classification process of the Random Forest classifier on a microcontroller to evaluate the energy efficiency of this seizure detector. In conclusion, the feature set in combination with Random Forest classifier is an energy efficient hardware implementation that shows an improvement of detection sensitivity and specificity compared to the presently available closed-loop intervention in epilepsy while preserving a low detection delay.