What are the assets and weaknesses of HFO detectors? A benchmark framework based on realistic simulations

Thalamic stereoelectroencephalography for neuromodulation target selection: Proof of concept and review of literature of pulvinar direct electrical stimulation

In patients with drug-resistant epilepsy (DRE) who are not candidates for resective surgery, various thalamic nuclei, including the anterior, centromedian, and pulvinar nuclei, have been extensively investigated as targets for neuromodulation. However, the therapeutic effects of different targets for thalamic neuromodulation on various types of epilepsy are not well understood. Here, we present a 32-year-old patient with multifocal bilateral temporoparieto-occipital epilepsy and bilateral malformations of cortical development (MCDs) who underwent bilateral stereoelectroencephalographic (SEEG) recordings of the aforementioned three thalamic nuclei bilaterally. The change in the rate of interictal epileptiform discharges (IEDs) from baseline were compared in temporal, central, parietal, and occipital regions after direct electrical stimulation (DES) of each thalamic nucleus. A significant decrease in the rate of IEDs (33% from baseline) in the posterior quadrant regions was noted in the ipsilateral as well as contralateral hemisphere following DES of the pulvinar. A scoping review was also performed to better understand the current standpoint of pulvinar thalamic stimulation in the treatment of DRE. The therapeutic effect of neuromodulation can differ among thalamic nuclei targets and epileptogenic zones (EZs). In patients with multifocal EZs with extensive MCDs, personalized thalamic targeting could be achieved through DES with thalamic SEEG electrodes.

Practical measurements distinguishing physiological and pathological stereoelectroencephalography channels based on high-frequency oscillations in the human brain

OBJECTIVE

The present study aimed to identify various distinguishing features for use in the accurate classification of stereoelectroencephalography (SEEG) channels based on high-frequency oscillations (HFOs) inside and outside the epileptogenic zone (EZ).

METHODS

HFOs were detected in patients with focal epilepsy who underwent SEEG. Subsequently, HFOs within the seizure-onset and early spread zones were defined as pathological HFOs, whereas others were defined as physiological. Three features of HFOs were identified at the channel level, namely, morphological repetition, rhythmicity, and phase-amplitude coupling (PAC). A machine-learning (ML) classifier was then built to distinguish two HFO types at the channel level by application of the above-mentioned features, and the contributions were quantified. Further verification of the characteristics and classifier performance was performed in relation to various conscious states, imaging results, EZ location, and surgical outcomes.

RESULTS

Thirty-five patients were included in this study, from whom 166  104 pathological HFOs in 255 channels and 53 374 physiological HFOs in 282 channels were entered into the analysis pipeline. The results revealed that the morphological repetitions of pathological HFOs were markedly higher than those of the physiological HFOs; this was also observed for rhythmicity and PAC. The classifier exhibited high accuracy in differentiating between the two forms of HFOs, as indicated by an area under the curve (AUC) of 0.89. Both PAC and rhythmicity contributed significantly to this distinction. The subgroup analyses supported these findings.

SIGNIFICANCE

The suggested HFO features can accurately distinguish between pathological and physiological channels substantially improving its usefulness in clinical localization.

PLAIN LANGUAGE SUMMARY

In this study, we computed three quantitative features associated with HFOs in each SEEG channel and then constructed a machine learning-based classifier for the classification of pathological and physiological channels. The classifier performed well in distinguishing the two channel types under different levels of consciousness as well as in terms of imaging results, EZ location, and patient surgical outcomes.

Recent advances in clinical electroencephalography

PURPOSE OF REVIEW

Clinical electroencephalography (EEG) is a conservative medical field. This explains likely the significant gap between clinical practice and new research developments. This narrative review discusses possible causes of this discrepancy and how to circumvent them. More specifically, we summarize recent advances in three applications of clinical EEG: source imaging (ESI), high-frequency oscillations (HFOs) and EEG in critically ill patients.

RECENT FINDINGS

Recently published studies on ESI provide further evidence for the accuracy and clinical utility of this method in the multimodal presurgical evaluation of patients with drug-resistant focal epilepsy, and opened new possibilities for further improvement of the accuracy. HFOs have received much attention as a novel biomarker in epilepsy. However, recent studies questioned their clinical utility at the level of individual patients. We discuss the impediments, show up possible solutions and highlight the perspectives of future research in this field. EEG in the ICU has been one of the major driving forces in the development of clinical EEG. We review the achievements and the limitations in this field.

SUMMARY

This review will promote clinical implementation of recent advances in EEG, in the fields of ESI, HFOs and EEG in the intensive care.

Epileptic tissue localization using graph-based networks in the high frequency oscillation range of intracranial electroencephalography

PURPOSE

High frequency oscillations (HFOs) are an emerging biomarker of epilepsy. However, very few studies have investigated the functional connectivity of interictal iEEG signals in the frequency range of HFOs. Here, we study the corresponding functional networks using graph theory, and we assess their predictive value for automatic electrode classification in a cohort of 20 drug resistant patients.

METHODS

Coherence-based connectivity analysis was performed on the iEEG recordings, and six different local graph measures were computed in both sub-bands of the HFO frequency range (80-250 Hz and 250-500 Hz). Correlation analysis was implemented between the local graph measures and the ripple and fast ripple rates. Finally, the WEKA software was employed for training and testing different predictive models on the aforementioned local graph measures.

RESULTS

The ripple rate was significantly correlated with five out of six local graph measures in the functional network. For fast ripples, their rate was also significantly (but negatively) correlated with most of the local metrics. The results from WEKA showed that the Logistic Regression algorithm was able to classify highly HFO-contaminated electrodes with an accuracy of 82.5 % for ripples and 75.4 % for fast ripples.

CONCLUSION

Functional connectivity networks in the HFO band could represent an alternative to the direct use of distinct HFO events, while also providing important insights about hub epileptic areas that can represent possible surgical targets. Automatic electrode classification through FC-based classifiers can help bypass the burden of manual HFO annotation, providing at the same time similar amount of information about the epileptic tissue.

Tuning Microelectrodes' Impedance to Improve Fast Ripples Recording

Epilepsy is a chronic neurological disorder characterized by recurrent seizures resulting from abnormal neuronal hyperexcitability. In the case of pharmacoresistant epilepsy requiring resection surgery, the identification of the Epileptogenic Zone (EZ) is critical. Fast Ripples (FRs; 200-600 Hz) are one of the promising biomarkers that can aid in EZ delineation. However, recording FRs requires physically small electrodes. These microelectrodes suffer from high impedance, which significantly impacts FRs' observability and detection. In this study, we investigated the potential of a conductive polymer coating to enhance FR observability. We employed biophysical modeling to compare two types of microelectrodes: Gold (Au) and Au coated with the conductive polymer poly(3,4-ethylenedioxythiophene)-poly(styrene sulfonate) (Au/PEDOT:PSS). These electrodes were then implanted into the CA1 hippocampal neural network of epileptic mice to record FRs during epileptogenesis. The results showed that the polymer-coated electrodes had a two-order lower impedance as well as a higher transfer function amplitude and cut-off frequency. Consequently, FRs recorded with the PEDOT:PSS-coated microelectrode yielded significantly higher signal energy compared to the uncoated one. The PEDOT:PSS coating improved the observability of the recorded FRs and thus their detection. This work paves the way for the development of signal-specific microelectrode designs that allow for better targeting of pathological biomarkers.

High-rate leading spikes in propagating spike sequences predict seizure outcome in surgical patients with temporal lobe epilepsy

Inter-ictal spikes aid in the diagnosis of epilepsy and in planning surgery of medication-resistant epilepsy. However, the localizing information from spikes can be unreliable because spikes can propagate, and the burden of spikes, often assessed as a rate, does not always correlate with the seizure onset zone or seizure outcome. Recent work indicates identifying where spikes regularly emerge and spread could localize the seizure network. Thus, the current study sought to better understand where and how rates of single and coupled spikes, and especially brain regions with high-rate and leading spike of a propagating sequence, informs the extent of the seizure network. In 37 patients with medication-resistant temporal lobe seizures, who had surgery to treat their seizure disorder, an algorithm detected spikes in the pre-surgical depth inter-ictal EEG. A separate algorithm detected spike propagation sequences and identified the location of leading and downstream spikes in each sequence. We analysed the rate and power of single spikes on each electrode and coupled spikes between pairs of electrodes, and the proportion of sites with high-rate, leading spikes in relation to the seizure onset zone of patients seizure free (n = 19) and those with continuing seizures (n = 18). We found increased rates of single spikes in mesial temporal seizure onset zone (ANOVA, P < 0.001, η2 = 0.138), and increased rates of coupled spikes within, but not between, mesial-, lateral- and extra-temporal seizure onset zone of patients with continuing seizures (P < 0.001; η2 = 0.195, 0.113 and 0.102, respectively). In these same patients, there was a higher proportion of brain regions with high-rate leaders, and each sequence contained a greater number of spikes that propagated with a higher efficiency over a longer distance outside the seizure onset zone than patients seizure free (Wilcoxon, P = 0.0172). The proportion of high-rate leaders in and outside the seizure onset zone could predict seizure outcome with area under curve = 0.699, but not rates of single or coupled spikes (0.514 and 0.566). Rates of coupled spikes to a greater extent than single spikes localize the seizure onset zone and provide evidence for inter-ictal functional segregation, which could be an adaptation to avert seizures. Spike rates, however, have little value in predicting seizure outcome. High-rate spike sites leading propagation could represent sources of spikes that are important components of an efficient seizure network beyond the clinical seizure onset zone, and like the seizure onset zone these, too, need to be removed, disconnected or stimulated to increase the likelihood for seizure control.

Intrinsic and secondary epileptogenicity in focal cortical dysplasia type II

OBJECTIVE

Favorable seizure outcome is reported following resection of bottom-of-sulcus dysplasia (BOSD). We assessed the distribution of epileptogenicity and dysplasia in and around BOSD to better understand this clinical outcome and the optimal surgical approach.

METHODS

We studied 27 children and adolescents with magnetic resonance imaging (MRI)-positive BOSD who underwent epilepsy surgery; 85% became seizure-free postresection (median = 5.0 years follow-up). All patients had resection of the dysplastic sulcus, and 11 had additional resection of the gyral crown (GC) or adjacent gyri (AG). Markers of epileptogenicity were relative cortical hypometabolism on preoperative 18 F-fluorodeoxyglucose (FDG) positron emission tomography (PET), and spiking, ripples, fast ripples, spike-high-frequency oscillation cross-rate, and phase amplitude coupling (PAC) on preresection and postresection electrocorticography (ECoG), all analyzed at the bottom-of-sulcus (BOS), top-of-sulcus (TOS), GC, and AG. Markers of dysplasia were increased cortical thickness on preoperative MRI, and dysmorphic neuron density and variant allele frequency of somatic MTOR mutations in resected tissue, analyzed at similar locations.

RESULTS

Relative cortical metabolism was significantly reduced and ECoG markers were significantly increased at the BOS compared to other regions. Apart from spiking and PAC, which were greater at the TOS compared to the GC, there were no significant differences in PET and other ECoG markers between the TOS, GC, and AG, suggesting a cutoff of epileptogenicity at the TOS rather than a tapering gradient on the cortical surface. MRI and tissue markers of dysplasia were all maximal in the BOS, reduced in the TOS, and mostly absent in the GC. Spiking and PAC reduced significantly over the GC after resection of the dysplastic sulcus.

SIGNIFICANCE

These findings support the concept that dysplasia and intrinsic epileptogenicity are mostly limited to the dysplastic sulcus in BOSD and support resection or ablation confined to the MRI-visible lesion as a first-line surgical approach. 18 F-FDG PET and ECoG abnormalities in surrounding cortex seem to be secondary phenomena.

Variability of Single Pulse Electrical Stimulation Responses Recorded with Intracranial Electroencephalography in Epileptic Patients

Cohort studies of brain stimulations performed with stereo-electroencephalographic (SEEG) electrodes in epileptic patients allow to derive large scale functional connectivity. It is known, however, that brain responses to electrical or magnetic stimulation techniques are not always reproducible. Here, we study variability of responses to single pulse SEEG electrical stimulation. We introduce a second-order probability analysis, i.e. we extend estimation of connection probabilities, defined as the proportion of responses trespassing a statistical threshold (determined in terms of Z-score with respect to spontaneous neuronal activity before stimulation) over all responses and derived from a number of individual measurements, to an analysis of pairs of measurements.Data from 445 patients were processed. We found that variability between two equivalent measurements is substantial in particular conditions. For long ( > ~ 90 mm) distances between stimulating and recording sites, and threshold value Z = 3, correlation between measurements drops almost to zero. In general, it remains below 0.5 when the threshold is smaller than Z = 4 or the stimulating current intensity is 1 mA. It grows with an increase of either of these factors. Variability is independent of interictal spiking rates in the stimulating and recording sites.We conclude that responses to SEEG stimulation in the human brain are variable, i.e. in a subject at rest, two stimulation trains performed at the same electrode contacts and with the same protocol can give discrepant results. Our findings highlight an advantage of probabilistic interpretation of such results even in the context of a single individual.

Comparison of beamformer and ICA for dynamic connectivity analysis: A simultaneous MEG-SEEG study

Magnetoencephalography (MEG) is a powerful tool for estimating brain connectivity with both good spatial and temporal resolution. It is particularly helpful in epilepsy to characterize non-invasively the epileptic networks. However, using MEG to map brain networks requires solving a difficult inverse problem that introduces uncertainty in the activity localization and connectivity measures. Our goal here was to compare independent component analysis (ICA) followed by dipole source localization and the linearly constrained minimum-variance beamformer (LCMV-BF) for characterizing regions with interictal epileptic activity and their dynamic connectivity. After a simulation study, we compared ICA and LCMV-BF results with intracerebral EEG (stereotaxic EEG, SEEG) recorded simultaneously in 8 epileptic patients, which provide a unique 'ground truth' to which non-invasive results can be confronted. We compared the signal time courses extracted applying ICA and LCMV-BF on MEG data to that of SEEG, both for the actual signals and the dynamic connectivity computed using cross-correlation (evolution of links in time). With our simulations, we illustrated the different effect of the temporal and spatial correlation among sources on the two methods. While ICA was more affected by the temporal correlation but robust against spatial configurations, LCMV-BF showed opposite behavior. Moreover, ICA seems more suited to retrieve the simulated networks. In case of real patient data, good MEG/SEEG correlation and good localization were obtained in 6 out of 8 patients. In 4 of them ICA had the best performance (higher correlation, lower localization distance). In terms of dynamic connectivity, the evolution in time of the cross-correlation links could be retrieved in 5 patients out of 6, however, with more variable results in terms of correlation and distance. In two patients LCMV-BF had better results than ICA. In one patient the two methods showed equally good outcomes, and in the remaining two patients ICA performed best. In conclusion, our results obtained by exploiting simultaneous MEG/SEEG recordings suggest that ICA and LCMV-BF have complementary qualities for retrieving the dynamics of interictal sources and their network interactions.

EEG biomarker candidates for the identification of epilepsy

Electroencephalography (EEG) is one of the main pillars used for the diagnosis and study of epilepsy, readily employed after a possible first seizure has occurred. The most established biomarker of epilepsy, in case seizures are not recorded, are interictal epileptiform discharges (IEDs). In clinical practice, however, IEDs are not always present and the EEG may appear completely normal despite an underlying epileptic disorder, often leading to difficulties in the diagnosis of the disease. Thus, finding other biomarkers that reliably predict whether an individual suffers from epilepsy even in the absence of evident epileptic activity would be extremely helpful, since they could allow shortening the period of diagnostic uncertainty and consequently decreasing the risk of seizure. To date only a few EEG features other than IEDs seem to be promising candidates able to distinguish between epilepsy, i.e. > 60 % risk of recurrent seizures, or other (pathological) conditions. The aim of this narrative review is to provide an overview of the EEG-based biomarker candidates for epilepsy and the techniques employed for their identification.

Source-sink connectivity: a novel interictal EEG marker for seizure localization

Over 15 million epilepsy patients worldwide have drug-resistant epilepsy. Successful surgery is a standard of care treatment but can only be achieved through complete resection or disconnection of the epileptogenic zone, the brain region(s) where seizures originate. Surgical success rates vary between 20% and 80%, because no clinically validated biological markers of the epileptogenic zone exist. Localizing the epileptogenic zone is a costly and time-consuming process, which often requires days to weeks of intracranial EEG (iEEG) monitoring. Clinicians visually inspect iEEG data to identify abnormal activity on individual channels occurring immediately before seizures or spikes that occur interictally (i.e. between seizures). In the end, the clinical standard mainly relies on a small proportion of the iEEG data captured to assist in epileptogenic zone localization (minutes of seizure data versus days of recordings), missing opportunities to leverage these largely ignored interictal data to better diagnose and treat patients. IEEG offers a unique opportunity to observe epileptic cortical network dynamics but waiting for seizures increases patient risks associated with invasive monitoring. In this study, we aimed to leverage interictal iEEG data by developing a new network-based interictal iEEG marker of the epileptogenic zone. We hypothesized that when a patient is not clinically seizing, it is because the epileptogenic zone is inhibited by other regions. We developed an algorithm that identifies two groups of nodes from the interictal iEEG network: those that are continuously inhibiting a set of neighbouring nodes ('sources') and the inhibited nodes themselves ('sinks'). Specifically, patient-specific dynamical network models were estimated from minutes of iEEG and their connectivity properties revealed top sources and sinks in the network, with each node being quantified by source-sink metrics. We validated the algorithm in a retrospective analysis of 65 patients. The source-sink metrics identified epileptogenic regions with 73% accuracy and clinicians agreed with the algorithm in 93% of seizure-free patients. The algorithm was further validated by using the metrics of the annotated epileptogenic zone to predict surgical outcomes. The source-sink metrics predicted outcomes with an accuracy of 79% compared to an accuracy of 43% for clinicians' predictions (surgical success rate of this dataset). In failed outcomes, we identified brain regions with high metrics that were untreated. When compared with high frequency oscillations, the most commonly proposed interictal iEEG feature for epileptogenic zone localization, source-sink metrics outperformed in predictive power (by a factor of 1.2), suggesting they may be an interictal iEEG fingerprint of the epileptogenic zone.

A multi-head self-attention deep learning approach for detection and recommendation of neuromagnetic high frequency oscillations in epilepsy

Magnetoencephalography is a noninvasive neuromagnetic technology to record epileptic activities for the pre-operative localization of epileptogenic zones, which has received increasing attention in the diagnosis and surgery of epilepsy. As reported by recent studies, pathological high frequency oscillations (HFOs), when utilized as a biomarker to localize the epileptogenic zones, result in a significant reduction in seizure frequency, even seizure elimination in around 80% of cases. Thus, objective, rapid, and automatic detection and recommendation of HFOs are highly desirable for clinicians to alleviate the burden of reviewing a large amount of MEG data from a given patient. Despite the advantage, the performance of existing HFOs rarely satisfies the clinical requirement. Consequently, no HFOs have been successfully applied to real clinical applications so far. In this work, we propose a multi-head self-attention-based detector for recommendation, termed MSADR, to detect and recommend HFO signals. Taking advantage of the state-of-the-art multi-head self-attention mechanism in deep learning, the proposed MSADR achieves a more superior accuracy of 88.6% than peer machine learning models in both detection and recommendation tasks. In addition, the robustness of MSADR is also extensively assessed with various ablation tests, results of which further demonstrate the effectiveness and generalizability of the proposed approach.

Startle-Induced Epileptic Spasms: A Clinical and Video-EEG Study

Objective

This study aimed to delineate the detailed characteristics of startle-induced epileptic spasms (ES) and explore the brain regions where startle-induced ES originated.

Methods

Among 581 patients with ES registered in our database, 30 were diagnosed with startle-induced ES according to video-electroencephalogram (EEG) and seizure semiology and were included in this study. Patients' clinical characteristics and ictal high-frequency oscillations (HFOs) were analyzed.

Results

Mean age at the onset of startle-induced ES was 28.1 months. Half of the patients had structural etiology, two of whom were diagnosed with co-existing structural and genetic etiologies. The focal neuroimaging abnormalities were predominant in the frontal cortex (9/15, 60.0%). Fifteen patients (50%) had prominent interictal epileptiform discharges in the frontal and anterior temporal. Ictal HFOs counts of the startle-induced ES in the anterior region were significantly higher than those in the posterior regions (p &lt; 0.05). Five patients (16.7%) became seizure-free ≥6 months, and ten (33.3%) showed startle-induced ES cessation ≥6 months. All patients except one had mild to severe psychomotor developmental delay after the onset of seizures.

Conclusion

Patients with startle-induced ES typically had brain lesions and showed drug-resistant. The neuroimaging and EEG findings, including ictal HFOs, support that startle-induced ES often originates from the frontal cortex.

Effect of Training on Visual Identification of High Frequency Oscillations-A Delphi-Style Intervention

Objective

We examined the effect of a simple Delphi-method feedback on visual identification of high frequency oscillations (HFOs) in the ripple (80–250 Hz) band, and assessed the impact of this training intervention on the interrater reliability and generalizability of HFO evaluations.

Methods

We employed a morphology detector to identify potential HFOs at two thresholds and presented them to visual reviewers to assess the probability of each epoch containing an HFO. We recruited 19 board-certified epileptologists with various levels of experience to complete a series of HFO evaluations during three sessions. A Delphi-style intervention was used to provide feedback on the performance of each reviewer relative to their peers. A delayed-intervention paradigm was used, in which reviewers received feedback either before or after the second session. ANOVAs were used to assess the effect of the intervention on the reviewers' evaluations. Generalizability theory was used to assess the interrater reliability before and after the intervention.

Results

The intervention, regardless of when it occurred, resulted in a significant reduction in the variability between reviewers in both groups (p_GroupDI = 0.037, p_GroupEI = 0.003). Prior to the delayed-intervention, the group receiving the early intervention showed a significant reduction in variability (p_GroupEI = 0.041), but the delayed-intervention group did not (p_GroupDI = 0.414). Following the intervention, the projected number of reviewers required to achieve strong generalizability decreased from 35 to 16.

Significance

This study shows a robust effect of a Delphi-style intervention on the interrater variability, reliability, and generalizability of HFO evaluations. The observed decreases in HFO marking discrepancies across 14 of the 15 reviewers are encouraging: they are necessarily associated with an increase in interrater reliability, and therefore with a corresponding decrease in the number of reviewers required to achieve strong generalizability. Indeed, the reliability of all reviewers following the intervention was similar to that of experienced reviewers prior to intervention. Therefore, a Delphi-style intervention could be implemented either to sufficiently train any reviewer, or to further refine the interrater reliability of experienced reviewers. In either case, a Delphi-style intervention would help facilitate the standardization of HFO evaluations and its implementation in clinical care.

BrainQuake: An Open-Source Python Toolbox for the Stereoelectroencephalography Spatiotemporal Analysis

Intracranial stereoelectroencephalography (SEEG) is broadly used in the presurgical evaluation of intractable epilepsy, due to its high temporal resolution in neural activity recording and high spatial resolution within suspected epileptogenic zones. Neurosurgeons or technicians face the challenge of conducting a workflow of post-processing operations with the multimodal data (e.g., MRI, CT, and EEG) after the implantation surgery, such as brain surface reconstruction, electrode contact localization, and SEEG data analysis. Several software or toolboxes have been developed to take one or more steps in the workflow but without an end-to-end solution. In this study, we introduced BrainQuake, an open-source Python software for the SEEG spatiotemporal analysis, integrating modules and pipelines in surface reconstruction, electrode localization, seizure onset zone (SOZ) prediction based on ictal and interictal SEEG analysis, and final visualizations, each of which is highly automated with a user-friendly graphical user interface (GUI). BrainQuake also supports remote communications with a public server, which is facilitated with automated and standardized preprocessing pipelines, high-performance computing power, and data curation management to provide a time-saving and compatible platform for neurosurgeons and researchers.

Interictal spikes with and without high-frequency oscillation have different single-neuron correlates

Interictal epileptiform discharges (IEDs) are a widely used biomarker in patients with epilepsy but lack specificity. It has been proposed that there are truly epileptogenic and less pathological or even protective IEDs. Recent studies suggest that highly pathological IEDs are characterized by high-frequency oscillations (HFOs). Here, we aimed to dissect these 'HFO-IEDs' at the single-neuron level, hypothesizing that the underlying mechanisms are distinct from 'non-HFO-IEDs'. Analysing hybrid depth electrode recordings from patients with temporal lobe epilepsy, we found that single-unit firing rates were higher in HFO- than in non-HFO-IEDs. HFO-IEDs were characterized by a pronounced pre-peak increase in firing, which coincided with the preferential occurrence of HFOs, whereas in non-HFO-IEDs, there was only a mild pre-peak increase followed by a post-peak suppression. Comparing each unit's firing during HFO-IEDs to its baseline activity, we found many neurons with a significant increase during the HFO component or ascending part, but almost none with a decrease. No such imbalance was observed during non-HFO-IEDs. Finally, comparing each unit's firing directly between HFO- and non-HFO-IEDs, we found that most cells had higher rates during HFO-IEDs and, moreover, identified a distinct subset of neurons with a significant preference for this IED subtype. In summary, our study reveals that HFO- and non-HFO-IEDs have different single-unit correlates. In HFO-IEDs, many neurons are moderately activated, and some participate selectively, suggesting that both types of increased firing contribute to highly pathological IEDs.

A brain atlas of axonal and synaptic delays based on modelling of cortico-cortical evoked potentials

Epilepsy presurgical investigation may include focal intracortical single-pulse electrical stimulations with depth electrodes, which induce cortico-cortical evoked potentials at distant sites because of white matter connectivity. Cortico-cortical evoked potentials provide a unique window on functional brain networks because they contain sufficient information to infer dynamical properties of large-scale brain connectivity, such as preferred directionality and propagation latencies.

 Here, we developed a biologically informed modelling approach to estimate the neural physiological parameters of brain functional networks from the cortico-cortical evoked potentials recorded in a large multicentric database. Specifically, we considered each cortico-cortical evoked potential as the output of a transient stimulus entering the stimulated region, which directly propagated to the recording region. Both regions were modelled as coupled neural mass models, the parameters of which were estimated from the first cortico-cortical evoked potential component, occurring before 80 ms, using dynamic causal modelling and Bayesian model inversion. This methodology was applied to the data of 780 patients with epilepsy from the F-TRACT database, providing a total of 34 354 bipolar stimulations and 774 445 cortico-cortical evoked potentials. The cortical mapping of the local excitatory and inhibitory synaptic time constants and of the axonal conduction delays between cortical regions was obtained at the population level using anatomy-based averaging procedures, based on the Lausanne2008 and the HCP-MMP1 parcellation schemes, containing 130 and 360 parcels, respectively. To rule out brain maturation effects, a separate analysis was performed for older (>15 years) and younger patients (<15 years). In the group of older subjects, we found that the cortico-cortical axonal conduction delays between parcels were globally short (median = 10.2 ms) and only 16% were larger than 20 ms. This was associated to a median velocity of 3.9 m/s. Although a general lengthening of these delays with the distance between the stimulating and recording contacts was observed across the cortex, some regions were less affected by this rule, such as the insula for which almost all efferent and afferent connections were faster than 10 ms. Synaptic time constants were found to be shorter in the sensorimotor, medial occipital and latero-temporal regions, than in other cortical areas. Finally, we found that axonal conduction delays were significantly larger in the group of subjects younger than 15 years, which corroborates that brain maturation increases the speed of brain dynamics.

To our knowledge, this study is the first to provide a local estimation of axonal conduction delays and synaptic time constants across the whole human cortex in vivo, based on intracerebral electrophysiological recordings.

Neuromagnetic high frequency spikes are a new and noninvasive biomarker for localization of epileptogenic zones

OBJECTIVE

One barrier hindering high frequency brain signals (HFBS, >80 Hz) from wide clinical applications is that the brain generates both pathological and physiological HFBS. This study was to find specific biomarkers for localizing epileptogenic zones (EZs).

METHODS

Twenty three children with drug-resistant epilepsy and age/sex matched healthy controls were studied with magnetoencephalography (MEG). High frequency oscillations (HFOs, > 4 oscillatory waveforms) and high frequency spikes (HFSs, > 1 spiky or sharp waveforms) in 80-250 Hz and 250-600 Hz bands were blindly detected with an artificial intelligence method and validated with visual inspection. The magnitude of HFOs and HFSs were quantified with spectral analyses. Sources of HFSs and HFOs were localized and compared with clinical EZs determined by invasive recordings and surgical outcomes.

RESULTS

HFOs in 80-250 Hz and 250-600 Hz were identified in both epilepsy patients (18/23, 12/23, respectively) and healthy controls (6/23, 4/23, respectively). HFSs in 80-250 Hz and 250-600 Hz were detected in patients (16/23, 11/23, respectively) but not in healthy controls. A combination of HFOs and HFSs localized EZs for 22 (22/23, 96%) patients.

CONCLUSIONS

The results indicate, for the first time, that HFSs are a newer and more specific biomarker than HFOs for localizing EZs because HFOs appeared in both epilepsy patients and healthy controls while HFSs appeared only in epilepsy patients.

Distinction of Physiologic and Epileptic Ripples: An Electrical Stimulation Study

Ripple oscillations (80-250 Hz) are a promising biomarker of epileptic activity, but are also involved in memory consolidation, which impairs their value as a diagnostic tool. Distinguishing physiologic from epileptic ripples has been particularly challenging because usually, invasive recordings are only performed in patients with refractory epilepsy. Here, we identified 'healthy' brain areas based on electrical stimulation and hypothesized that these regions specifically generate 'pure' ripples not coupled to spikes. Intracranial electroencephalography (EEG) recorded with subdural grid electrodes was retrospectively analyzed in 19 patients with drug-resistant focal epilepsy. Interictal spikes and ripples were automatically detected in slow-wave sleep using the publicly available Delphos software. We found that rates of spikes, ripples and ripples coupled to spikes ('spike-ripples') were higher inside the seizure-onset zone (p < 0.001). A comparison of receiver operating characteristic curves revealed that spike-ripples slightly delineated the seizure-onset zone channels, but did this significantly better than spikes (p < 0.001). Ripples were more frequent in the eloquent neocortex than in the remaining non-seizure onset zone areas (p < 0.001). This was due to the higher rates of 'pure' ripples (p < 0.001; median rates 3.3/min vs. 1.4/min), whereas spike-ripple rates were not significantly different (p = 0.87). 'Pure' ripples identified 'healthy' channels significantly better than chance (p < 0.001). Our findings suggest that, in contrast to epileptic spike-ripples, 'pure' ripples are mainly physiological. They may be considered, in addition to electrical stimulation, to delineate eloquent cortex in pre-surgical patients. Since we applied open source software for detection, our approach may be generally suited to tackle a variety of research questions in epilepsy and cognitive science.

Noninvasive high-frequency oscillations riding spikes delineates epileptogenic sources

Millions of people affected by epilepsy may undergo surgical resection of the epileptic tissues to stop seizures if such epileptic foci can be accurately delineated. High-frequency oscillations (HFOs), existing in electroencephalography, are highly correlated with epileptic brain, which is promising for guiding successful neurosurgery. However, it is unclear whether and how pathological HFOs can be differentiated to localize the epileptogenic tissues given the presence of various nonepileptic high-frequency activities. Here, we show morphological and source imaging evidence that pathological HFOs can be identified by the concurrence of epileptiform spikes. We describe a framework to delineate the underlying epileptogenicity using this biomarker. Our work may offer translational tools to improve treatments by noninvasively demarking pathological activities and hence epileptic foci.

High-frequency oscillations (HFOs) are a promising biomarker for localizing epileptogenic brain and guiding successful neurosurgery. However, the utility and translation of noninvasive HFOs, although highly desirable, is impeded by the difficulty in differentiating pathological HFOs from nonepileptiform high-frequency activities and localizing the epileptic tissue using noninvasive scalp recordings, which are typically contaminated with high noise levels. Here, we show that the consistent concurrence of HFOs with epileptiform spikes (pHFOs) provides a tractable means to identify pathological HFOs automatically, and this in turn demarks an epileptiform spike subgroup with higher epileptic relevance than the other spikes in a cohort of 25 temporal epilepsy patients (including a total of 2,967 interictal spikes and 1,477 HFO events). We found significant morphological distinctions of HFOs and spikes in the presence/absence of this concurrent status. We also demonstrated that the proposed pHFO source imaging enhanced localization of epileptogenic tissue by 162% (∼5.36 mm) for concordance with surgical resection and by 186% (∼12.48 mm) with seizure-onset zone determined by invasive studies, compared to conventional spike imaging, and demonstrated superior congruence with the surgical outcomes. Strikingly, the performance of spike imaging was selectively boosted by the presence of spikes with pHFOs, especially in patients with multitype spikes. Our findings suggest that concurrent HFOs and spikes reciprocally discriminate pathological activities, providing a translational tool for noninvasive presurgical diagnosis and postsurgical evaluation in vulnerable patients.

VEP atlas: An anatomic and functional human brain atlas dedicated to epilepsy patients

BACKGROUND

Several automated parcellation atlases of the human brain have been developed over the past decades, based on various criteria, and have been applied in basic and clinical research.

NEW METHOD

Here we present the Virtual Epileptic Patient (VEP) atlas that offers a new automated brain region parcellation and labeling, which has been developed for the specific use in the domains of epileptology and functional neurosurgery and is able to apply at individual patient's level.

RESULTS

It comprises 162 brain regions, including 73 cortical and 8 subcortical regions per hemisphere. We demonstrate the successful application of the VEP atlas in a cohort of 50 retrospective patients. The structural organization is complemented by the functional variation of stereotactic intracerebral EEG (SEEG) signal data features establishing brain region-specific 3d-maps.

COMPARISON WITH EXISTING METHODS

The VEP atlas integrates both anatomical and functional definitions in the same atlas, adapted to applications for epilepsy patients and individualizable.

CONCLUSION

The covariation of structural and functional organization is the basis for current efforts of patient-specific large-scale brain network modeling exploiting virtual brain technologies for the identification of the epileptogenic regions in an ongoing prospective clinical trial EPINOV.

Automatic vs. Manual Detection of High Frequency Oscillations in Intracranial Recordings From the Human Temporal Lobe

Background: High frequency oscillations (HFOs) have attracted great interest among neuroscientists and epileptologists in recent years. Not only has their occurrence been linked to epileptogenesis, but also to physiologic processes, such as memory consolidation. There are at least two big challenges for HFO research. First, detection, when performed manually, is time consuming and prone to rater biases, but when performed automatically, it is biased by artifacts mimicking HFOs. Second, distinguishing physiologic from pathologic HFOs in patients with epilepsy is problematic. Here we automatically and manually detected HFOs in intracranial EEGs (iEEG) of patients with epilepsy, recorded during a visual memory task in order to assess the feasibility of the different detection approaches to identify task-related ripples, supporting the physiologic nature of HFOs in the temporal lobe. Methods: Ten patients with unclear seizure origin and bilaterally implanted macroelectrodes took part in a visual memory consolidation task. In addition to iEEG, scalp EEG, electrooculography (EOG), and facial electromyography (EMG) were recorded. iEEG channels contralateral to the suspected epileptogenic zone were inspected visually for HFOs. Furthermore, HFOs were marked automatically using an RMS detector and a Stockwell classifier. We compared the two detection approaches and assessed a possible link between task performance and HFO occurrence during encoding and retrieval trials. Results: HFO occurrence rates were significantly lower when events were marked manually. The automatic detection algorithm was greatly biased by filter-artifacts. Surprisingly, EOG artifacts as seen on scalp electrodes appeared to be linked to many HFOs in the iEEG. Occurrence rates could not be associated to memory performance, and we were not able to detect strictly defined "clear" ripples. Conclusion: Filtered graphoelements in the EEG are known to mimic HFOs and thus constitute a problem. So far, in invasive EEG recordings mostly technical artifacts and filtered epileptiform discharges have been considered as sources for these "false" HFOs. The data at hand suggests that even ocular artifacts might bias automatic detection in invasive recordings. Strict guidelines and standards for HFO detection are necessary in order to identify artifact-derived HFOs, especially in conditions when cognitive tasks might produce a high amount of artifacts.

Distinguishing false and true positive detections of high frequency oscillations

Objective.

High frequency oscillations (HFOs) are a promising biomarker of tissue that instigates seizures. However, ambiguous data and random background fluctuations can cause any HFO detector (human or automated) to falsely label non-HFO data as an HFO (a false positive detection). The objective of this paper was to identify quantitative features of HFOs that distinguish between true and false positive detections.

Approach.

Feature selection was performed using background data in multi-day, interictal intracranial recordings from ten patients. We selected the feature most similar between randomly selected segments of background data and HFOs detected in surrogate background data (false positive detections by construction). We then compared these results with fuzzy clustering of detected HFOs in clinical data to verify the feature’s applicability. We validated the feature is sensitive to false versus true positive HFO detections by using an independent data set (six subjects) scored for HFOs by three human reviewers. Lastly, we compared the effect of redacting putative false positive HFO detections on the distribution of HFOs across channels and their association with seizure onset zone (SOZ) and resected volume (RV).

Main results.

Of the 15 analyzed features, the analysis selected only skewness of the curvature (skewCurve). The feature was validated in human scored data to be associated with distinguishing true and false positive HFO detections. Automated HFO detections with higher skewCurve were more focal based on entropy measures and had increased localization to both the SOZ and RV.

Significance.

We identified a quantitative feature of HFOs which helps distinguish between true and false positive detections. Redacting putative false positive HFO detections improves the specificity of HFOs as a biomarker of epileptic tissue.

Model-Guided Design of Microelectrodes for HFO Recording

High Frequency Oscillations (HFOs, 200-600 Hz) are recognized as a biomarker of epileptogenic brain areas. This work aims at designing novel microelectrodes in order to optimize the recording and further detection of HFOs in brain (intracerebral electroencephalography, iEEG). The quality of the recorded iEEG signals is highly dependent on the electrode contact impedance, which is determined by the characteristics of the recording electrode (geometry, position, material). These properties are essential for the observability of HFOs. In this study, a previously published hippocampal neural network model is used for the simulation of interictal HFOs. An additional microelectrode model layer is implemented in order to simulate the impact of using different types and characteristics of microelectrodes on the recorded HFOs. Results indicate that a small layer PEDOT/PSS and PEDOT/CNT on microelectrodes can effectively decrease their impedance resulting in the increase of HFOs observability. This model-based study can lead to the actual design of new electrodes that will ultimately contribute to improved diagnosis prior to invasive therapies.

Interictal Fast Ripples Are Associated With the Seizure-Generating Lesion in Patients With Dual Pathology

Rationale: Patients with dual pathology have two potentially epileptogenic lesions: One in the hippocampus and one in the neocortex. If epilepsy surgery is considered, stereotactic electroencephalography (SEEG) may reveal which of the lesions is seizure-generating, but frequently, some uncertainty remains. We aimed to investigate whether interictal high-frequency oscillations (HFOs), which are a promising biomarker of epileptogenicity, are associated with the primary focus.

Methods: We retrospectively analyzed 16 patients with dual pathology. They were grouped according to their seizure-generating lesion, as suggested by ictal SEEG. An automated detector was applied to identify interictal epileptic spikes, ripples (80–250 Hz), ripples co-occurring with spikes (IES-ripples) and fast ripples (250–500 Hz). We computed a ratio R to obtain an indicator of whether rates were higher in the hippocampal lesion (R close to 1), higher in the neocortical lesion (R close to −1), or more or less similar (R close to 0).

Results: Spike and HFO rates were higher in the hippocampal than in the neocortical lesion (p < 0.001), particularly in seizure onset zone channels. Seizures originated exclusively in the hippocampus in 5 patients (group 1), in both lesions in 7 patients (group 2), and exclusively in the neocortex in 4 patients (group 3). We found a significant correlation between the patients' primary focus and the ratio R_{fast ripples}, i.e., the proportion of interictal fast ripples detected in this lesion (p < 0.05). No such correlation was observed for interictal epileptic spikes (p = 0.69), ripples (p = 0.60), and IES-ripples (p = 0.54). In retrospect, interictal fast ripples would have correctly “predicted” the primary focus in 69% of our patients (p < 0.01).

Conclusions: We report a correlation between interictal fast ripple rate and the primary focus, which was not found for epileptic spikes. Fast ripple analysis could provide helpful information for generating a hypothesis on seizure-generating networks, especially in cases with few or no recorded seizures.

Automatic and Accurate Epilepsy Ripple and Fast Ripple Detection via Virtual Sample Generation and Attention Neural Networks

About 1% of the population around the world suffers from epilepsy. The success of epilepsy surgery depends critically on pre-operative localization of epileptogenic zones. High frequency oscillations including ripples (80-250 Hz) and fast ripples (250-500 Hz) are commonly used as biomarkers to localize epileptogenic zones. Recent literature demonstrated that fast ripples indicate epileptogenic zones better than ripples. Thus, it is crucial to accurately detect fast ripples from ripples signals of magnetoencephalography for improving outcome of epilepsy surgery. This paper proposes an automatic and accurate ripple and fast ripple detection method that employs virtual sample generation and neural networks with an attention mechanism. We evaluate our proposed detector on patient data with 50 ripples and 50 fast ripples labeled by two experts. The experimental results show that our new detector outperforms multiple traditional machine learning models. In particular, our method can achieve a mean accuracy of 89.3% and an average area under the receiver operating characteristic curve of 0.88 in 50 repeats of random subsampling validation. In addition, we experimentally demonstrate the effectiveness of virtual sample generation, attention mechanism, and architecture of neural network models.

Integrated Automatic Detection, Classification and Imaging of High Frequency Oscillations With Stereoelectroencephalography

Objective

During presurgical evaluation for focal epilepsy patients, the evidence supporting the use of high frequency oscillations (HFOs) for delineating the epileptogenic zone (EZ) increased over the past decade. This study aims to develop and validate an integrated automatic detection, classification and imaging pipeline of HFOs with stereoelectroencephalography (SEEG) to narrow the gap between HFOs quantitative analysis and clinical application.

Methods

The proposed pipeline includes stages of channel inclusion, candidate HFOs detection and automatic labeling with four trained convolutional neural network (CNN) classifiers and HFOs sorting based on occurrence rate and imaging. We first evaluated the initial detector using an open simulated dataset. After that, we validated our full algorithm in a 20-patient cohort against three assumptions based on previous studies. Classified HFOs results were compared with seizure onset zone (SOZ) channels for their concordance. The receiver operating characteristic (ROC) curve and the corresponding area under the curve (AUC) were calculated representing the prediction ability of the labeled HFOs outputs for SOZ.

Results

The initial detector demonstrated satisfactory performance on the simulated dataset. The four CNN classifiers converged quickly during training, and the accuracies on the validation dataset were above 95%. The localization value of HFOs was significantly improved by HFOs classification. The AUC values of the 20 testing patients increased after HFO classification, indicating a satisfactory prediction value of the proposed algorithm for EZ identification.

Conclusion

Our detector can provide robust HFOs analysis results revealing EZ at the individual level, which may ultimately push forward the transitioning of HFOs analysis into a meaningful part of the presurgical evaluation and surgical planning.

Pitfalls in Scalp High-Frequency Oscillation Detection From Long-Term EEG Monitoring

Aims: Intracranially recorded high-frequency oscillations (>80 Hz) are considered a candidate epilepsy biomarker. Recent studies claimed their detectability on the scalp surface. We aimed to investigate the applicability of high-frequency oscillation analysis to routine surface EEG obtained at an epilepsy monitoring unit. Methods: We retrospectively analyzed surface EEGs of 18 patients with focal epilepsy and six controls, recorded during sleep under maximal medication withdrawal. As a proof of principle, the occurrence of motor task-related events during wakefulness was analyzed in a subsample of six patients with seizure- or syncope-related motor symptoms. Ripples (80-250 Hz) and fast ripples (>250 Hz) were identified by semi-automatic detection. Using semi-parametric statistics, differences in spontaneous and task-related occurrence rates were examined within subjects and between diagnostic groups considering the factors diagnosis, brain region, ripple type, and task condition. Results: We detected high-frequency oscillations in 17 out of 18 patients and in four out of six controls. Results did not show statistically significant differences in the mean rates of event occurrences, neither regarding the laterality of the epileptic focus, nor with respect to active and inactive task conditions, or the moving hand laterality. Significant differences in general spontaneous incidence [WTS(1) = 9.594; p = 0.005] that indicated higher rates of fast ripples compared to ripples, notably in patients with epilepsy compared to the control group, may be explained by variations in data quality. Conclusion: The current analysis methods are prone to biases. A common agreement on a standard operating procedure is needed to ensure reliable and economic detection of high-frequency oscillations.

SGM: a novel time-frequency algorithm based on unsupervised learning improves high-frequency oscillation detection in epilepsy

OBJECTIVE

We propose a novel automated method called the S-Transform Gaussian Mixture detection algorithm (SGM) to detect high-frequency oscillations (HFO) combining the strengths of different families of previously published detectors.

APPROACH

This algorithm does not depend on parameter tuning on a subject (or database) basis, uses time-frequency characteristics, and relies on non-supervised classification to determine if the events standing out from the baseline activity are HFO or not. SGM consists of three steps: the first stage computes the signal baseline using the entropy of the autocorrelation; the second uses the S-Transform to obtain several time-frequency features (area, entropy, and time and frequency widths); and in the third stage Gaussian mixture models cluster time-frequency features to decide if events correspond to HFO-like activity. To validate the SGM algorithm we tested its performance in simulated and real environments.

MAIN RESULTS

We assessed the algorithm on a publicly available simulated stereoelectroencephalographic (SEEG) database with varying signal-to-noise ratios (SNR), obtaining very good results for medium and high SNR signals. We further tested the SGM algorithm on real signals from patients with focal epilepsy, in which HFO detection was performed visually by experts, yielding a high agreement between experts and SGM.

SIGNIFICANCE

The SGM algorithm displayed proper performance in simulated and real environments and therefore can be used for non-supervised detection of HFO. This non-supervised algorithm does not require previous labelling by experts or parameter adjustment depending on the subject or database considered. SGM is not a computationally intensive algorithm, making it suitable to detect and characterize HFO in long-term SEEG recordings.

Unsupervised Detection of High-Frequency Oscillations Using Time-Frequency Maps and Computer Vision

High-frequency oscillations >80 Hz (HFOs) have unique features distinguishing them from spikes and artifactual components that can be well-evidenced in the time-frequency representations. We introduce an unsupervised HFO detector that uses computer-vision algorithms to detect HFO landmarks on two-dimensional (2D) time-frequency maps. To validate the detector, we introduce an analytical model of the HFO based on a sinewave having a Gaussian envelope, for which analytical equations in time-frequency space can be derived, allowing us to establish a direct correspondence between common HFO detection criteria in the time domain with the ones in the frequency domain, used by the computer-vision detection algorithm. The detector identifies potential HFO events on the time-frequency representation, which are classified as true HFOs if criteria regarding the HFO's frequency, amplitude, and duration are met. The detector is validated on simulated HFOs according to the analytical model, in the presence of noise, with different signal-to-noise ratios (SNRs) ranging from −9 to 0 dB. The detector's sensitivity was 0.64 at an SNR of −9 dB, 0.98 at −6 dB, and >0.99 at −3 dB and 0 dB, while its positive prediction value was >0.95, regardless of the SNR. Using the same simulation dataset, our detector is benchmarked against four previously published HFO detectors. The F-measure, a combined metric that takes into account both sensitivity and positive prediction value, was used to compare detection algorithms. Our detector surpassed the other detectors at −6, −3, and 0 dB and had the second best F-score at −9 dB SNR after the MNI detector (0.77 vs. 0.83). The ability to detect HFOs in clinical recordings has been tested on a set of 36 intracranial electroencephalogram (EEG) channels in six patients, with 89% of the detections being validated by two independent reviewers. The results demonstrate that the unsupervised detection of HFOs based on their 2D features in time-frequency maps is feasible and has a performance comparable or better than the most used HFO detectors.

Automatic detection of high-frequency-oscillations and their sub-groups co-occurring with interictal-epileptic-spikes

OBJECTIVE

High-frequency-oscillations (HFO) and interictal-epileptic-spikes (IES) are spatial biomarkers of the epileptogenic-zone. Those HFO spatially and temporally co-occurring with IES (IES-HFO) are potentially superior biomarkers, their use is however challenged by the difficulty in detecting the low amplitude HFO riding the high-amplitude and steep-waveform of IES. We aim to develop an automatic HFO detector with an improved performance with respect to current methods and that also correctly distinguishes IES-HFO from IES occurring in isolation (isol-IES). We also aim to validate the biomarker-value of the automatic detections by utilizing them to localize a surrogate of the epileptogenic-zone.

APPROACH

We developed automatic-detectors of HFO-Ripples (80-250 Hz), HFO-FastRipples (250-500 Hz) and IES based on kernelized support-vector-machines (SVM). The training of the HFO-detectors was based on visually marked HFO and the parameter optimization during this training-process promoted the correct discernment between IES-HFO and isol-IES. Both HFO-detectors were benchmarked against other published detectors using databases with both visually marked and simulated gold-standards. The IES-detector was trained with a publicly available simulated database and tested against both simulated and visually marked gold-standards. To validate the detections' biomarker-value, the detectors were run on intracranial-EEGs from 8 patients and each with durations of 2-3 days, the detections' cumulated per-channel occurrence-rate was then used to predict the seizure-onset-zone as a surrogate of the epileptogenic-zone.

MAIN RESULTS

The HFO-detectors obtained at least 21 more F1-score points than previously published algorithms at the lowest signal-to-noise-ratio. The success achieved when discerning between IES-HFO and isol-IES was comparable to that of other published algorithms. The automatically detected IES-HFO and IES-Ripples showed the best biomarker-value to localize the epileptogenic-zone.

SIGNIFICANCE

The developed detectors are publicly available and provide a reliable tool to further study HFO, IES-HFO and their value as biomarkers. The putative higher biomarker value from IES-HFO was validated on automatically-detected, long-term data.

Are high-frequency oscillations better biomarkers of the epileptogenic zone than spikes?

PURPOSE OF REVIEW

Precise localization of the epileptogenic zone is imperative for the success of resective surgery of drug-resistant epileptic patients. To decrease the number of surgical failures, clinical research has been focusing on finding new biomarkers. For the past decades, high-frequency oscillations (HFOs, 80-500 Hz) have ousted interictal spikes - the classical interictal marker - from the research spotlight. Many studies have claimed that HFOs were more linked to epileptogenicity than spikes. This present review aims at refining this statement in light of recent studies.

RECENT FINDINGS

Analysis based on single-patient characteristics has not been able to determine which of HFOs or spikes were better marker of epileptogenic tissues. Physiological HFOs are one of the main obstacles to translate HFOs to clinical practice as separating them from pathological HFOs remains a challenge. Fast ripples (a subgroup of HFOs, 250-500 Hz) which are mostly pathological are not found in all epileptogenic tissues.

SUMMARY

Quantified measures of HFOs and spikes give complementary results, but many barriers still persist in applying them in clinical routine. The current way of testing HFO and spike detectors and their performance in delineating the epileptogenic zone is debatable and still lacks practicality. Solutions to handle physiological HFOs have been proposed but are still at a preliminary stage.

NREM sleep is the state of vigilance that best identifies the epileptogenic zone in the interictal electroencephalogram

OBJECTIVE

Interictal epileptiform anomalies such as epileptiform discharges or high-frequency oscillations show marked variations across the sleep-wake cycle. This study investigates which state of vigilance is the best to localize the epileptogenic zone (EZ) in interictal intracranial electroencephalography (EEG).

METHODS

Thirty patients with drug-resistant epilepsy undergoing stereo-EEG (SEEG)/sleep recording and subsequent open surgery were included; 13 patients (43.3%) had good surgical outcome (Engel class I). Sleep was scored following standard criteria. Multiple features based on the interictal EEG (interictal epileptiform discharges, high-frequency oscillations, univariate and bivariate features) were used to train a support vector machine (SVM) model to classify SEEG contacts placed in the EZ. The performance of the algorithm was evaluated by the mean area under the receiver-operating characteristic (ROC) curves (AUCs) and positive predictive values (PPVs) across 10-minute sections of wake, non-rapid eye movement sleep (NREM) stages N2 and N3, REM sleep, and their combination.

RESULTS

Highest AUCs were achieved in NREM sleep stages N2 and N3 compared to wakefulness and REM (P < .01). There was no improvement when using a combination of all four states (P > .05); the best performing features in the combined state were selected from NREM sleep. There were differences between good (Engel I) and poor (Engel II-IV) outcomes in PPV (P < .05) and AUC (P < .01) across all states. The SVM multifeature approach outperformed spikes and high-frequency oscillations (P < .01) and resulted in results similar to those of the seizure-onset zone (SOZ; P > .05).

SIGNIFICANCE

Sleep improves the localization of the EZ with best identification obtained in NREM sleep stages N2 and N3. Results based on the multifeature classification in 10 minutes of NREM sleep were not different from the results achieved by the SOZ based on 12.7 days of seizure monitoring. This finding might ultimately result in a more time-efficient intracranial presurgical investigation of focal epilepsy.

Multi-feature localization of epileptic foci from interictal, intracranial EEG

OBJECTIVE

When considering all patients with focal drug-resistant epilepsy, as high as 40-50% of patients suffer seizure recurrence after surgery. To achieve seizure freedom without side effects, accurate localization of the epileptogenic tissue is crucial before its resection. We investigate an automated, fast, objective mapping process that uses only interictal data.

METHODS

We propose a novel approach based on multiple iEEG features, which are used to train a support vector machine (SVM) model for classification of iEEG electrodes as normal or pathologic using 30 min of inter-ictal recording.

RESULTS

The tissue under the iEEG electrodes, classified as epileptogenic, was removed in 17/18 excellent outcome patients and was not entirely resected in 8/10 poor outcome patients. The overall best result was achieved in a subset of 9 excellent outcome patients with the area under the receiver operating curve = 0.95.

CONCLUSION

SVM models combining multiple iEEG features show better performance than algorithms using a single iEEG marker. Multiple iEEG and connectivity features in presurgical evaluation could improve epileptogenic tissue localization, which may improve surgical outcome and minimize risk of side effects.

SIGNIFICANCE

In this study, promising results were achieved in localization of epileptogenic regions by SVM models that combine multiple features from 30 min of inter-ictal iEEG recordings.

Redaction of false high frequency oscillations due to muscle artifact improves specificity to epileptic tissue

OBJECTIVE

High Frequency Oscillations (HFOs) are a promising biomarker of epilepsy. HFOs are typically acquired on intracranial electrodes, but contamination from muscle artifacts is still problematic in HFO analysis. This paper evaluates the effect of myogenic artifacts on intracranial HFO detection and how to remove them.

METHODS

Intracranial EEG was recorded in 31 patients. HFOs were detected for the entire recording using an automated algorithm. When available, simultaneous scalp EEG was used to identify periods of muscle artifact. Those markings were used to train an automated scalp EMG detector and an intracranial EMG detector. Specificity to epileptic tissue was evaluated by comparison with seizure onset zone and resected volume in patients with good outcome.

RESULTS

EMG artifacts are frequent and produce large numbers of false HFOs, especially in the anterior temporal lobe. The scalp and intracranial EMG detectors both had high accuracy. Removing false HFOs improved specificity to epileptic tissue.

CONCLUSIONS

Evaluation of HFOs requires accounting for the effect of muscle artifact. We present two tools that effectively mitigate the effect of muscle artifact on HFOs.

SIGNIFICANCE

Removing muscle artifacts improves the specificity of HFOs to epileptic tissue. Future HFO work should account for this effect, especially when using automated algorithms or when scalp electrodes are not present.

MEEGIPS-A Modular EEG Investigation and Processing System for Visual and Automated Detection of High Frequency Oscillations

High frequency oscillations (HFOs) are electroencephalographic correlates of brain activity detectable in a frequency range above 80 Hz. They co-occur with physiological processes such as saccades, movement execution, and memory formation, but are also related to pathological processes in patients with epilepsy. Localization of the seizure onset zone, and, more specifically, of the to-be resected area in patients with refractory epilepsy seems to be supported by the detection of HFOs. The visual identification of HFOs is very time consuming with approximately 8 h for 10 min and 20 channels. Therefore, automated detection of HFOs is highly warranted. So far, no software for visual marking or automated detection of HFOs meets the needs of everyday clinical practice and research. In the context of the currently available tools and for the purpose of related local HFO study activities we aimed at converging the advantages of clinical and experimental systems by designing and developing a comprehensive and extensible software framework for HFO analysis that, on the one hand, focuses on the requirements of clinical application and, on the other hand, facilitates the integration of experimental code and algorithms. The development project included the definition of use cases, specification of requirements, software design, implementation, and integration. The work comprised the engineering of component-specific requirements, component design, as well as component- and integration-tests. A functional and tested software package is the deliverable of this activity. The project MEEGIPS, a Modular EEG Investigation and Processing System for visual and automated detection of HFOs, introduces a highly user friendly software that includes five of the most prominent automated detection algorithms. Future evaluation of these, as well as implementation of further algorithms is facilitated by the modular software architecture.

Localization of the Epileptogenic Zone Using High Frequency Oscillations

For patients with drug-resistant focal epilepsy, surgery is the therapy of choice in order to achieve seizure freedom. Epilepsy surgery foremost requires the identification of the epileptogenic zone (EZ), defined as the brain area indispensable for seizure generation. The current gold standard for identification of the EZ is the seizure-onset zone (SOZ). The fact, however that surgical outcomes are unfavorable in 40-50% of well-selected patients, suggests that the SOZ is a suboptimal biomarker of the EZ, and that new biomarkers resulting in better postsurgical outcomes are needed. Research of recent years suggested that high-frequency oscillations (HFOs) are a promising biomarker of the EZ, with a potential to improve surgical success in patients with drug-resistant epilepsy without the need to record seizures. Nonetheless, in order to establish HFOs as a clinical biomarker, the following issues need to be addressed. First, evidence on HFOs as a clinically relevant biomarker stems predominantly from retrospective assessments with visual marking, leading to problems of reproducibility and reliability. Prospective assessments of the use of HFOs for surgery planning using automatic detection of HFOs are needed in order to determine their clinical value. Second, disentangling physiologic from pathologic HFOs is still an unsolved issue. Considering the appearance and the topographic location of presumed physiologic HFOs could be immanent for the interpretation of HFO findings in a clinical context. Third, recording HFOs non-invasively via scalp electroencephalography (EEG) and magnetoencephalography (MEG) is highly desirable, as it would provide us with the possibility to translate the use of HFOs to the scalp in a large number of patients. This article reviews the literature regarding these three issues. The first part of the article focuses on the clinical value of invasively recorded HFOs in localizing the EZ, the detection of HFOs, as well as their separation from physiologic HFOs. The second part of the article focuses on the current state of the literature regarding non-invasively recorded HFOs with emphasis on findings and technical considerations regarding their localization.

High-frequency oscillations are not better biomarkers of epileptogenic tissues than spikes

OBJECTIVE

High-frequency oscillations (HFOs) in intracerebral EEG (stereoelectroencephalography; SEEG) are considered as better biomarkers of epileptogenic tissues than spikes. How this can be applied at the patient level remains poorly understood. We investigated how well HFOs and spikes can predict epileptogenic regions with a large spatial sampling at the patient level.

METHODS

We analyzed non-REM sleep SEEG recordings sampled at 2,048Hz of 30 patients. Ripples (Rs; 80-250Hz), fast ripples (FRs; 250-500Hz), and spikes were automatically detected. Rates of these markers and several combinations-spikes co-occurring with HFOs or FRs and cross-rate (Spk⊗HFO)-were compared to a quantified measure of the seizure onset zone (SOZ) by performing a receiver operating characteristic analysis for each patient individually. We used a Wilcoxon signed-rank test corrected for false-discovery rate to assess whether a marker was better than the others for predicting the SOZ.

RESULTS

A total of 2,930 channels was analyzed (median of 100 channels per patient). The HFOs or any of its variants were not statistically better than spikes. Only one feature, the cross-rate, was better than all the other markers. Moreover, fast ripples, even though very specific, were not delineating all epileptogenic tissues.

INTERPRETATION

At the patient level, the performance of HFOs is weakened by the presence of strong physiological HFO generators. Fast ripples are not sensitive enough to be the unique biomarker of epileptogenicity. Nevertheless, combining HFOs and spikes using our proposed measure-the cross-rate-is a better strategy than using only one marker. Ann Neurol 2018;83:84-97.

A Stacked Sparse Autoencoder-Based Detector for Automatic Identification of Neuromagnetic High Frequency Oscillations in Epilepsy

High-frequency oscillations (HFOs) are spontaneous magnetoencephalography (MEG) patterns that have been acknowledged as a putative biomarker to identify epileptic foci. Correct detection of HFOs in the MEG signals is crucial for the accurate and timely clinical evaluation. Since the visual examination of HFOs is time-consuming, error-prone, and with poor inter-reviewer reliability, an automatic HFOs detector is highly desirable in clinical practice. However, the existing approaches for HFOs detection may not be applicable for MEG signals with noisy background activity. Therefore, we employ the stacked sparse autoencoder (SSAE) and propose an SSAE-based MEG HFOs (SMO) detector to facilitate the clinical detection of HFOs. To the best of our knowledge, this is the first attempt to conduct HFOs detection in MEG using deep learning methods. After configuration optimization, our proposed SMO detector is outperformed other classic peer models by achieving 89.9% in accuracy, 88.2% in sensitivity, and 91.6% in specificity. Furthermore, we have tested the performance consistency of our model using various validation schemes. The distribution of performance metrics demonstrates that our model can achieve steady performance.

High-Frequency Oscillations in the Scalp Electroencephalogram: Mission Impossible without Computational Intelligence

High-frequency oscillations (HFOs) in the electroencephalogram (EEG) are thought to be a promising marker for epileptogenicity. A number of automated detection algorithms have been developed for reliable analysis of invasively recorded HFOs. However, invasive recordings are not widely applicable since they bear risks and costs, and the harm of the surgical intervention of implantation needs to be weighted against the informational benefits of the invasive examination. In contrast, scalp EEG is widely available at low costs and does not bear any risks. However, the detection of HFOs on the scalp represents a challenge that was taken on so far mostly via visual detection. Visual detection of HFOs is, in turn, highly time-consuming and subjective. In this review, we discuss that automated detection algorithms for detection of HFOs on the scalp are highly warranted because the available algorithms were all developed for invasively recorded EEG and do not perform satisfactorily in scalp EEG because of the low signal-to-noise ratio and numerous artefacts as well as physiological activity that obscures the tiny phenomena in the high-frequency range.

Evaluating resective surgery targets in epilepsy patients: A comparison of quantitative EEG methods

BACKGROUND

Quantitative analysis of intracranial EEG is a promising tool to assist clinicians in the planning of resective brain surgery in patients suffering from pharmacoresistant epilepsies. Quantifying the accuracy of such tools, however, is nontrivial as a ground truth to verify predictions about hypothetical resections is missing.

NEW METHOD

As one possibility to address this, we use customized hypotheses tests to examine the agreement of the methods on a common set of patients. One method uses machine learning techniques to enable the predictive modeling of EEG time series. The other estimates nonlinear interrelation between EEG channels. Both methods were independently shown to distinguish patients with excellent post-surgical outcome (Engel class I) from those without improvement (Engel class IV) when assessing the electrodes associated with the tissue that was actually resected during brain surgery. Using the AND and OR conjunction of both methods we evaluate the performance gain that can be expected when combining them.

RESULTS

Both methods' assessments correlate strongly positively with the similarity between a hypothetical resection and the corresponding actual resection in class I patients. Moreover, the Spearman rank correlation between the methods' patient rankings is significantly positive.

COMPARISON WITH EXISTING METHOD(S)

To our best knowledge, this is the first study comparing surgery target assessments from fundamentally differing techniques.

CONCLUSIONS

Although conceptually completely independent, there is a relation between the predictions obtained from both methods. Their broad consensus supports their application in clinical practice to provide physicians additional information in the process of presurgical evaluation.

Kurtosis-Based Detection of Intracranial High-Frequency Oscillations for the Identification of the Seizure Onset Zone

Pathological High-Frequency Oscillations (HFOs) have been recently proposed as potential biomarker of the seizure onset zone (SOZ) and have shown superior accuracy to interictal epileptiform discharges in delineating its anatomical boundaries. Characterization of HFOs is still in its infancy and this is reflected in the heterogeneity of analysis and reporting methods across studies and in clinical practice. The clinical approach to HFOs identification and quantification usually still relies on visual inspection of EEG data. In this study, we developed a pipeline for the detection and analysis of HFOs. This includes preliminary selection of the most informative channels exploiting statistical properties of the pre-ictal and ictal intracranial EEG (iEEG) time series based on spectral kurtosis, followed by wavelet-based characterization of the time-frequency properties of the signal. We performed a preliminary validation analyzing EEG data in the ripple frequency band (80-250 Hz) from six patients with drug-resistant epilepsy who underwent pre-surgical evaluation with stereo-EEG (SEEG) followed by surgical resection of pathologic brain areas, who had at least two-year positive post-surgical outcome. In this series, kurtosis-driven selection and wavelet-based detection of HFOs had average sensitivity of 81.94% and average specificity of 96.03% in identifying the HFO area which overlapped with the SOZ as defined by clinical presurgical workup. Furthermore, the kurtosis-based channel selection resulted in an average reduction in computational time of 66.60%.

Progress and Remaining Challenges in the Application of High Frequency Oscillations as Biomarkers of Epileptic Brain

High-frequency oscillations (HFOs: 100 - 600 Hz) have been widely proposed as biomarkers of epileptic brain tissue. In addition, HFOs over a broader range of frequencies spanning 30 - 2000 Hz are potential biomarkers of both physiological and pathological brain processes. The majority of the results from humans with focal epilepsy have focused on HFOs recorded directly from the brain with intracranial EEG (iEEG) in the high gamma (65 - 100 Hz), ripple (100 - 250 Hz), and fast ripple (250 - 600 Hz) frequency ranges. These results are supplemented by reports of HFOs recorded with iEEG in the low gamma (30 - 65Hz) and very high frequency (500 - 2000 Hz) ranges. Visual detection of HFOs is laborious and limited by poor inter-rater agreement; and the need for accurate, reproducible automated HFOs detection is well recognized. In particular, the clinical translation of HFOs as a biomarker of the epileptogenic brain has been limited by the ability to reliably detect and accurately classify HFOs as physiological or pathological. Despite these challenges, there has been significant progress in the field, which is the subject of this review. Furthermore, we provide data and corresponding analytic code in an effort to promote reproducible research and accelerate clinical translation.