Identification of preseizure States in epilepsy: a data-driven approach for multichannel EEG recordings

Time-series-analysis-based detection of critical transitions in real-world non-autonomous systems

Real-world non-autonomous systems are open, out-of-equilibrium systems that evolve in and are driven by temporally varying environments. Such systems can show multiple timescale and transient dynamics together with transitions to very different and, at times, even disastrous dynamical regimes. Since such critical transitions disrupt the systems' intended or desired functionality, it is crucial to understand the underlying mechanisms, to identify precursors of such transitions, and to reliably detect them in time series of suitable system observables to enable forecasts. This review critically assesses the various steps of investigation involved in time-series-analysis-based detection of critical transitions in real-world non-autonomous systems: from the data recording to evaluating the reliability of offline and online detections. It will highlight pros and cons to stimulate further developments, which would be necessary to advance understanding and forecasting nonlinear behavior such as critical transitions in complex systems.

Unsupervised EEG preictal interval identification in patients with drug-resistant epilepsy

Typical seizure prediction models aim at discriminating interictal brain activity from pre-seizure electrographic patterns. Given the lack of a preictal clinical definition, a fixed interval is widely used to develop these models. Recent studies reporting preictal interval selection among a range of fixed intervals show inter- and intra-patient preictal interval variability, possibly reflecting the heterogeneity of the seizure generation process. Obtaining accurate labels of the preictal interval can be used to train supervised prediction models and, hence, avoid setting a fixed preictal interval for all seizures within the same patient. Unsupervised learning methods hold great promise for exploring preictal alterations on a seizure-specific scale. Multivariate and univariate linear and nonlinear features were extracted from scalp electroencephalography (EEG) signals collected from 41 patients with drug-resistant epilepsy undergoing presurgical monitoring. Nonlinear dimensionality reduction was performed for each group of features and each of the 226 seizures. We applied different clustering methods in searching for preictal clusters located until 2 h before the seizure onset. We identified preictal patterns in 90% of patients and 51% of the visually inspected seizures. The preictal clusters manifested a seizure-specific profile with varying duration (22.9 ± 21.0 min) and starting time before seizure onset (47.6 ± 27.3 min). Searching for preictal patterns on the EEG trace using unsupervised methods showed that it is possible to identify seizure-specific preictal signatures for some patients and some seizures within the same patient.

Analysis of Morbidity and Outcomes Associated With Use of Subdural Grids vs Stereoelectroencephalography in Patients With Intractable Epilepsy

Importance

A major change has occurred in the evaluation of epilepsy with the availability of robotic stereoelectroencephalography (SEEG) for seizure localization. However, the comparative morbidity and outcomes of this minimally invasive procedure relative to traditional subdural electrode (SDE) implantation are unknown.

Objective

To perform a comparative analysis of the relative efficacy, procedural morbidity, and epilepsy outcomes consequent to SEEG and SDE in similar patient populations and performed by a single surgeon at 1 center.

Design, Setting and Participants

Overall, 239 patients with medically intractable epilepsy underwent 260 consecutive intracranial electroencephalographic procedures to localize their epilepsy. Procedures were performed from November 1, 2004, through June 30, 2017, and data were analyzed in June 2017 and August 2018.

Interventions

Implantation of SDE using standard techniques vs SEEG using a stereotactic robot, followed by resection or laser ablation of the seizure focus.

Main Outcomes and Measures

Length of surgical procedure, surgical complications, opiate use, and seizure outcomes using the Engel Epilepsy Surgery Outcome Scale.

Results

Of the 260 cases included in the study (54.6% female; mean [SD] age at evaluation, 30.3 [13.1] years), the SEEG (n = 121) and SDE (n = 139) groups were similar in age (mean [SD], 30.1 [12.2] vs 30.6 [13.8] years), sex (47.1% vs 43.9% male), numbers of failed anticonvulsants (mean [SD], 5.7 [2.5] vs 5.6 [2.5]), and duration of epilepsy (mean [SD], 16.4 [12.0] vs17.2 [12.1] years). A much greater proportion of SDE vs SEEG cases were lesional (99 [71.2%] vs 53 [43.8%]; P < .001). Seven symptomatic hemorrhagic sequelae (1 with permanent neurological deficit) and 3 infections occurred in the SDE cohort with no clinically relevant complications in the SEEG cohort, a marked difference in complication rates (P = .003). A greater proportion of SDE cases resulted in resection or ablation compared with SEEG cases (127 [91.4%] vs 90 [74.4%]; P < .001). Favorable epilepsy outcomes (Engel class I [free of disabling seizures] or II [rare disabling seizures]) were observed in 57 of 75 SEEG cases (76.0%) and 59 of 108 SDE cases (54.6%; P = .003) amongst patients undergoing resection or ablation, at 1 year. An analysis of only nonlesional cases revealed good outcomes in 27 of 39 cases (69.2%) vs 9 of 26 cases (34.6%) at 12 months in SEEG and SDE cohorts, respectively (P = .006). When considering all patients undergoing evaluation, not just those undergoing definitive procedures, favorable outcomes (Engel class I or II) for SEEG compared with SDE were similar (57 of 121 [47.1%] vs 59 of 139 [42.4%] at 1 year; P = .45).

Conclusions and Relevance

This direct comparison of large matched cohorts undergoing SEEG and SDE implantation reveals distinctly better procedural morbidity favoring SEEG. These modalities intrinsically evaluate somewhat different populations, with SEEG being more versatile and applicable to a range of scenarios, including nonlesional and bilateral cases, than SDE. The significantly favorable adverse effect profile of SEEG should factor into decision making when patients with pharmacoresistant epilepsy are considered for intracranial evaluations.

Precursors of seizures due to specific spatial-temporal modifications of evolving large-scale epileptic brain networks

Knowing when, where, and how seizures are initiated in large-scale epileptic brain networks remains a widely unsolved problem. Seizure precursors – changes in brain dynamics predictive of an impending seizure – can now be identified well ahead of clinical manifestations, but either the seizure onset zone or remote brain areas are reported as network nodes from which seizure precursors emerge. We aimed to shed more light on the role of constituents of evolving epileptic networks that recurrently transit into and out of seizures. We constructed such networks from more than 3200 hours of continuous intracranial electroencephalograms recorded in 38 patients with medication refractory epilepsy. We succeeded in singling out predictive edges and predictive nodes. Their particular characteristics, namely edge weight respectively node centrality (a fundamental concept of network theory), from the pre-ictal periods of 78 out of 97 seizures differed significantly from the characteristics seen during inter-ictal periods. The vast majority of predictive nodes were connected by most of the predictive edges, but these nodes never played a central role in the evolving epileptic networks. Interestingly, predictive nodes were entirely associated with brain regions deemed unaffected by the focal epileptic process. We propose a network mechanism for a transition into the pre-seizure state, which puts into perspective the role of the seizure onset zone in this transition and highlights the necessity to reassess current concepts for seizure generation and seizure prevention.

Closed-Loop Brain Stimulation for Drug-Resistant Epilepsy: Towards an Evidence-Based Approach to Personalized Medicine

Closed-loop brain stimulation is one of the few treatments available for patients who are ineligible for traditional surgical resection of the epileptogenic zone, due to having generalized epilepsy, multifocal epilepsy, or focal epilepsy localized to an eloquent brain region. Due to its clinical efficacy and potential to delivery personalized therapy based on an individual's own intracerebral electrophysiology, this treatment is becoming an important part of clinical practice, despite a limited understanding of how to program detection and stimulation parameters for optimal, patient-specific benefit. To bring this challenge into focus, we review the evolution of neural stimulation for epilepsy, provide a technical overview of the RNS System (the only FDA-approved closed-loop device), and discuss the major challenges of working with a closed-loop device. We then propose an evidence-based solution for individualizing therapy that is driven by a bottom-up informatics approach.

Weighted and directed interactions in evolving large-scale epileptic brain networks

Epilepsy can be regarded as a network phenomenon with functionally and/or structurally aberrant connections in the brain. Over the past years, concepts and methods from network theory substantially contributed to improve the characterization of structure and function of these epileptic networks and thus to advance understanding of the dynamical disease epilepsy. We extend this promising line of research and assess-with high spatial and temporal resolution and using complementary analysis approaches that capture different characteristics of the complex dynamics-both strength and direction of interactions in evolving large-scale epileptic brain networks of 35 patients that suffered from drug-resistant focal seizures with different anatomical onset locations. Despite this heterogeneity, we find that even during the seizure-free interval the seizure onset zone is a brain region that, when averaged over time, exerts strongest directed influences over other brain regions being part of a large-scale network. This crucial role, however, manifested by averaging on the population-sample level only - in more than one third of patients, strongest directed interactions can be observed between brain regions far off the seizure onset zone. This may guide new developments for individualized diagnosis, treatment and control.

Predictability of uncontrollable multifocal seizures - towards new treatment options

Drug-resistant, multifocal, non-resectable epilepsies are among the most difficult epileptic disorders to manage. An approach to control previously uncontrollable seizures in epilepsy patients would consist of identifying seizure precursors in critical brain areas combined with delivering a counteracting influence to prevent seizure generation. Predictability of seizures with acceptable levels of sensitivity and specificity, even in an ambulatory setting, has been repeatedly shown, however, in patients with a single seizure focus only. We did a study to assess feasibility of state-of-the-art, electroencephalogram-based seizure-prediction techniques in patients with uncontrollable multifocal seizures. We obtained significant predictive information about upcoming seizures in more than two thirds of patients. Unexpectedly, the emergence of seizure precursors was confined to non-affected brain areas. Our findings clearly indicate that epileptic networks, spanning lobes and hemispheres, underlie generation of seizures. Our proof-of-concept study is an important milestone towards new therapeutic strategies based on seizure-prediction techniques for clinical practice.

Optogenetic tools for modulating and probing the epileptic network

Epilepsy affects roughly 1% of the population worldwide. Although effective treatments with antiepileptic drugs are available, more than 20% of patients have seizures that are refractory to medical therapy and many patients experience adverse effects. Hence, there is a continued need for novel therapies for those patients. A new technique called "optogenetics" may offer a new hope for these refractory patients. Optogenetics is a technology based on the combination of optics and genetics, which can control or record neural activity with light. Following delivery of light-sensitive opsin genes such as channelrhodopsin-2 (ChR2), halorhodopsin (NpHR), and others into brain, excitation or inhibition of specific neurons in precise brain areas can be controlled by illumination at different wavelengths with very high temporal and spatial resolution. Neuromodulation with the optogenetics toolbox have already been shown to be effective at treating seizures in animal models of epilepsy. This review will outline the most recent advances in epilepsy research with optogenetic techniques and discuss how this technology can contribute to our understanding and treatment of epilepsy in the future.

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.

Is ictal cognitive dysfunction predictable?

OBJECTIVE

Accurate prediction of electrographic seizure onset may reduce injuries and improve quality of life in pharmaco-resistant epileptics. However, because sub-clinical, far out-number clinical seizures, indiscriminate issuance of warnings may have a paralyzing effect on these patients. This study investigates the predictability of ictal cognitive dysfunction.

METHODS

Latency and percentage of correct responses to a reaction time test triggered by automated seizure detections were compared to those obtained inter-ictally in 14 subjects undergoing surgery evaluation. Since accurate prediction of seizures is elusive, early detection was used, as it indirectly but reliably investigates for the existence of a cognitive pre-ictal state.

RESULTS

Significant differences between ictal and inter-ictal cognitive performance were not uncovered until late into the temporal evolution of "focal" seizures.

CONCLUSIONS

These observations suggest that cognitive dysfunction is unpredictable in seizures originating from discrete cortical regions, as the transition into unawareness seems abrupt.

SIGNIFICANCE

Prediction of electrographic seizure onsets with worthwhile accuracy would likely result in large numbers of daily warnings, the great majority for sub-clinical seizures. This outcome would considerably increase, without safety justification, patients' psychological burden inherent to each forecast, thus further diminishing quality of life.

Imaging preictal hemodynamic changes in neocortical epilepsy

OBJECT

The ability to predict seizure occurrence is extremely important to trigger abortive therapies and to warn patients and their caregivers. Optical imaging of hemodynamic parameters such as blood flow, blood volume, and tissue and hemoglobin oxygenation has already been shown to successfully localize epileptic events with high spatial and temporal resolution. The ability to actually predict seizure occurrence using hemodynamic parameters is less well explored.

METHODS

In this article, the authors critically review data from the literature on neocortical epilepsy and optical imaging, and they discuss the preictal hemodynamic changes and their application in neurosurgery.

RESULTS

Recent optical mapping studies have demonstrated preictal hemodynamic changes in both human and animal neocortex.

CONCLUSIONS

Optical measurements of blood flow and oxygenation may become increasingly important for predicting and localizing epileptic events. The ability to successfully predict ictal onsets may be useful to trigger closed-loop abortive therapies.

Anticipating the unobserved: prediction of subclinical seizures

Subclinical seizures (SCS) have rarely been considered in the diagnosis and therapy of epilepsy and have not been systematically analyzed in studies on seizure prediction. Here, we investigate whether predictions of subclinical seizures are feasible and how their occurrence may affect the performance of prediction algorithms. Using the European database of long-term recordings of surface and invasive electroencephalography data, we analyzed the data from 21 patients with SCS, including in total 413 clinically manifest seizures (CS) and 3341 SCS. Based on the mean phase coherence we investigated the predictive performance of CS and SCS. The two types of seizures had similar prediction sensitivities. Significant performance was found considerably more often for SCS than for CS, especially for patients with invasive recordings. When analyzing false alarms triggered by predicting CS, a significant number of these false predictions were followed by SCS for 9 of 21 patients. Although currently observed prediction performance may not be deemed sufficient for clinical applications for the majority of the patients, it can be concluded that the prediction of SCS is feasible on a similar level as for CS and allows a prediction of more of the seizures impairing patients, possibly also reducing the number of false alarms that were in fact correct predictions of CS. This article is part of a Supplemental Special Issue entitled The Future of Automated Seizure Detection and Prediction.