Seizure prediction: Any better than chance?

"Brain state network dynamics in pediatric epilepsy: Chaotic attractor transition ensemble network"

Traditional scalp EEG signal analysis in pediatric epilepsy is limited by poor spatial resolution, susceptibility to noise and artifacts, and difficulty in accurately localizing epileptic activity, especially from deep or interconnected brain regions. Additionally, such methods often overlook the dynamic nature of brain states and seizure propagation, while reliance on visual inspection introduces variability in interpretation. These limitations hinder precise seizure detection and the mechanistic understanding of brain network dynamics. Here, we offer an alternative approach that addresses these challenges, and eventually enables effective clinical interventions to improve patient outcomes. By incorporating chaos and dynamical systems theory, we present and validate a novel ensemble framework, Chaotic Attractor Transition Ensemble Network for Epilepsy (CATE-NET), which identifies neuro-dynamical signatures underlying pediatric epilepsy, facilitating the discrimination between physiological brain activity and seizure-induced signal irregularities. CATE-NET is modularly designed to leverage nonlinear dynamics of EEG signals and chaotic attractors, particularly the Rössler chaotic attractor to model scalp EEG data. This is followed by a long short-term memory network module for the automatic analysis of brain states. The final module utilizes probabilistic graphing to map the output of the LSTM to state transition graphs, between pre-ictal, inter-ictal, ictal, and ictal-free brain states. Model metrics include a classification accuracy of 0.98, sensitivity of 0.76, specificity of 0.84, and an AUC value of 0.91 when distinguishing among ictal, inter-ictal, and ictal-free brain states. Additionally, the system integrates flexible horizon windows of 10, 20, and 30 min to determine brain state transitions. We demonstrate that nonlinear dynamics present in epileptic brain states derived from the Rössler chaotic attractor are effective features to compute brain state analysis and visualize pediatric epileptic brain state topology. CATE-NET introduces a novel platform for brain state analysis, feature extraction, and topological mapping in pediatric epilepsy by combining chaotic attractors, deep learning, and probabilistic graphing. By integrating explainable AI (XAI), the framework clarifies how chaotic attractor patterns and probabilistic transitions contribute to brain state classifications, seizure state dynamic transitions. This approach reveals the spatial organization and EEG signal dynamics of pediatric epileptic brain states, allowing integration with clinical EEG equipment to potentially improve seizure management and real time decision making.

Seizure forecasting: Bifurcations in the long and winding road

To date, the unpredictability of seizures remains a source of suffering for people with epilepsy, motivating decades of research into methods to forecast seizures. Originally, only few scientists and neurologists ventured into this niche endeavor, which, given the difficulty of the task, soon turned into a long and winding road. Over the past decade, however, our narrow field has seen a major acceleration, with trials of chronic electroencephalographic devices and the subsequent discovery of cyclical patterns in the occurrence of seizures. Now, a burgeoning science of seizure timing is emerging, which in turn informs best forecasting strategies for upcoming clinical trials. Although the finish line might be in view, many challenges remain to make seizure forecasting a reality. This review covers the most recent scientific, technical, and medical developments, discusses methodology in detail, and sets a number of goals for future studies.

Seizure forecasting: Where do we stand?

A lot of mileage has been made recently on the long and winding road toward seizure forecasting. Here we briefly review some selected milestones passed along the way, which were discussed at the International Conference for Technology and Analysis of Seizures-ICTALS 2022-convened at the University of Bern, Switzerland. Major impetus was gained recently from wearable and implantable devices that record not only electroencephalography, but also data on motor behavior, acoustic signals, and various signals of the autonomic nervous system. This multimodal monitoring can be performed for ultralong timescales covering months or years. Accordingly, features and metrics extracted from these data now assess seizure dynamics with a greater degree of completeness. Most prominently, this has allowed the confirmation of the long-suspected cyclical nature of interictal epileptiform activity, seizure risk, and seizures. The timescales cover daily, multi-day, and yearly cycles. Progress has also been fueled by approaches originating from the interdisciplinary field of network science. Considering epilepsy as a large-scale network disorder yielded novel perspectives on the pre-ictal dynamics of the evolving epileptic brain. In addition to discrete predictions that a seizure will take place in a specified prediction horizon, the community broadened the scope to probabilistic forecasts of a seizure risk evolving continuously in time. This shift of gears triggered the incorporation of additional metrics to quantify the performance of forecasting algorithms, which should be compared to the chance performance of constrained stochastic null models. An imminent task of utmost importance is to find optimal ways to communicate the output of seizure-forecasting algorithms to patients, caretakers, and clinicians, so that they can have socioeconomic impact and improve patients' well-being.

On temporal scale-free non-periodic stimulation and its mechanisms as an infinite improbability drive of the brain's functional connectogram

Rationalized development of electrical stimulation (ES) therapy is of paramount importance. Not only it will foster new techniques and technologies with increased levels of safety, efficacy, and efficiency, but it will also facilitate the translation from basic research to clinical practice. For such endeavor, design of new technologies must dialogue with state-of-the-art neuroscientific knowledge. By its turn, neuroscience is transitioning-a movement started a couple of decades earlier-into adopting a new conceptual framework for brain architecture, in which time and thus temporal patterns plays a central role in the neuronal representation of sampled data from the world. This article discusses how neuroscience has evolved to understand the importance of brain rhythms in the overall functional architecture of the nervous system and, consequently, that neuromodulation research should embrace this new conceptual framework. Based on such support, we revisit the literature on standard (fixed-frequency pulsatile stimuli) and mostly non-standard patterns of ES to put forward our own rationale on how temporally complex stimulation schemes may impact neuromodulation strategies. We then proceed to present a low frequency, on average (thus low energy), scale-free temporally randomized ES pattern for the treatment of experimental epilepsy, devised by our group and termed NPS (Non-periodic Stimulation). The approach has been shown to have robust anticonvulsant effects in different animal models of acute and chronic seizures (displaying dysfunctional hyperexcitable tissue), while also preserving neural function. In our understanding, accumulated mechanistic evidence suggests such a beneficial mechanism of action may be due to the natural-like characteristic of a scale-free temporal pattern that may robustly compete with aberrant epileptiform activity for the recruitment of neural circuits. Delivering temporally patterned or random stimuli within specific phases of the underlying oscillations (i.e., those involved in the communication within and across brain regions) could both potentiate and disrupt the formation of neuronal assemblies with random probability. The usage of infinite improbability drive here is obviously a reference to the "The Hitchhiker's Guide to the Galaxy" comedy science fiction classic, written by Douglas Adams. The parallel is that dynamically driving brain functional connectogram, through neuromodulation, in a manner that would not favor any specific neuronal assembly and/or circuit, could re-stabilize a system that is transitioning to fall under the control of a single attractor. We conclude by discussing future avenues of investigation and their potentially disruptive impact on neurotechnology, with a particular interest in NPS implications in neural plasticity, motor rehabilitation, and its potential for clinical translation.

Temporally Targeted Interactions With Pathologic Oscillations as Therapeutical Targets in Epilepsy and Beyond

Self-organized neuronal oscillations rely on precisely orchestrated ensemble activity in reverberating neuronal networks. Chronic, non-malignant disorders of the brain are often coupled to pathological neuronal activity patterns. In addition to the characteristic behavioral symptoms, these disturbances are giving rise to both transient and persistent changes of various brain rhythms. Increasing evidence support the causal role of these "oscillopathies" in the phenotypic emergence of the disease symptoms, identifying neuronal network oscillations as potential therapeutic targets. While the kinetics of pharmacological therapy is not suitable to compensate the disease related fine-scale disturbances of network oscillations, external biophysical modalities (e.g., electrical stimulation) can alter spike timing in a temporally precise manner. These perturbations can warp rhythmic oscillatory patterns via resonance or entrainment. Properly timed phasic stimuli can even switch between the stable states of networks acting as multistable oscillators, substantially changing the emergent oscillatory patterns. Novel transcranial electric stimulation (TES) approaches offer more reliable neuronal control by allowing higher intensities with tolerable side-effect profiles. This precise temporal steerability combined with the non- or minimally invasive nature of these novel TES interventions make them promising therapeutic candidates for functional disorders of the brain. Here we review the key experimental findings and theoretical background concerning various pathological aspects of neuronal network activity leading to the generation of epileptic seizures. The conceptual and practical state of the art of temporally targeted brain stimulation is discussed focusing on the prevention and early termination of epileptic seizures.

An Overview of EEG-based Machine Learning Methods in Seizure Prediction and Opportunities for Neurologists in this Field

The unpredictability of epileptic seizures is one of the most problematic aspects of the field of epilepsy. Methods or devices capable of detecting seizures minutes before they occur may help prevent injury or even death and significantly improve the quality of life. Machine learning (ML) is an emerging technology that can markedly enhance algorithm performance by interpreting data. ML has gained increasing attention from medical researchers in recent years. Its epilepsy applications range from the localization of the epileptic region, predicting the medical or surgical outcome of epilepsy, and automated electroencephalography (EEG) analysis to seizure prediction. While ML has good prospects with regard to detecting epileptic seizures via EEG signals, many clinicians are still unfamiliar with this field. This work briefly summarizes the history and recent significant progress made in this field and clarifies the essential components of the automatic seizure detection system using ML methodologies for clinicians. This review also proposes how neurologists can actively contribute to ensure improvements in seizure prediction using EEG-based ML.

Technical considerations for evaluating clinical prediction indices: a case study for predicting code blue events with MEWS

Objective.There have been many efforts to develop tools predictive of health deterioration in hospitalized patients, but comprehensive evaluation of their predictive ability is often lacking to guide implementation in clinical practice. In this work, we propose new techniques and metrics for evaluating the performance of predictive alert algorithms and illustrate the advantage of capturing the timeliness and the clinical burden of alerts through the example of the modified early warning score (MEWS) applied to the prediction of in-hospital code blue events.Approach. Different implementations of MEWS were calculated from available physiological parameter measurements collected from the electronic health records of ICU adult patients. The performance of MEWS was evaluated using conventional and a set of non-conventional metrics and approaches that take into account the timeliness and practicality of alarms as well as the false alarm burden.Main results. MEWS calculated using the worst-case measurement (i.e. values scoring 3 points in the MEWS definition) over 2 h intervals significantly reduced the false alarm rate by over 50% (from 0.19/h to 0.08/h) while maintaining similar sensitivity levels as MEWS calculated from raw measurements (∼80%). By considering a prediction horizon of 12 h preceding a code blue event, a significant improvement in the specificity (∼60%), the precision (∼155%), and the work-up to detection ratio (∼50%) could be achieved, at the cost of a relatively marginal decrease in sensitivity (∼10%).Significance. Performance aspects pertaining to the timeliness and burden of alarms can aid in understanding the potential utility of a predictive alarm algorithm in clinical settings.

Prediction of Seizure Recurrence. A Note of Caution

Great strides have been made recently in documenting that machine-learning programs can predict seizure occurrence in people who have epilepsy. Along with this progress have come claims that appear to us to be a bit premature. We anticipate that many people will benefit from seizure prediction. We also doubt that all will benefit. Although machine learning is a useful tool for aiding discovery, we believe that the greatest progress will come from deeper understanding of seizures, epilepsy, and the EEG features that enable seizure prediction. In this essay, we lay out reasons for optimism and skepticism.

Seizure Cycles in Focal Epilepsy

Importance

Focal epilepsy is characterized by the cyclical recurrence of seizures, but, to our knowledge, the prevalence and patterns of seizure cycles are unknown.

Objective

To establish the prevalence, strength, and temporal patterns of seizure cycles over timescales of hours to years.

Design, Setting, and Participants

This retrospective cohort study analyzed data from continuous intracranial electroencephalography (cEEG) and seizure diaries collected between January 19, 2004, and May 18, 2018, with durations up to 10 years. A total of 222 adults with medically refractory focal epilepsy were selected from 256 total participants in a clinical trial of an implanted responsive neurostimulation device. Selection was based on availability of cEEG and/or self-reports of disabling seizures.

Exposures

Antiseizure medications and responsive neurostimulation, based on clinical indications.

Main Outcomes and Measures

Measures involved (1) self-reported daily seizure counts, (2) cEEG-based hourly counts of electrographic seizures, and (3) detections of interictal epileptiform activity (IEA), which fluctuates in daily (circadian) and multiday (multidien) cycles. Outcomes involved descriptive characteristics of cycles of IEA and seizures: (1) prevalence, defined as the percentage of patients with a given type of seizure cycle; (2) strength, defined as the degree of consistency with which seizures occur at certain phases of an underlying cycle, measured as the phase-locking value (PLV); and (3) seizure chronotypes, defined as patterns in seizure timing evident at the group level.

Results

Of the 222 participants, 112 (50%) were male, and the median age was 35 years (range, 18-66 years). The prevalence of circannual (approximately 1 year) seizure cycles was 12% (24 of 194), the prevalence of multidien (approximately weekly to approximately monthly) seizure cycles was 60% (112 of 186), and the prevalence of circadian (approximately 24 hours) seizure cycles was 89% (76 of 85). Strengths of circadian (mean [SD] PLV, 0.34 [0.18]) and multidien (mean [SD] PLV, 0.34 [0.17]) seizure cycles were comparable, whereas circannual seizure cycles were weaker (mean [SD] PLV, 0.17 [0.10]). Across individuals, circadian seizure cycles showed 5 peaks: morning, mid-afternoon, evening, early night, and late night. Multidien cycles of IEA showed peak periodicities centered around 7, 15, 20, and 30 days. Independent of multidien period length, self-reported and electrographic seizures consistently occurred during the days-long rising phase of multidien cycles of IEA.

Conclusions and Relevance

Findings in this large cohort establish the high prevalence of plural seizure cycles and help explain the natural variability in seizure timing. The results have the potential to inform the scheduling of diagnostic studies, the delivery of time-varying therapies, and the design of clinical trials in epilepsy.

A Study of EEG Feature Complexity in Epileptic Seizure Prediction

The purpose of this study is (1) to provide EEG feature complexity analysis in seizure prediction by inter-ictal and pre-ital data classification and, (2) to assess the between-subject variability of the considered features. In the past several decades, there has been a sustained interest in predicting epilepsy seizure using EEG data. Most methods classify features extracted from EEG, which they assume are characteristic of the presence of an epilepsy episode, for instance, by distinguishing a pre-ictal interval of data (which is in a given window just before the onset of a seizure) from inter-ictal (which is in preceding windows following the seizure). To evaluate the difficulty of this classification problem independently of the classification model, we investigate the complexity of an exhaustive list of 88 features using various complexity metrics, i.e., the Fisher discriminant ratio, the volume of overlap, and the individual feature efficiency. Complexity measurements on real and synthetic data testbeds reveal that that seizure prediction by pre-ictal/inter-ictal feature distinction is a problem of significant complexity. It shows that several features are clearly useful, without decidedly identifying an optimal set.

Electrophysiological Correlates of Brain Health Help Diagnose Epilepsy and Lateralize Seizure Focus

The absence of epileptiform activity in a scalp electroencephalogram (EEG) recorded from a potential epilepsy patient can cause delays in clinical care delivery. Here we present a machine-learning-based approach to find evidence for epilepsy in scalp EEGs that do not contain any epileptiform activity, according to expert visual review (i.e., "normal" EEGs). We found that deviations in the EEG features representing brain health, such as the alpha rhythm, can indicate the potential for epilepsy and help lateralize seizure focus, even when commonly recognized epileptiform features are absent. Hence, we developed a machine-learning-based approach that utilizes alpha-rhythm-related features to classify 1) whether an EEG was recorded from an epilepsy patient, and 2) if so, the seizure-generating side of the patient's brain. We evaluated our approach using "normal" scalp EEGs of 48 patients with drug-resistant focal epilepsy and 144 healthy individuals, and a naive Bayes classifier achieved area under ROC curve (AUC) values of 0.81 and 0.72 for the two classification tasks, respectively. These findings suggest that our methodology is useful in the absence of interictal epileptiform activity and can enhance the probability of diagnosing epilepsy at the earliest possible time.

Seizure prediction - ready for a new era

Epilepsy is a common disorder characterized by recurrent seizures. An overwhelming majority of people with epilepsy regard the unpredictability of seizures as a major issue. More than 30 years of international effort have been devoted to the prediction of seizures, aiming to remove the burden of unpredictability and to couple novel, time-specific treatment to seizure prediction technology. A highly influential review published in 2007 concluded that insufficient evidence indicated that seizures could be predicted. Since then, several advances have been made, including successful prospective seizure prediction using intracranial EEG in a small number of people in a trial of a real-time seizure prediction device. In this Review, we examine advances in the field, including EEG databases, seizure prediction competitions, the prospective trial mentioned and advances in our understanding of the mechanisms of seizures. We argue that these advances, together with statistical evaluations, set the stage for a resurgence in efforts towards the development of seizure prediction methodologies. We propose new avenues of investigation involving a synergy between mechanisms, models, data, devices and algorithms and refine the existing guidelines for the development of seizure prediction technology to instigate development of a solution that removes the burden of the unpredictability of seizures.

Prediction error connectivity: A new method for EEG state analysis

Several models have been proposed to explain brain regional and interregional communication, the majority of them using methods that tap the frequency domain, like spectral coherence. Considering brain interareal communication as binary interactions, we describe a novel method devised to predict dynamics and thus highlight abrupt changes marked by unpredictability. Based on a variable-order Markov model algorithm developed in-house for data compression, the prediction error connectivity (PEC) estimates network transitions by calculating error matrices (EMs). We analysed 20 h of EEG signals of virtual networks generated with a neural mass model. Subnetworks changed through time (2 of 5 signals), from normal to normal or pathological states. PEC was superior to spectral coherence in detecting all considered transitions, especially in broad and ripple bands. Subsequently, EMs of real data were classified using a support vector machine in order to capture the transition from interictal to preictal state and calculate seizure risk. A single seizure was randomly selected for training. Through this approach it was possible to establish a threshold that the calculated risk consistently overcame minutes before the events. Using either spectral coherence or PEC we created 1000 models that successfully predicted 6 seizures (100% sensibility), a whole cluster recorded in a patient with hippocampal epilepsy. However, PEC resulted superior to coherence in terms of true seizure free time and amount of false warnings. Indeed, the best PEC model predicted 96% of interictal time (vs. 83% of coherence) of about 20 h of stereo-EEG. This analysis was extended to patients with neo/mesocortical temporal, neocortical frontal, parietal and occipital lobe epilepsy. Again PEC showed high performance, allowing the prediction of 31 events distributed across 10 days with ROC AUCs that reached 98% (average 93 ± 5%) in 6 different patients. Moreover, considering another state transition, PEC could classify and forecast up to 88% (average 85 ± 3%) of the REM phase both in deep and scalp EEG. In conclusion, PEC is a novel approach that relies on pattern analysis in the time-domain. We believe that this method can be successfully employed both for the study of brain connectivity, and also implemented in real-life solutions for seizure detection and prediction.

Deep brain stimulation probing performance is enhanced by pairing stimulus with epileptic seizure

The unpredictability of spontaneous and recurrent seizures significantly impairs the quality of life of patients with epilepsy. Probing neural network excitability with deep brain electrical stimulation (DBS) has shown promising results predicting pathological shifts in brain states. This work presents a proof-of-principal that active electroencephalographic (EEG) probing, as a seizure predictive tool, is enhanced by pairing DBS and the electrographic seizure itself. The ictogenic model used consisted of inducing seizures by continuous intravenous infusion of pentylenetetrazol (PTZ - 2.5 mg/ml/min) while a probing DBS was delivered to the thalamus (TH) or amygdaloid complex to detect changes prior to seizure onset. Cortical electrophysiological recordings were performed before, during, and after PTZ infusion. Thalamic DBS probing, but not amygdaloid, was able to predict seizure onset without any observable proconvulsant effects. However, previously pairing amygdaloid DBS and epileptic polyspike discharges (day-1) elicited distinct preictal cortically recorded evoked response (CRER) (day-2) when compared with control groups that received the same amount of electrical pulses at different moments of the ictogenic progress at day-1. In conclusion, our results have demonstrated that the pairing strategy potentiated the detection of an altered brain state prior to the seizure onset. The EEG probing enhancement method opens many possibilities for both diagnosis and treatment of epilepsy.

Automatic Computer-Based Detection of Epileptic Seizures

Automatic computer-based seizure detection and warning devices are important for objective seizure documentation, for SUDEP prevention, to avoid seizure related injuries and social embarrassments as a consequence of seizures, and to develop on demand epilepsy therapies. Automatic seizure detection systems can be based on direct analysis of epileptiform discharges on scalp-EEG or intracranial EEG, on the detection of motor manifestations of epileptic seizures using surface electromyography (sEMG), accelerometry (ACM), video detection systems and mattress sensors and finally on the assessment of changes of physiologic parameters accompanying epileptic seizures measured by electrocardiography (ECG), respiratory monitors, pulse oximetry, surface temperature sensors, and electrodermal activity. Here we review automatic seizure detection based on scalp-EEG, ECG, and sEMG. Different seizure types affect preferentially different measurement parameters. While EEG changes accompany all types of seizures, sEMG and ACM are suitable mainly for detection of seizures with major motor manifestations. Therefore, seizure detection can be optimized by multimodal systems combining several measurement parameters. While most systems provide sensitivities over 70%, specificity expressed as false alarm rates still needs to be improved. Patients' acceptance and comfort of a specific device are of critical importance for its long-term application in a meaningful clinical way.

Automated detection of epileptic ripples in MEG using beamformer-based virtual sensors

OBJECTIVE

In epilepsy, high-frequency oscillations (HFOs) are expressively linked to the seizure onset zone (SOZ). The detection of HFOs in the noninvasive signals from scalp electroencephalography (EEG) and magnetoencephalography (MEG) is still a challenging task. The aim of this study was to automate the detection of ripples in MEG signals by reducing the high-frequency noise using beamformer-based virtual sensors (VSs) and applying an automatic procedure for exploring the time-frequency content of the detected events.

APPROACH

Two-hundred seconds of MEG signal and simultaneous iEEG were selected from nine patients with refractory epilepsy. A two-stage algorithm was implemented. Firstly, beamforming was applied to the whole head to delimitate the region of interest (ROI) within a coarse grid of MEG-VS. Secondly, a beamformer using a finer grid in the ROI was computed. The automatic detection of ripples was performed using the time-frequency response provided by the Stockwell transform. Performance was evaluated through comparisons with simultaneous iEEG signals.

MAIN RESULTS

ROIs were located within the seizure-generating lobes in the nine subjects. Precision and sensitivity values were 79.18% and 68.88%, respectively, by considering iEEG-detected events as benchmarks. A higher number of ripples were detected inside the ROI compared to the same region in the contralateral lobe.

SIGNIFICANCE

The evaluation of interictal ripples using non-invasive techniques can help in the delimitation of the epileptogenic zone and guide placement of intracranial electrodes. This is the first study that automatically detects ripples in MEG in the time domain located within the clinically expected epileptic area taking into account the time-frequency characteristics of the events through the whole signal spectrum. The algorithm was tested against intracranial recordings, the current gold standard. Further studies should explore this approach to enable the localization of noninvasively recorded HFOs to help during pre-surgical planning and to reduce the need for invasive diagnostics.

Towards an Online Seizure Advisory System-An Adaptive Seizure Prediction Framework Using Active Learning Heuristics

In the last decade, seizure prediction systems have gained a lot of attention because of their enormous potential to largely improve the quality-of-life of the epileptic patients. The accuracy of the prediction algorithms to detect seizure in real-world applications is largely limited because the brain signals are inherently uncertain and affected by various factors, such as environment, age, drug intake, etc., in addition to the internal artefacts that occur during the process of recording the brain signals. To deal with such ambiguity, researchers transitionally use active learning, which selects the ambiguous data to be annotated by an expert and updates the classification model dynamically. However, selecting the particular data from a pool of large ambiguous datasets to be labelled by an expert is still a challenging problem. In this paper, we propose an active learning-based prediction framework that aims to improve the accuracy of the prediction with a minimum number of labelled data. The core technique of our framework is employing the Bernoulli-Gaussian Mixture model (BGMM) to determine the feature samples that have the most ambiguity to be annotated by an expert. By doing so, our approach facilitates expert intervention as well as increasing medical reliability. We evaluate seven different classifiers in terms of the classification time and memory required. An active learning framework built on top of the best performing classifier is evaluated in terms of required annotation effort to achieve a high level of prediction accuracy. The results show that our approach can achieve the same accuracy as a Support Vector Machine (SVM) classifier using only 20 % of the labelled data and also improve the prediction accuracy even under the noisy condition.

A forward-looking review of seizure prediction

PURPOSE OF REVIEW

Seizure prediction has made important advances over the last decade, with the recent demonstration that prospective seizure prediction is possible, though there remain significant obstacles to broader application. In this review, we will describe insights gained from long-term trials, with the aim of identifying research goals for the next decade.

RECENT FINDINGS

Unexpected results from these studies, including strong and highly individual relationships between spikes and seizures, diurnal patterns of seizure activity, and the coexistence of different seizure populations within individual patients exhibiting distinctive dynamics, have caused us to re-evaluate many prior assumptions in seizure prediction studies and suggest alternative strategies that could be employed in the search for algorithms providing greater clinical utility. Advances in analytical approaches, particularly deep-learning techniques, harbour great promise and in combination with less-invasive systems with sufficiently power-efficient computational capacity will bring broader clinical application within reach.

SUMMARY

We conclude the review with an exercise in wishful thinking, which asks what the ideal seizure prediction dataset would look like and how these data should be manipulated to maximize benefits for patients. The motivation for structuring the review in this way is to create a forward-looking, optimistic critique of the existing methodologies.

Predicting epileptic seizures from scalp EEG based on attractor state analysis

BACKGROUND AND OBJECTIVE

Epilepsy is the second most common disease of the brain. Epilepsy makes it difficult for patients to live a normal life because it is difficult to predict when seizures will occur. In this regard, if seizures could be predicted a reasonable period of time before their occurrence, epilepsy patients could take precautions against them and improve their safety and quality of life. In this paper, we investigate a novel seizure precursor based on attractor state analysis for seizure prediction.

METHODS

We analyze the transition process from normal to seizure attractor state and investigate a precursor phenomenon seen before reaching the seizure attractor state. From the result of an analysis, we define a quantified spectral measure in scalp EEG for seizure prediction. From scalp EEG recordings, the Fourier coefficients of six EEG frequency bands are extracted, and the defined spectral measure is computed based on the coefficients for each half-overlapped 20-second-long window. The computed spectral measure is applied to seizure prediction using a low-complexity methodology.

RESULTS

Within scalp EEG, we identified an early-warning indicator before an epileptic seizure occurs. Getting closer to the bifurcation point that triggers the transition from normal to seizure state, the power spectral density of low frequency bands of the perturbation of an attractor in the EEG, showed a relative increase. A low-complexity seizure prediction algorithm using this feature was evaluated, using ∼583h of scalp EEG in which 143 seizures in 16 patients were recorded. With the test dataset, the proposed method showed high sensitivity (86.67%) with a false prediction rate of 0.367h^-1 and average prediction time of 45.3min.

CONCLUSIONS

A novel seizure prediction method using scalp EEG, based on attractor state analysis, shows potential for application with real epilepsy patients. This is the first study in which the seizure-precursor phenomenon of an epileptic seizure is investigated based on attractor-based analysis of the macroscopic dynamics of the brain. With the scalp EEG, we first propose use of a spectral feature identified for seizure prediction, in which the dynamics of an attractor are excluded, and only the perturbation dynamics from the attractor are considered.

A Realistic Seizure Prediction Study Based on Multiclass SVM

A patient-specific algorithm, for epileptic seizure prediction, based on multiclass support-vector machines (SVM) and using multi-channel high-dimensional feature sets, is presented. The feature sets, combined with multiclass classification and post-processing schemes aim at the generation of alarms and reduced influence of false positives. This study considers 216 patients from the European Epilepsy Database, and includes 185 patients with scalp EEG recordings and 31 with intracranial data. The strategy was tested over a total of 16,729.80[Formula: see text]h of inter-ictal data, including 1206 seizures. We found an overall sensitivity of 38.47% and a false positive rate per hour of 0.20. The performance of the method achieved statistical significance in 24 patients (11% of the patients). Despite the encouraging results previously reported in specific datasets, the prospective demonstration on long-term EEG recording has been limited. Our study presents a prospective analysis of a large heterogeneous, multicentric dataset. The statistical framework based on conservative assumptions, reflects a realistic approach compared to constrained datasets, and/or in-sample evaluations. The improvement of these results, with the definition of an appropriate set of features able to improve the distinction between the pre-ictal and nonpre-ictal states, hence minimizing the effect of confounding variables, remains a key aspect.

On predicting epileptic seizures from intracranial electroencephalography

This study investigates the sensitivity and specificity of predicting epileptic seizures from intracranial electroencephalography (iEEG). A monitoring system is studied to generate an alarm upon detecting a precursor of an epileptic seizure. The iEEG traces of ten patients suffering from medically intractable epilepsy were used to build a prediction model. From the iEEG recording of each patient, power spectral densities were calculated and classified using support vector machines. The prediction results varied across patients. For seven patients, seizures were predicted with 100% sensitivity without any false alarms. One patient showed good sensitivity but lower specificity, and the other two patients showed lower sensitivity and specificity. Predictive analytics based on the spectral feature of iEEG performs well for some patients but not all. This result highlights the need for patient-specific prediction models and algorithms.

A Phase-Locked Loop Epilepsy Network Emulator

Most seizure forecasting employs statistical learning techniques that lack a representation of the network interactions that give rise to seizures. We present an epilepsy network emulator (ENE) that uses a network of interconnected phase-locked loops (PLLs) to model synchronous, circuit-level oscillations between electrocorticography (ECoG) electrodes. Using ECoG data from a canine-epilepsy model (Davis et al. 2011) and a physiological entropy measure (approximate entropy or ApEn, Pincus 1995), we demonstrate the entropy of the emulator phases increases dramatically during ictal periods across all ECoG recording sites and across all animals in the sample. Further, this increase precedes the observable voltage spikes that characterize seizure activity in the ECoG data. These results suggest that the ENE is sensitive to phase-domain information in the neural circuits measured by ECoG and that an increase in the entropy of this measure coincides with increasing likelihood of seizure activity. Understanding this unpredictable phase-domain electrical activity present in ECoG recordings may provide a target for seizure detection and feedback control.

Is Using Threshold-Crossing Method and Single Type of Features Sufficient to Achieve Realistic Application of Seizure Prediction?

Objective This study aims to verify whether the simple threshold-crossing method can work well enough to achieve the realistic application of seizure prediction on the basis of a large public database, and examines how a more complex classifier can improve prediction performance. It also verified whether the combination of multiple types of features with a complex classifier can improve prediction performance. Method Phase synchronization and spectral power features were extracted from electroencephalogram recordings. The threshold-crossing method and a support vector machine (SVM) were used to identify preictal and interictal samples. Based on the type of selected features and the manner of classification, 5 different methods were conducted on 19 patients. The performances of these methods were directly compared and tested using a random predictor. In-sample optimization problems were avoided in the feature and parameter selection procedure to obtain credible results. Results The threshold-crossing method could only obtain satisfying prediction results for approximately half of the selected patients. The SVM classifier could significantly improve prediction performance compared with the threshold-crossing method for both types of features. Although the average performance was further improved when both types of features were combined with the SVM classifier, the improvement was insignificant. Conclusion A complex classifier, such as the SVM, is recommended in a realistic prediction device, although it will increase the complexity of the device. Indeed, the simple threshold-crossing method performs well enough for some of the patients. The combination of phase synchronization and spectral power features is unnecessary because of the increased computation complexity.

Seizure Forecasting and the Preictal State in Canine Epilepsy

The ability to predict seizures may enable patients with epilepsy to better manage their medications and activities, potentially reducing side effects and improving quality of life. Forecasting epileptic seizures remains a challenging problem, but machine learning methods using intracranial electroencephalographic (iEEG) measures have shown promise. A machine-learning-based pipeline was developed to process iEEG recordings and generate seizure warnings. Results support the ability to forecast seizures at rates greater than a Poisson random predictor for all feature sets and machine learning algorithms tested. In addition, subject-specific neurophysiological changes in multiple features are reported preceding lead seizures, providing evidence supporting the existence of a distinct and identifiable preictal state.

Crowdsourcing reproducible seizure forecasting in human and canine epilepsy

See Mormann and Andrzejak (doi:10.1093/brain/aww091) for a scientific commentary on this article.  

Seizures are thought to arise from an identifiable pre-ictal state. Brinkmann et al. report the results of an online, open-access seizure forecasting competition using intracranial EEG recordings from canines with naturally occurring epilepsy and human patients undergoing presurgical monitoring. The winning algorithms forecast seizures at rates significantly greater than chance.

See Mormann and Andrzejak (doi:10.1093/brain/aww091) for a scientific commentary on this article.  

Accurate forecasting of epileptic seizures has the potential to transform clinical epilepsy care. However, progress toward reliable seizure forecasting has been hampered by lack of open access to long duration recordings with an adequate number of seizures for investigators to rigorously compare algorithms and results. A seizure forecasting competition was conducted on kaggle.com using open access chronic ambulatory intracranial electroencephalography from five canines with naturally occurring epilepsy and two humans undergoing prolonged wide bandwidth intracranial electroencephalographic monitoring. Data were provided to participants as 10-min interictal and preictal clips, with approximately half of the 60 GB data bundle labelled (interictal/preictal) for algorithm training and half unlabelled for evaluation. The contestants developed custom algorithms and uploaded their classifications (interictal/preictal) for the unknown testing data, and a randomly selected 40% of data segments were scored and results broadcasted on a public leader board. The contest ran from August to November 2014, and 654 participants submitted 17 856 classifications of the unlabelled test data. The top performing entry scored 0.84 area under the classification curve. Following the contest, additional held-out unlabelled data clips were provided to the top 10 participants and they submitted classifications for the new unseen data. The resulting area under the classification curves were well above chance forecasting, but did show a mean 6.54 ± 2.45% (min, max: 0.30, 20.2) decline in performance. The kaggle.com model using open access data and algorithms generated reproducible research that advanced seizure forecasting. The overall performance from multiple contestants on unseen data was better than a random predictor, and demonstrates the feasibility of seizure forecasting in canine and human epilepsy.

All together now: Analogies between chimera state collapses and epileptic seizures

Conceptually and structurally simple mathematical models of coupled oscillator networks can show a rich variety of complex dynamics, providing fundamental insights into many real-world phenomena. A recent and not yet fully understood example is the collapse of coexisting synchronous and asynchronous oscillations into a globally synchronous motion found in networks of identical oscillators. Here we show that this sudden collapse is promoted by a further decrease of synchronization, rather than by critically high synchronization. This strikingly counterintuitive mechanism can be found also in nature, as we demonstrate on epileptic seizures in humans. Analyzing spatiotemporal correlation profiles derived from intracranial electroencephalographic recordings (EEG) of seizures in epilepsy patients, we found a pronounced decrease of correlation at the seizure onsets. Applying our findings in a closed-loop control scheme to models of coupled oscillators in chimera states, we succeed in both provoking and preventing outbreaks of global synchronization. Our findings not only advance the understanding of networks of coupled dynamics but can open new ways to control them, thus offering a vast range of potential new applications.

Time-dependent degree-degree correlations in epileptic brain networks: from assortative to dissortative mixing

We investigate the long-term evolution of degree-degree correlations (assortativity) in functional brain networks from epilepsy patients. Functional networks are derived from continuous multi-day, multi-channel electroencephalographic data, which capture a wide range of physiological and pathophysiological activities. In contrast to previous studies which all reported functional brain networks to be assortative on average, even in case of various neurological and neurodegenerative disorders, we observe large fluctuations in time-resolved degree-degree correlations ranging from assortative to dissortative mixing. Moreover, in some patients these fluctuations exhibit some periodic temporal structure which can be attributed, to a large extent, to daily rhythms. Relevant aspects of the epileptic process, particularly possible pre-seizure alterations, contribute marginally to the observed long-term fluctuations. Our findings suggest that physiological and pathophysiological activity may modify functional brain networks in a different and process-specific way. We evaluate factors that possibly influence the long-term evolution of degree-degree correlations.

Forecasting Seizures Using Intracranial EEG Measures and SVM in Naturally Occurring Canine Epilepsy

Management of drug resistant focal epilepsy would be greatly assisted by a reliable warning system capable of alerting patients prior to seizures to allow the patient to adjust activities or medication. Such a system requires successful identification of a preictal, or seizure-prone state. Identification of preictal states in continuous long- duration intracranial electroencephalographic (iEEG) recordings of dogs with naturally occurring epilepsy was investigated using a support vector machine (SVM) algorithm. The dogs studied were implanted with a 16-channel ambulatory iEEG recording device with average channel reference for a mean (st. dev.) of 380.4 (+87.5) days producing 220.2 (+104.1) days of intracranial EEG recorded at 400 Hz for analysis. The iEEG records had 51.6 (+52.8) seizures identified, of which 35.8 (+30.4) seizures were preceded by more than 4 hours of seizure-free data. Recorded iEEG data were stratified into 11 contiguous, non-overlapping frequency bands and binned into one-minute synchrony features for analysis. Performance of the SVM classifier was assessed using a 5-fold cross validation approach, where preictal training data were taken from 90 minute windows with a 5 minute pre-seizure offset. Analysis of the optimal preictal training time was performed by repeating the cross validation over a range of preictal windows and comparing results. We show that the optimization of feature selection varies for each subject, i.e. algorithms are subject specific, but achieve prediction performance significantly better than a time-matched Poisson random predictor (p<0.05) in 5/5 dogs analyzed.

Measuring predictability of autonomous network transitions into bursting dynamics

Understanding spontaneous transitions between dynamical modes in a network is of significant importance. These transitions may separate pathological and normal functions of the brain. In this paper, we develop a set of measures that, based on spatio-temporal features of network activity, predict autonomous network transitions from asynchronous to synchronous dynamics under various conditions. These metrics quantify spike-timing distributions within a narrow time window as a function of the relative location of the active neurons. We applied these metrics to investigate the properties of these transitions in excitatory-only and excitatory-and-inhibitory networks and elucidate how network topology, noise level, and cellular heterogeneity affect both the reliability and the timeliness of the predictions. The developed measures can be calculated in real time and therefore potentially applied in clinical situations.

Assessing directionality and strength of coupling through symbolic analysis: an application to epilepsy patients

Inferring strength and direction of interactions from electroencephalographic (EEG) recordings is of crucial importance to improve our understanding of dynamical interdependencies underlying various physiological and pathophysiological conditions in the human epileptic brain. We here use approaches from symbolic analysis to investigate--in a time-resolved manner--weighted and directed, short- to long-ranged interactions between various brain regions constituting the epileptic network. Our observations point to complex spatial-temporal interdependencies underlying the epileptic process and their role in the generation of epileptic seizures, despite the massive reduction of the complex information content of multi-day, multi-channel EEG recordings through symbolization. We discuss limitations and potential future improvements of this approach.

Effects of responsive electrical brain stimulation on intracranial electroencephalogram spikes

OBJECTIVES

Responsive cortical electrical stimulation with implanted devices is under investigation for seizures. While designed to terminate seizures, might this stimulation also affect the underlying epileptic process of seizure generation?

MATERIALS AND METHODS

Four patients undergoing intracranial electroencephalogram (EEG) for seizure localization had an external responsive neurostimulator (eRNS) connected to their seizure-onset zones. The eRNS detected interictal EEG spikes and stimulated at the focus. We quantified spikes at three locations: (1) near stimulation, (2) remote but in the same lobe as stimulation, and (3) in different lobe from stimulation. Ten-minute windows were analyzed at three times: (1) baseline, (2) after the first four hours of stimulation, and (3) poststimulation. One blinded investigator performed manual spike counts. Quantitative measures were total spikes, spike-free intervals (continuous ten-sec segments with no spikes), and spike clusters (one-sec intervals with three or more spikes).

RESULTS

Some changes in spikes occurred in each patient, but no uniform pattern emerged. Two general observations were made: (1) spike counts within a given patient exhibited internally consistent changes with stimulation; (2) across patients, the nature of spike count changes varied, indicating patient-to-patient variability. For example, poststimulation, two patients had more and two patients had fewer total spikes. However, when spikes decreased near stimulation, they decreased at other sites, and when spikes increased near stimulation, they increased at other sites.

CONCLUSIONS

Changes in spike occurrence, organization, and topography with stimulation suggest the eRNS affected spike generation and may affect the underlying interictal epileptic process. Case-to-case variability may be due to individual patient factors, and its significance is yet to be determined.

Forecasting seizures in dogs with naturally occurring epilepsy

Seizure forecasting has the potential to create new therapeutic strategies for epilepsy, such as providing patient warnings and delivering preemptive therapy. Progress on seizure forecasting, however, has been hindered by lack of sufficient data to rigorously evaluate the hypothesis that seizures are preceded by physiological changes, and are not simply random events. We investigated seizure forecasting in three dogs with naturally occurring focal epilepsy implanted with a device recording continuous intracranial EEG (iEEG). The iEEG spectral power in six frequency bands: delta (0.1–4 Hz), theta (4–8 Hz), alpha (8–12 Hz), beta (12–30 Hz), low-gamma (30–70 Hz), and high-gamma (70–180 Hz), were used as features. Logistic regression classifiers were trained to discriminate labeled pre-ictal and inter-ictal data segments using combinations of the band spectral power features. Performance was assessed on separate test data sets via 10-fold cross-validation. A total of 125 spontaneous seizures were detected in continuous iEEG recordings spanning 6.5 to 15 months from 3 dogs. When considering all seizures, the seizure forecasting algorithm performed significantly better than a Poisson-model chance predictor constrained to have the same time in warning for all 3 dogs over a range of total warning times. Seizure clusters were observed in all 3 dogs, and when the effect of seizure clusters was decreased by considering the subset of seizures separated by at least 4 hours, the forecasting performance remained better than chance for a subset of algorithm parameters. These results demonstrate that seizures in canine epilepsy are not randomly occurring events, and highlight the feasibility of long-term seizure forecasting using iEEG monitoring.

Channel selection for epilepsy seizure prediction method based on machine learning

The studies on seizure prediction problem have shown great improvement these years. Machine learning based seizure prediction method shows great performance by doing pattern recognition on high-dimensional bivariate synchronization features. However, the computation loading of the machine learning based method may be too high to meet wearable or implantable devices with the power and area constraints. In this work, channel selection is proposed to reduce the channel number from 22 to less than 6 channels and therefore more than 93.73% of the computation loading is saved through the method. The best result shows successful rate of 60.6% in 3-channel cases of ECoG database and successful rate of 70% in 3-channel cases of EEG database.

Seizure prediction in patients with mesial temporal lobe epilepsy using EEG measures of state similarity

OBJECTIVES

In patients with intractable epilepsy, predicting seizures above chance and with clinically acceptable performance has yet to be demonstrated. In this study, an intracranial EEG-based seizure prediction method using measures of similarity with a reference state is proposed.

METHODS

1565 h of continuous intracranial EEG data from 17 patients with mesial temporal lobe epilepsy were investigated. The recordings included 175 seizures. In each patient the data was split into a training set and a testing set. EEG segments were analyzed using continuous wavelet transform. During training, a reference state was defined in the immediate preictal data and used to derive three features quantifying the discrimination between preictal and interictal states. A classifier was then trained in the feature space. Its performance was assessed using testing set and compared with a random predictor for statistical validation.

RESULTS

Better than random prediction performance was achieved in 7 patients. The sensitivity was higher than 85%, the warning rate was less than 0.35/h and the proportion of time under warning was less than 30%.

CONCLUSION

Seizures are predicted above chance in 41% of patients using measures of state similarity.

SIGNIFICANCE

Sensitivity and specificity levels are potentially interesting for closed-loop seizure control applications.

Toward new paradigms of seizure detection

Great effort has been made toward defining and characterizing the pre-ictal state. Many studies have pursued the idea that there are recognizable electrographic (EEG-based) features which occur before overt clinical seizure activity. However, development of reliable EEG-based seizure detection and prediction algorithms has been difficult. In this review, we discuss the concepts of seizure detection vs. prediction and the pre-ictal "clinical milieu" and "EEG milieu". We proceed to discuss novel concepts of seizure detection based on the pre-ictal "physiological milieu"; in particular, we indicate some early evidence for the hypothesis that pre-ictal cell swelling/extracellular space constriction can be detected with novel optical methods. Development and validation of optical seizure detection technology could provide an entirely new translational approach for the many patients with intractable epilepsy.

Seizure prediction with bipolar spectral power features using Adaboost and SVM classifiers

This paper presents the results of our study on finding a lower complexity and yet a robust seizure prediction method using intracranial electroencephalogram (iEEG) recordings. We compare two classifiers: a low-complexity Adaboost and the more complex support vector machine (SVM). Adaboost is a linear classier using decision stumps, and SVM uses a nonlinear Gaussian kernel. Bipolar and/or time-differential spectral power features of different sub-bands are extracted from the iEEG signal. Adaboost is used to simultaneously classify as well as rank the features. Eliminating the low discriminating features reduces computational complexity and power consumption. The top features selected by Adaboost were also used as a feature set for SVM classification. The outputs of classifiers are regularized by applying a moving-average window and a threshold is used to generate alarms. The proposed methods were applied on 8 invasive recordings selected from the EPILEPSIAE database, the European database of EEG seizure recordings. Doublecross validation is used by separating data sets for training and optimization from testing. The key conclusion is that Adaboost performs slightly better than SVM using a reduced feature set on average with significantly less complexity resulting in a sensitivity of 77.1% (27 of 35 seizures in 873 h recordings) and a false alarm rate of 0.18 per hour.

ANALYSIS OF ABSENCE SEIZURE GENERATION USING EEG SPATIAL-TEMPORAL REGULARITY MEASURES

Epileptic seizures are thought to be generated and to evolve through an underlying anomaly of synchronization in the activity of groups of neuronal populations. The related dynamic scenario of state transitions is revealed by detecting changes in the dynamical properties of Electroencephalography (EEG) signals. The recruitment procedure ending with the crisis can be explored through a spatial-temporal plot from which to extract suitable descriptors that are able to monitor and quantify the evolving synchronization level from the EEG tracings. In this paper, a spatial-temporal analysis of EEG recordings based on the concept of permutation entropy (PE) is proposed. The performance of PE are tested on a database of 24 patients affected by absence (generalized) seizures. The results achieved are compared to the dynamical behavior of the EEG of 40 healthy subjects. Being PE a feature which is dependent on two parameters, an extensive study of the sensitivity of the performance of PE with respect to the parameters' setting was carried out on scalp EEG. Once the optimal PE configuration was determined, its ability to detect the different brain states was evaluated. According to the results here presented, it seems that the widely accepted model of "jump" transition to absence seizure should be in some cases coupled (or substituted) by a gradual transition model characteristic of self-organizing networks. Indeed, it appears that the transition to the epileptic status is heralded before the preictal state, ever since the interictal stages. As a matter of fact, within the limits of the analyzed database, the frontal-temporal scalp areas appear constantly associated to PE levels higher compared to the remaining electrodes, whereas the parieto-occipital areas appear associated to lower PE values. The EEG of healthy subjects neither shows any similar dynamic behavior nor exhibits any recurrent portrait in PE topography.

In vivo detection of cortical optical changes associated with seizure activity with optical coherence tomography

The most common technology for seizure detection is with electroencephalography (EEG), which has low spatial resolution and minimal depth discrimination. Optical techniques using near-infrared (NIR) light have been used to improve upon EEG technology and previous research has suggested that optical changes, specifically changes in near-infrared optical scattering, may precede EEG seizure onset in in vivo models. Optical coherence tomography (OCT) is a high resolution, minimally invasive imaging technique, which can produce depth resolved cross-sectional images. In this study, OCT was used to detect changes in optical properties of cortical tissue in vivo in mice before and during the induction of generalized seizure activity. We demonstrated that a significant decrease (P < 0.001) in backscattered intensity during seizure progression can be detected before the onset of observable manifestations of generalized (stage-5) seizures. These results indicate the feasibility of minimally-invasive optical detection of seizures with OCT.

Epileptic seizures from abnormal networks: why some seizures defy predictability

Seizure prediction has proven to be difficult in clinically realistic environments. Is it possible that fluctuations in cortical firing could influence the onset of seizures in an ictal zone? To test this, we have now used neural network simulations in a computational model of cortex having a total of 65,536 neurons with intercellular wiring patterned after histological data. A spatially distributed Poisson driven background input representing the activity of neighboring cortex affected 1% of the neurons. Gamma distributions were fit to the interbursting phase intervals, a non-parametric test for randomness was applied, and a dynamical systems analysis was performed to search for period-1 orbits in the intervals. The non-parametric analysis suggests that intervals are being drawn at random from their underlying joint distribution and the dynamical systems analysis is consistent with a nondeterministic dynamical interpretation of the generation of bursting phases. These results imply that in a region of cortex with abnormal connectivity analogous to a seizure focus, it is possible to initiate seizure activity with fluctuations of input from the surrounding cortical regions. These findings suggest one possibility for ictal generation from abnormal focal epileptic networks. This mechanism additionally could help explain the difficulty in predicting partial seizures in some patients.

The role of high-quality EEG databases in the improvement and assessment of seizure prediction methods

Initially, seizure prediction was based on the analysis of brief EEG segments preceding clinically manifest seizures. Whereas such approaches suggested that the sensitivities of various EEG-derived features in predicting seizures were high, the inclusion of longer interictal periods and the combined assessment of sensitivity and specificity and the application of statistical validation methods have put into question the validity of such claims. We here show that the duration of EEG on which analyses are based and the number of seizures assessed negatively correlate with the reported sensitivities of prediction studies. Methodological aspects of seizure prediction are discussed in the framework of currently existing databases and of the newly established European Union database. This article is part of a Supplemental Special Issue entitled The Future of Automated Seizure Detection and Prediction.

Seizure probability in animal models of acquired epilepsy: a perspective on the concept of the preictal state

The concept of a preictal state is based on the belief that it may be possible to predict seizures before they occur. The preictal state is viewed as a time period when a seizure is practically inevitable, or at least a period of greatly increased seizure probability. Changes in seizure frequency may provide insight into how seizure probability increases after brain injury. Here, time-dependent changes in the frequency of spontaneous recurrent seizures after brain injury are summarized from published, nearly continuous, electrographic (EEG) recordings of kainate-treated rats and neonatal rats subjected to hypoxia-ischemia. For these animal models, seizure frequency - and thus seizure probability - was a sigmoid function of time after the brain injury. This observation differs from the traditional view, where the development of epilepsy after brain injury is a step-function of time, and the latent period is the time between a brain injury and the first spontaneous seizure. Based on backward extrapolation of the plots of seizure frequency versus time, these data suggest that seizure probability increases continuously during the latent period. Also, spontaneous recurrent seizures frequently occurred in clusters, suggesting that the intra-cluster seizure intervals are periods of high seizure probability. Thus, seizure probability progressively increases as a function of time after an epileptogenic brain injury, and is particularly high between seizures within a cluster, as compared to the time between clusters. These data suggest that the detectors of the preictal state need to be accurate (and tested) over a very wide range of seizure probabilities, and that studies on the physiological events that occur during seizure clusters may provide insight on the properties of the preictal state.