High frequency oscillations in intra-operative electrocorticography before and after epilepsy surgery

Prognostic Value of Complete Resection of the High-Frequency Oscillation Area in Intracranial EEG: A Systematic Review and Meta-Analysis

BACKGROUND AND OBJECTIVES

High-frequency oscillations (HFOs; ripples 80-250 Hz; fast ripples [FRs] 250-500 Hz) recorded with intracranial electrodes generated excitement and debate about their potential to localize epileptogenic foci. We performed a systematic review and meta-analysis on the prognostic value of complete resection of the HFOs-area (crHFOs-area) for epilepsy surgical outcome in intracranial EEG (iEEG) accessing multiple subgroups.

METHODS

We searched PubMed, Embase, and Web of Science for original research from inception to October 27, 2022. We defined favorable surgical outcome (FSO) as Engel class I, International League Against Epilepsy class 1, or seizure-free status. The prognostic value of crHFOs-area for FSO was assessed by (1) the pooled FSO proportion after crHFOs-area; (2) FSO for crHFOs-area vs without crHFOs-area; and (3) the predictive performance. We defined high combined prognostic value as FSO proportion >80% + FSO crHFOs-area >without crHFOs-area + area under the curve (AUC) >0.75 and examined this for the clinical subgroups (study design, age, diagnostic type, HFOs-identification method, HFOs-rate thresholding, and iEEG state). Temporal lobe epilepsy (TLE) was compared with extra-TLE through dichotomous variable analysis. Individual patient analysis was performed for sex, affected hemisphere, MRI findings, surgery location, and pathology.

RESULTS

Of 1,387 studies screened, 31 studies (703 patients) met our eligibility criteria. Twenty-seven studies (602 patients) analyzed FRs and 20 studies (424 patients) ripples. Pooled FSO proportion after crHFOs-area was 81% (95% CI 76%-86%) for FRs and 82% (73%-89%) for ripples. Patients with crHFOs-area achieved more often FSO than those without crHFOs-area (FRs odds ratio [OR] 6.38, 4.03-10.09, p < 0.001; ripples 4.04, 2.32-7.04, p < 0.001). The pooled AUCs were 0.81 (0.77-0.84) for FRs and 0.76 (0.72-0.79) for ripples. Combined prognostic value was high in 10 subgroups: retrospective, children, long-term iEEG, threshold (FRs and ripples) and automated detection and interictal (FRs). FSO after complete resection of FRs-area (crFRs-area) was achieved less often in people with TLE than extra-TLE (OR 0.37, 0.15-0.89, p = 0.006). Individual patient analyses showed that crFRs-area was seen more in patients with FSO with than without MRI lesions (p = 0.02 after multiple correction).

DISCUSSION

Complete resection of the brain area with HFOs is associated with good postsurgical outcome. Its prognostic value holds, especially for FRs, for various subgroups. The use of HFOs for extra-TLE patients requires further evidence.

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.

Learn how to interpret and use intracranial EEG findings

Epilepsy surgery is the therapy of choice for many patients with drug-resistant focal epilepsy. Recognizing and describing ictal and interictal patterns with intracranial electroencephalography (EEG) recordings is important in order to most efficiently leverage advantages of this technique to accurately delineate the seizure-onset zone before undergoing surgery. In this seminar in epileptology, we address learning objective "1.4.11 Recognize and describe ictal and interictal patterns with intracranial recordings" of the International League against Epilepsy curriculum for epileptologists. We will review principal considerations of the implantation planning, summarize the literature for the most relevant ictal and interictal EEG patterns within and beyond the Berger frequency spectrum, review invasive stimulation for seizure and functional mapping, discuss caveats in the interpretation of intracranial EEG findings, provide an overview on special considerations in children and in subdural grids/strips, and review available quantitative/signal analysis approaches. To be as practically oriented as possible, we will provide a mini atlas of the most frequent EEG patterns, highlight pearls for its not infrequently challenging interpretation, and conclude with two illustrative case examples. This article shall serve as a useful learning resource for trainees in clinical neurophysiology/epileptology by providing a basic understanding on the concepts of invasive intracranial EEG.

Interictal magnetoencephalography abnormalities to guide intracranial electrode implantation and predict surgical outcome

Intracranial EEG is the gold standard technique for epileptogenic zone localization but requires a preconceived hypothesis of the location of the epileptogenic tissue. This placement is guided by qualitative interpretations of seizure semiology, MRI, EEG and other imaging modalities, such as magnetoencephalography. Quantitative abnormality mapping using magnetoencephalography has recently been shown to have potential clinical value. We hypothesized that if quantifiable magnetoencephalography abnormalities were sampled by intracranial EEG, then patients' post-resection seizure outcome may be better. Thirty-two individuals with refractory neocortical epilepsy underwent magnetoencephalography and subsequent intracranial EEG recordings as part of presurgical evaluation. Eyes-closed resting-state interictal magnetoencephalography band power abnormality maps were derived from 70 healthy controls as a normative baseline. Magnetoencephalography abnormality maps were compared to intracranial EEG electrode implantation, with the spatial overlap of intracranial EEG electrode placement and cerebral magnetoencephalography abnormalities recorded. Finally, we assessed if the implantation of electrodes in abnormal tissue and subsequent resection of the strongest abnormalities determined by magnetoencephalography and intracranial EEG corresponded to surgical success. We used the area under the receiver operating characteristic curve as a measure of effect size. Intracranial electrodes were implanted in brain tissue with the most abnormal magnetoencephalography findings-in individuals that were seizure-free postoperatively (T = 3.9, P = 0.001) but not in those who did not become seizure-free. The overlap between magnetoencephalography abnormalities and electrode placement distinguished surgical outcome groups moderately well (area under the receiver operating characteristic curve = 0.68). In isolation, the resection of the strongest abnormalities as defined by magnetoencephalography and intracranial EEG separated surgical outcome groups well, area under the receiver operating characteristic curve = 0.71 and area under the receiver operating characteristic curve = 0.74, respectively. A model incorporating all three features separated surgical outcome groups best (area under the receiver operating characteristic curve = 0.80). Intracranial EEG is a key tool to delineate the epileptogenic zone and help render individuals seizure-free postoperatively. We showed that data-driven abnormality maps derived from resting-state magnetoencephalography recordings demonstrate clinical value and may help guide electrode placement in individuals with neocortical epilepsy. Additionally, our predictive model of postoperative seizure freedom, which leverages both magnetoencephalography and intracranial EEG recordings, could aid patient counselling of expected outcome.

Localizing epileptogenic zones with high-frequency oscillations and directed connectivity

PURPOSE

Precise localization of the epileptogenic zone (EZ) is essential for epilepsy surgery. Existing methods often fail to detect slow onset patterns or similar neural activities presented in the recorded signals. To address this issue, we propose a new measure to quantify epileptogenicity, i.e., the connectivity high-frequency epileptogenicity index (cHFEI).

METHODS

The cHFEI method combines directed connectivity and high-frequency oscillations (HFOs) to measure the epileptogenicity of regions involved in a brain network. By applying this method to stereoelectroencephalography (SEEG) recordings of 49 seizures in 20 patients, we calculated the accuracy, sensitivity, and precision with a visually identified epileptogenic zone as a reference. The performance was evaluated by the confusion matrix and the area under the receiver operating characteristic (ROC) curve.

RESULTS

Epileptic network estimation based on cHFEI successfully distinguished brain regions involved in seizure onset from the propagation network. Moreover, cHFEI outperformed other existing detection methods in the estimation of EZs in all patients, with an average area under the ROC curve of 0.88 and an accuracy of 0.85.

CONCLUSIONS

cHFEI can characterize EZ in a robust manner despite various seizure onset patterns and has potential application in epilepsy therapy.

Quantitative approaches to guide epilepsy surgery from intracranial EEG

Over the past 10 years, the drive to improve outcomes from epilepsy surgery has stimulated widespread interest in methods to quantitatively guide epilepsy surgery from intracranial EEG (iEEG). Many patients fail to achieve seizure freedom, in part due to the challenges in subjective iEEG interpretation. To address this clinical need, quantitative iEEG analytics have been developed using a variety of approaches, spanning studies of seizures, interictal periods, and their transitions, and encompass a range of techniques including electrographic signal analysis, dynamical systems modeling, machine learning and graph theory. Unfortunately, many methods fail to generalize to new data and are sensitive to differences in pathology and electrode placement. Here, we critically review selected literature on computational methods of identifying the epileptogenic zone from iEEG. We highlight shared methodological challenges common to many studies in this field and propose ways that they can be addressed. One fundamental common pitfall is a lack of open-source, high-quality data, which we specifically address by sharing a centralized high-quality, well-annotated, multicentre dataset consisting of >100 patients to support larger and more rigorous studies. Ultimately, we provide a road map to help these tools reach clinical trials and hope to improve the lives of future patients.

Epileptogenic high-frequency oscillations present larger amplitude both in mesial temporal and neocortical regions

High-frequency oscillations (HFO) are a promising biomarker for the identification of epileptogenic tissue. While HFO rates have been shown to predict seizure outcome, it is not yet clear whether their morphological features might improve this prediction. We validated HFO rates against seizure outcome and delineated the distribution of HFO morphological features. We collected stereo-EEG recordings from 20 patients (231 electrodes; 1,943 contacts). We computed HFO rates (the co-occurrence of ripples and fast ripples) through a validated automated detector during non-rapid eye movement sleep. Applying machine learning, we delineated HFO morphological features within and outside epileptogenic tissue across mesial temporal lobe (MTL) and Neocortex. HFO rates predicted seizure outcome with 85% accuracy, 79% specificity, 100% sensitivity, 100% negative predictive value, and 67% positive predictive value. The analysis of HFO features showed larger amplitude in the epileptogenic tissue, similar morphology for epileptogenic HFO in MTL and Neocortex, and larger amplitude for physiological HFO in MTL. We confirmed HFO rates as a reliable biomarker for epilepsy surgery and characterized the potential clinical relevance of HFO morphological features. Our results support the prospective use of HFO in epilepsy surgery and contribute to the anatomical mapping of HFO morphology.

Local neuronal excitation and global inhibition during epileptic fast ripples in humans

Understanding the neuronal basis of epileptic activity is a major challenge in neurology. Cellular integration into larger scale networks is all the more challenging. In the local field potential, interictal epileptic discharges can be associated with fast ripples (200-600 Hz), which are a promising marker of the epileptogenic zone. Yet, how neuronal populations in the epileptogenic zone and in healthy tissue are affected by fast ripples remain unclear. Here, we used a novel 'hybrid' macro-micro depth electrode in nine drug-resistant epileptic patients, combining classic depth recording of local field potentials (macro-contacts) and two or three tetrodes (four micro-wires bundled together) enabling up to 15 neurons in local circuits to be simultaneously recorded. We characterized neuronal responses (190 single units) with the timing of fast ripples (2233 fast ripples) on the same hybrid and other electrodes that target other brain regions. Micro-wire recordings reveal signals that are not visible on macro-contacts. While fast ripples detected on the closest macro-contact to the tetrodes were always associated with fast ripples on the tetrodes, 82% of fast ripples detected on tetrodes were associated with detectable fast ripples on the nearest macro-contact. Moreover, neuronal recordings were taken in and outside the epileptogenic zone of implanted epileptic subjects and they revealed an interlay of excitation and inhibition across anatomical scales. While fast ripples were associated with increased neuronal activity in very local circuits only, they were followed by inhibition in large-scale networks (beyond the epileptogenic zone, even in healthy cortex). Neuronal responses to fast ripples were homogeneous in local networks but differed across brain areas. Similarly, post-fast ripple inhibition varied across recording locations and subjects and was shorter than typical inter-fast ripple intervals, suggesting that this inhibition is a fundamental refractory process for the networks. These findings demonstrate that fast ripples engage local and global networks, including healthy tissue, and point to network features that pave the way for new diagnostic and therapeutic strategies. They also reveal how even localized pathological brain dynamics can affect a broad range of cognitive functions.

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.

High-Frequency Oscillations and Epileptogenic Network

Epilepsy is a network disease caused by aberrant neocortical large-scale connectivity spanning regions on the scale of several centimeters. High-frequency oscillations, characterized by the 80-600 Hz signals in electroencephalography, have been proven to be a promising biomarker of epilepsy that can be used in assessing the severity and susceptibility of epilepsy as well as the location of the epileptogenic zone. However, the presence of a high-frequency oscillation network remains a topic of debate as high-frequency oscillations have been previously thought to be incapable of propagation, and the relationship between high-frequency oscillations and the epileptogenic network has rarely been discussed. Some recent studies reported that high-frequency oscillations may behave like networks that are closely relevant to the epileptogenic network. Pathological highfrequency oscillations are network-driven phenomena and elucidate epileptogenic network development; high-frequency oscillations show different characteristics coincident with the epileptogenic network dynamics, and cross-frequency coupling between high-frequency oscillations and other signals may mediate the generation and propagation of abnormal discharges across the network.

Spikes and High Frequency Oscillations in Lateral Neocortical Temporal Lobe Epilepsy: Can They Predict the Success Chance of Hippocampus-Sparing Resections?

Purpose

We investigated the distribution of spikes and HFOs recorded during intraoperative electrocorticography (ioECoG) and tried to elaborate a predictive model for postsurgical outcomes of patients with lateral neocortical temporal lobe epilepsy (TLE) whose mesiotemporal structures are left in situ.

Methods

We selected patients with temporal lateral neocortical epilepsy focus who underwent ioECoG-tailored resections without amygdalo–hippocampectomies. We visually marked spikes, ripples (80–250 Hz), and fast ripples (FRs; 250–500 Hz) on neocortical and mesiotemporal channels before and after resections. We looked for differences in event rates and resection ratios between good (Engel 1A) and poor outcome groups and performed logistic regression analysis to identify outcome predictors.

Results

Fourteen out of 24 included patients had a good outcome. The poor-outcome patients showed higher rates of ripples on neocortical channels distant from the resection in pre- and post-ioECoG than people with good outcomes (p_pre = 0.04, p_post = 0.05). Post-ioECoG FRs were found only in poor-outcome patients (N = 3). A prediction model based on regression analysis showed low rates of mesiotemporal post-ioECoG ripples (OR_mesio = 0.13, p_mesio = 0.04) and older age at epilepsy onset (OR = 1.76, p = 0.04) to be predictors of good seizure outcome.

Conclusion

HFOs in ioECoG may help to inform the neurosurgeon of the hippocampus-sparing resection success chance in patients with lateral neocortical TLE.

High-frequency oscillations detected by electroencephalography as biomarkers to evaluate treatment outcome, mirror pathological severity and predict susceptibility to epilepsy

High-frequency oscillations (HFOs) in the electroencephalography (EEG) have been extensively investigated as a potential biomarker of epileptogenic zones. The understanding of the role of HFOs in epilepsy has been advanced considerably over the past decade, and the use of scalp EEG facilitates recordings of HFOs. HFOs were initially applied in large scale in epilepsy surgery and are now being utilized in other applications. In this review, we summarize applications of HFOs in 3 subtopics: (1) HFOs as biomarkers to evaluate epilepsy treatment outcome; (2) HFOs as biomarkers to measure seizure propensity; (3) HFOs as biomarkers to reflect the pathological severity of epilepsy. Nevertheless, knowledge regarding the above clinical applications of HFOs remains limited at present. Further validation through prospective studies is required for its reliable application in the clinical management of individual epileptic patients.

Microscale dynamics of electrophysiological markers of epilepsy

OBJECTIVE

Interictal discharges (IIDs) and high frequency oscillations (HFOs) are established neurophysiologic biomarkers of epilepsy, while microseizures are less well studied. We used custom poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS) microelectrodes to better understand these markers' microscale spatial dynamics.

METHODS

Electrodes with spatial resolution down to 50 µm were used to record intraoperatively in 30 subjects. IIDs' degree of spread and spatiotemporal paths were generated by peak-tracking followed by clustering. Repeating HFO patterns were delineated by clustering similar time windows. Multi-unit activity (MUA) was analyzed in relation to IID and HFO timing.

RESULTS

We detected IIDs encompassing the entire array in 93% of subjects, while localized IIDs, observed across < 50% of channels, were seen in 53%. IIDs traveled along specific paths. HFOs appeared in small, repeated spatiotemporal patterns. Finally, we identified microseizure events that spanned 50-100 µm. HFOs covaried with MUA, but not with IIDs.

CONCLUSIONS

Overall, these data suggest that irritable cortex micro-domains may form part of an underlying pathologic architecture which could contribute to the seizure network.

SIGNIFICANCE

These results, supporting the possibility that epileptogenic cortex comprises a mosaic of irritable domains, suggests that microscale approaches might be an important perspective in devising novel seizure control therapies.

High-Frequency Oscillations on Interictal Epileptiform Discharges in Routinely Acquired Scalp EEG: Can It Be Used as a Prognostic Marker?

Introduction

Despite all the efforts for optimizing epilepsy management in children over the past decades, there is no clear consensus regarding whether to treat or not to treat epileptiform discharges (EDs) after a first unprovoked seizure or the optimal duration of therapy with anti-seizure medication (ASM). It is therefore highly needed to find markers on scalp electroencephalogram (EEG) that can help identify pathological EEG discharges that require treatment.

Aim of the study

This retrospective study aimed to identify whether the coexistence of ripples/high-frequency oscillations (HFOs) with interictal EDs (IEDs) in routinely acquired scalp EEG is associated with a higher risk of seizure recurrence and could be used as a prognostic marker.

Methods

100 children presenting with new onset seizure to Children’s Medical Center- Dallas during 2015–2016, who were not on ASM and had focal EDs on an awake and sleep EEG recorded with sample frequency of 500 HZ, were randomly identified by database review. EEGs were analyzed blinded to the data of the patients. HFOs were visually identified using review parameters including expanded time base and adjusted filter settings.

Results

The average age of patients was 6.3 years (±4.35 SD). HFOs were visually identified in 19% of the studied patients with an inter-rater reliability of 99% for HFO negative discharges and 78% agreement for identification of HFOs. HFOs were identified more often in the younger age group; however, they were identified in 11% of patients >5 years old. They were more frequently associated with spikes than with sharp waves and more often with higher amplitude EDs. Patients with HFOs were more likely to have a recurrence of seizures in the year after the first seizure (P < 0.05) and to continue to have seizures after 2 years (P < 0.0001). There was no statistically significant difference between the two groups with regards to continuing ASM after 2 years.

Conclusion

Including analysis for HFOs in routine EEG interpretation may increase the yield of the study and help guide the decision to either start or discontinue ASM. In the future, this may also help to identify pathological discharges with deleterious effects on the growing brain and set a new target for the management of epilepsy.

Application of a convolutional neural network for fully-automated detection of spike ripples in the scalp electroencephalogram

BACKGROUND

A reliable biomarker to identify cortical tissue responsible for generating epileptic seizures is required to guide prognosis and treatment in epilepsy. Combined spike ripple events are a promising biomarker for epileptogenic tissue that currently require expert review for accurate identification. This expert review is time consuming and subjective, limiting reproducibility and high-throughput applications.

NEW METHOD

To address this limitation, we develop a fully-automated method for spike ripple detection. The method consists of a convolutional neural network trained to compute the probability that a spectrogram image contains a spike ripple.

RESULTS

We validate the proposed spike ripple detector on expert-labeled data and show that this detector accurately separates subjects with low and high seizure risks.

COMPARISON WITH EXISTING METHOD

The proposed method performs as well as existing methods that require manual validation of candidate spike ripple events. The introduction of a fully automated method reduces subjectivity and increases rigor and reproducibility of this epilepsy biomarker.

CONCLUSION

We introduce and validate a fully-automated spike ripple detector to support utilization of this epilepsy biomarker in clinical and translational work.

Are HFOs in the Intra-operative ECoG Related to Hippocampal Sclerosis, Volume and IQ?

Temporal lobe epilepsy (TLE) is the most common form of refractory focal epilepsy and is often associated with hippocampal sclerosis (HS) and cognitive disturbances. Over the last decade, high frequency oscillations (HFOs) in the intraoperative electrocorticography (ioECoG) have been proposed to be biomarkers for the delineation of epileptic tissue but hippocampal ripples have also been associated with memory consolidation. Healthy hippocampi can show prolonged ripple activity in stereo- EEG. We aimed to identify how the HFO rates [ripples (80–250 Hz, fast ripples (250–500 Hz); prolonged ripples (80–250 Hz, 200–500 ms)] in the pre-resection ioECoG over subtemporal area (hippocampus) and lateral temporal neocortex relate to presence of hippocampal sclerosis, the hippocampal volume quantified on MRI and the severity of cognitive impairment in TLE patients. Volumetric measurement of hippocampal subregions was performed in 47 patients with TLE, who underwent ioECoG. Ripples, prolonged ripples, and fast ripples were visually marked and rates of HFOs were calculated. The intellectual quotient (IQ) before resection was determined. There was a trend toward higher rates of ripples and fast ripples in subtemporal electrodes vs. the lateral neocortex (ripples: 2.1 vs. 1.3/min; fast ripples: 0.9 vs. 0.2/min). Patients with HS showed higher rates of subtemporal fast ripples than other patients (Z = −2.51, p = 0.012). Prolonged ripples were only found in the lateral temporal neocortex. The normalized ratio (smallest/largest) of hippocampal volume was correlated to pre-resection IQ (r = 0.45, p = 0.015). There was no correlation between HFO rates and hippocampal volumes or HFO rates and IQ. To conclude, intra-operative fast ripples were a marker for HS, but ripples and fast ripples were not linearly correlated with either the amount of hippocampal atrophy, nor for pre-surgical IQ.

Can we use intraoperative high-frequency oscillations to guide tumor-related epilepsy surgery?

Objective

In people with low‐grade intrinsic brain tumors, an epileptic focus is often located close to the lesion. High‐frequency oscillations (HFOs) in electrocorticography (ECoG) might help to delineate this focus. We investigated the relationship between HFOs and low‐grade brain tumors and their potential value for tumor‐related epilepsy surgery.

Methods

We analyzed pre‐ and postresection intraoperative ECoG in 41 patients with refractory epilepsy and a low‐grade lesion. Electrodes were designated as overlying the tumor, adjacent resected tissue (peritumoral), or outside the resection bed using magnetic resonance imaging (MRI) and intraoperative photographs. We then used a semiautomated approach to detect HFOs as either ripples (80–250 Hz) or fast ripples (250–500 Hz).

Results

The rate of fast ripples was higher in electrodes covering tumor and peritumoral tissue than outside the resection (p = .04). Mesiotemporal tumors showed more ripples (p = .002), but not more fast ripples (p = .07), than superficial tumors. Rates of fast ripples were higher in glioma and extraventricular neurocytoma than in ganglioglioma or dysembryoplastic neuroepithelial tumor (DNET). The rate of ripples and fast ripples in postresection ECoG was not higher in patients with residual tumor tissue on MRI than those without. The rate of ripples in postresection ECoG was higher in patients with good than bad seizure outcome (p = .03). Fast ripples outside the resection and in post‐ECoG seem related to seizure recurrence.

Significance

Fast ripples in intraoperative ECoG can be used to help guide resection in tumor‐related epilepsy surgery. Preresection fast ripples occur predominantly in epileptogenic tumor and peritumoral tissue. Fast ripple rates are higher in glioma and extraventricular neurocytoma than in ganglioglioma and DNET.

Physiological Ripples Associated With Sleep Spindles Can Be Identified in Patients With Refractory Epilepsy Beyond Mesio-Temporal Structures

Introduction: High frequency oscillations (HFO) are promising biomarkers of epileptic tissue. While group analysis suggested a correlation between surgical removal of HFO generating tissue and seizure free outcome, HFO could not predict seizure outcome on an individual patient level. One possible explanation is the lack of differentiation between physiological and epileptic HFO. In the mesio-temporal lobe, a proportion of physiological ripples can be identified by their association with scalp sleep spindles. Spike associated ripples in contrast can be considered epileptic. This study investigated whether categorizing ripples by the co-occurrence with sleep spindles or spikes improves outcome prediction after surgery. Additionally, it aimed to investigate whether spindle-ripple association is limited to the mesio-temporal lobe structures or visible across the whole brain. Methods: We retrospectively analyzed EEG of 31 patients with chronic intracranial EEG. Sleep spindles in scalp EEG and ripples and epileptic spikes in iEEG were automatically detected. Three ripple subtypes were obtained: SpindleR, Non-SpindleR, and SpikeR. Rate ratios between removed and non-removed brain areas were calculated. We compared the distinct ripple subtypes and their rates in different brain regions, inside and outside seizure onset areas and between patients with good and poor seizure outcome. Results: SpindleR were found across all brain regions. SpikeR had significantly higher rates in the SOZ than in Non-SOZ channels. A significant positive correlation between removal of ripple-events and good outcome was found for the mixed ripple group (r_s = 0.43, p = 0.017) and for ripples not associated with spindles (r_s=0.40, p = 0.044). Also, a significantly high proportion of spikes associated with ripples were removed in seizure free patients (p = 0.036). Discussion: SpindleR are found in mesio-temporal and neocortical structures, indicating that ripple-spindle-coupling might have functional importance beyond mesio-temporal structures. Overall, the proportion of SpindleR was low and separating spindle and spike associated ripples did not improve outcome prediction in our patient group. SpindleR analysis therefore can be a tool to identify physiological events but needs to be used in combination with other methods to have clinical relevance.

Detecting High Frequency Oscillations for Stereoelectroencephalography in Epilepsy via Hypergraph Learning

Successful epilepsy surgeries depend highly on pre-operative localization of epileptogenic zones. Stereoelectroencephalography (SEEG) records interictal and ictal activities of the epilepsy in order to precisely find and localize epileptogenic zones in clinical practice. While it is difficult to find distinct ictal onset patterns generated the seizure onset zone from SEEG recordings in a confined region, high frequency oscillations are commonly considered as putative biomarkers for the identification of epileptogenic zones. Therefore, automatic and accurate detection of high frequency oscillations in SEEG signals is crucial for timely clinical evaluation. This work formulates the detection of high frequency oscillations as a signal segment classification problem and develops a hypergraph-based detector to automatically detect high frequency oscillations such that human experts can visually review SEEG signals. We evaluated our method on 4,000 signal segments from clinical SEEG recordings that contain both ictal and interictal data obtained from 19 patients who suffer from refractory focal epilepsy. The experimental results demonstrate the effectiveness of the proposed detector that can successfully localize interictal high frequency oscillations and outperforms multiple peer machine learning methods. In particular, the proposed detector achieved 90.7% in accuracy, 80.9% in sensitivity, and 96.9% in specificity.

Impact of predictive, preventive and precision medicine strategies in epilepsy

Over the last decade, advances in genetics, neuroimaging and EEG have enabled the aetiology of epilepsy to be identified earlier in the disease course than ever before. At the same time, progress in the study of experimental models of epilepsy has provided a better understanding of the mechanisms underlying the condition and has enabled the identification of therapies that target specific aetiologies. We are now witnessing the impact of these advances in our daily clinical practice. Thus, now is the time for a paradigm shift in epilepsy treatment from a reactive attitude, treating patients after the onset of epilepsy and the initiation of seizures, to a proactive attitude that is more broadly integrated into a 'P4 medicine' approach. This P4 approach, which is personalized, predictive, preventive and participatory, puts patients at the centre of their own care and, ultimately, aims to prevent the onset of epilepsy. This aim will be achieved by adapting epilepsy treatments not only to a given syndrome but also to a given patient and moving from the usual anti-seizure treatments to personalized treatments designed to target specific aetiologies. In this Review, we present the current state of this ongoing revolution, emphasizing the impact on clinical practice.

Trends in the use of automated algorithms for the detection of high-frequency oscillations associated with human epilepsy

High-frequency oscillations (HFOs) in intracranial electroencephalography (EEG) are a promising biomarker of the epileptogenic zone and tool for surgical planning. Many studies have shown that a high rate of HFOs (number per minute) is correlated with the seizure-onset zone, and complete removal of HFO-generating brain regions has been associated with seizure-free outcome after surgery. In order to use HFOs as a biomarker, these transient events must first be detected in electrophysiological data. Because visual detection of HFOs is time-consuming and subject to low interrater reliability, many automated algorithms have been developed, and they are being used increasingly for such studies. However, there is little guidance on how to select an algorithm, implement it in a clinical setting, and validate the performance. Therefore, we aim to review automated HFO detection algorithms, focusing on conceptual similarities and differences between them. We summarize the standard steps for data pre-processing, as well as post-processing strategies for rejection of false-positive detections. We also detail four methods for algorithm testing and validation, and we describe the specific goal achieved by each one. We briefly review direct comparisons of automated algorithms applied to the same data set, emphasizing the importance of optimizing detection parameters. Then, to assess trends in the use of automated algorithms and their potential for use in clinical studies, we review evidence for the relationship between automatically detected HFOs and surgical outcome. We conclude with practical recommendations and propose standards for the selection, implementation, and validation of automated HFO-detection algorithms.

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.

High frequency oscillations relate to cognitive improvement after epilepsy surgery in children

OBJECTIVE

To investigate how high frequency oscillations (HFOs; ripples 80-250 Hz, fast ripples (FRs) 250-500 Hz) and spikes in intra-operative electrocorticography (ioECoG) relate to cognitive outcome after epilepsy surgery in children.

METHODS

We retrospectively included 20 children who were seizure free after epilepsy surgery using ioECoG and determined their intelligence quotients (IQ) pre- and two years postoperatively. We analyzed whether the number of HFOs and spikes in pre- and postresection ioECoGs, and their change in the non-resected areas relate to cognitive improvement (with ≥ 5 IQ points increase considered to be clinically relevant (=IQ+ group) and < 5 IQ points as irrelevant (=IQ- group)).

RESULTS

The IQ+ group showed significantly more FRs in the resected tissue (p = 0.01) and less FRs in the postresection ioECoG (p = 0.045) compared to the IQ- group. Postresection decrease of ripples on spikes was correlated with postoperative cognitive improvement (correlation coefficient = -0.62 with p = 0.01).

CONCLUSIONS

Postoperative cognitive improvement was related to reduction of pathological HFOs signified by removing FR generating areas with subsequently less residual FRs, and decrease of ripples on spikes in the resection edge of the non-resected area.

SIGNIFICANCE

HFOs recorded in ioECoG could play a role as biomarkers in the prediction and understanding of cognitive outcome after epilepsy surgery.

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.

Graph index complexity as a novel surrogate marker of high frequency oscillations in delineating the seizure onset zone

OBJECTIVE

To investigate the Graph Index Complexity (uGIC) as a marker of high frequency oscillatory (HFO) activity, the seizure onset zone (SOZ), and surgical outcome.

METHODS

The SOZ, rates of HFOs at two thresholds (broad, strict), and uGIC were determined using EEG data from 41 patients. The correlation between HFOs and uGIC were calculated. HFOs and uGIC were compared within and outside the SOZ. Postsurgical outcome was compared to the colocalization of HFOs and resected SOZ.

RESULTS

There was significant correlation between uGIC and both broad (r = 0.69, p < 0.0005) and strict HFOs (r = 0.48, p < 0.0005). All were significantly greater within the SOZ overall, but only in 17/41 (strict, uGIC) or 18/41 (broad) patients. HFO markers were significantly greater within the SOZ for 8/15 patients with positive postsurgical outcomes, but not for any patients with negative outcomes (0/5).

CONCLUSION

The uGIC is a marker of HFO activity, while HFOs and uGIC are markers of the SOZ overall. Colocalization of HFOs and the SOZ has strong positive predictive value for postsurgical outcome, but poor negative predictive value.

SIGNIFICANCE

The uGIC is an objective surrogate marker of HFO activity independent of identifying discrete HFO events.

Quantitative analysis of intracranial electrocorticography signals using the concept of statistical parametric mapping

Statistical parametric mapping (SPM) is a technique with which one can delineate brain activity statistically deviated from the normative mean, and has been commonly employed in noninvasive neuroimaging and EEG studies. Using the concept of SPM, we developed a novel technique for quantification of the statistical deviation of an intracranial electrocorticography (ECoG) measure from the nonepileptic mean. We validated this technique using data previously collected from 123 patients with drug-resistant epilepsy who underwent resective epilepsy surgery. We determined how the measurement of statistical deviation of modulation index (MI) from the non-epileptic mean (rated by z-score) improved the performance of seizure outcome classification model solely based on conventional clinical, seizure onset zone (SOZ), and neuroimaging variables. Here, MI is a summary measure quantifying the strength of in-situ coupling between high-frequency activity at >150 Hz and slow wave at 3-4 Hz. We initially generated a normative MI atlas showing the mean and standard deviation of slow-wave sleep MI of neighboring non-epileptic channels of 47 patients, whose ECoG sampling involved all four lobes. We then calculated 'MI z-score' at each electrode site. SOZ had a greater 'MI z-score' compared to non-SOZ in the remaining 76 patients. Subsequent multivariate logistic regression analysis and receiver operating characteristic analysis to the combined data of all patients revealed that the full regression model incorporating all predictor variables, including SOZ and 'MI z-score', best classified the seizure outcome with sensitivity/specificity of 0.86/0.76. The model excluding 'MI z-score' worsened its sensitivity/specificity to 0.86/0.48. Furthermore, the leave-one-out analysis successfully cross-validated the full regression model. Measurement of statistical deviation of MI from the non-epileptic mean on invasive recording is technically feasible. Our analytical technique can be used to evaluate the utility of ECoG biomarkers in epilepsy presurgical evaluation.

Magnetoencephalography imaging of high frequency oscillations strengthens presurgical localization and outcome prediction

In patients with medically refractory epilepsy, resective surgery is the mainstay of therapy to achieve seizure freedom. However, ∼20-50% of cases have intractable seizures post-surgery due to the imprecise determination of epileptogenic zone. Recent intracranial studies suggest that high frequency oscillations between 80 and 200 Hz could serve as one of the consistent epileptogenicity biomarkers for localization of the epileptogenic zone. However, these high frequency oscillations are not adopted in the clinical setting because of difficult non-invasive detection. Here, we investigated non-invasive detection and localization of high frequency oscillations and its clinical utility in accurate pre-surgical assessment and post-surgical outcome prediction. We prospectively recruited 52 patients with medically refractory epilepsy who underwent standard pre-surgical workup including magnetoencephalography (MEG) followed by resective surgery after determination of the epileptogenic zone. The post-surgical outcome was assessed after 22.14 ± 10.05 months. Interictal epileptic spikes were expertly identified, and interictal epileptic oscillations across the neural activity frequency spectrum from 8 to 200 Hz were localized using adaptive spatial filtering methods. Localization results were compared with epileptogenic zone and resected cortex for congruence assessment and validated against the clinical outcome. The concordance rate of high frequency oscillations sources (80-200 Hz) with the presumed epileptogenic zone and the resected cortex were 75.0% and 78.8%, respectively, which is superior to that of other frequency bands and standard dipole fitting methods. High frequency oscillation sources corresponding with the resected cortex, had the best sensitivity of 78.0%, positive predictive value of 100% and an accuracy of 78.84% to predict the patient's surgical outcome, among all other frequency bands. If high frequency oscillation sources were spatially congruent with resected cortex, patients had an odds ratio of 5.67 and 82.4% probability of achieving a favourable surgical outcome. If high frequency oscillations sources were discordant with the epileptogenic zone or resection area, patient has an odds ratio of 0.18 and only 14.3% probability of achieving good outcome, and mostly tended to have an unfavourable outcome (χ2 = 5.22; P = 0.02; φ = -0.317). In receiver operating characteristic curve analyses, only sources of high-frequency oscillations demonstrated the best sensitivity and specificity profile in determining the patient's surgical outcome with area under the curve of 0.76, whereas other frequency bands indicate a poor predictive performance. Our study is the first non-invasive study to detect high frequency oscillations, address the efficacy of high frequency oscillations over the different neural oscillatory frequencies, localize them and clinically validate them with the post-surgical outcome in patients with medically refractory epilepsy. The evidence presented in the current study supports the fact that HFOs might significantly improve the presurgical assessment, and post-surgical outcome prediction, where it could widely be used in a clinical setting as a non-invasive biomarker.

Increased gamma and decreased fast ripple connections of epileptic tissue: A high-frequency directed network approach

OBJECTIVE

New insights into high-frequency electroencephalographic activity and network analysis provide potential tools to improve delineation of epileptic tissue and increase the chance of postoperative seizure freedom. Based on our observation of high-frequency oscillations "spreading outward" from the epileptic source, we hypothesize that measures of directed connectivity in the high-frequency range distinguish epileptic from healthy brain tissue.

METHODS

We retrospectively selected refractory epilepsy patients with a malformation of cortical development or tumor World Health Organization grade I/II who underwent epilepsy surgery with intraoperative electrocorticography for tailoring the resection based on spikes. We assessed directed functional connectivity in the theta (4-8 Hz), gamma (30-80 Hz), ripple (80-250 Hz), and fast ripple (FR; 250-500 Hz) bands using the short-time direct directed transfer function, and calculated the total, incoming, and outgoing propagation strength for each electrode. We compared network measures of electrodes covering the resected and nonresected areas separately for patients with good and poor outcome, and of electrodes with and without spikes, ripples, and FRs (group level: paired t test; patient level: Mann-Whitney U test). We selected the measure that could best identify the resected area and channels with epileptic events using the area under the receiver operating characteristic curve, and calculated the positive and negative predictive value, sensitivity, and specificity.

RESULTS

We found higher total and outstrength in the ripple and gamma bands in resected tissue in patients with good outcome (ripple_total : P = .01; ripple_out : P = .04; gamma_total : P = .01; gamma_out : P = .01). Channels with events showed lower total and instrength, and higher outstrength in the FR band, and higher total and outstrength in the ripple, gamma, and theta bands (FR_total : P = .05; FR_in : P < .01; FR_out : P = .02; gamma_total : P < .01; gamma_in : P = .01; gamma_out : P < .01; theta_total : P = .01; theta_out : P = .01). The total strength in the gamma band was most distinctive at the channel level (positive predictive value [PPV]_good  = 74%, PPV_poor  = 43%).

SIGNIFICANCE

Interictally, epileptic tissue is isolated in the FR band and acts as a driver up to the (fast) ripple frequency range. The gamma band total strength seems promising to delineate epileptic tissue intraoperatively.

A Magnetoencephalography Study of Pediatric Interictal Neuromagnetic Activity Changes and Brain Network Alterations Caused by Epilepsy in the High Frequency (80-1000 Hz)

More and more studies propose that high frequency brain signals are promising biomarkers of epileptogenic zone. In this paper, our aim is to investigate the neuromagnetic changes and brain network topological alterations during an interictal period at high frequency ranges (80-1000 Hz) between healthy controls and epileptic patients with Magnetoencephalography. We analyzed neuromagnetic activities with accumulated source imaging, and constructed brain network based on graph theory. Neuromagnetic activity changes and brain network alterations between two groups were analyzed in three frequency bands: ripple (80-250 Hz), fast ripples (FRs, 250-500 Hz), and very high frequency oscillations (VHFO, 500-1000 Hz). We found that epileptic patients showed significantly altered patterns of neuromagnetic source localization and altered brain network patterns. And, we also found that mean functional connectivity and the number of modules from epileptic patients significantly increased in the ripple and FRs bands, and mean clustering coefficient from epileptic patients significantly decreased in the ripple and FRs bands. We also found that the mean functional connectivity was positively correlated with duration of epilepsy in the ripple and VHFO bands, and the number of modules was positively correlated with the duration of epilepsy in the ripple, FRs, and VHFO bands. Our results indicate that epilepsy can alter patients' neuromagnetic activities and brain networks in the high-frequency ranges, and these alterations become more pathological as the duration of epilepsy grows longer.

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.

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.

Removing high-frequency oscillations: A prospective multicenter study on seizure outcome

OBJECTIVE

To evaluate the use of interictal high-frequency oscillations (HFOs) in epilepsy surgery for prediction of postsurgical seizure outcome in a prospective multicenter trial.

METHODS

We hypothesized that a seizure-free outcome could be expected in patients in whom the surgical planning included the majority of HFO-generating brain tissue while a poor seizure outcome could be expected in patients in whom only a few such areas were planned to be resected. Fifty-two patients were included from 3 tertiary epilepsy centers during a 1-year period. Ripples (80-250 Hz) and fast ripples (250-500 Hz) were automatically detected during slow-wave sleep with chronic intracranial EEG in 2 centers and acute intraoperative electrocorticography in 1 patient.

RESULTS

There was a correlation between the removal of HFO-generating regions and seizure-free outcome at the group level for all patients. No correlation was found, however, for the center-specific analysis, and an individual prognostication of seizure outcome was true in only 36 patients (67%). Moreover, some patients became seizure-free without removal of the majority of HFO-generating tissue. The investigation of influencing factors, including comparisons of visual and automatic analysis, using a threshold analysis for areas with high HFO activity, and excluding contacts bordering the resection, did not result in improved prognostication.

CONCLUSIONS

On an individual patient level, a prediction of outcome was not possible in all patients. This may be due to the analysis techniques used. Alternatively, HFOs may be less specific for epileptic tissue than earlier studies have indicated.

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

OBJECTIVE

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

APPROACH

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

MAIN RESULTS

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

SIGNIFICANCE

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

Generalizability of High Frequency Oscillation Evaluations in the Ripple Band

Objective: We examined the interrater reliability and generalizability of high-frequency oscillation (HFO) visual evaluations in the ripple (80–250 Hz) band, and established a framework for the transition of HFO analysis to routine clinical care. We were interested in the interrater reliability or epoch generalizability to describe how similar the evaluations were between reviewers, and in the reviewer generalizability to represent the consistency of the internal threshold each individual reviewer.

Methods: We studied 41 adult epilepsy patients (mean age: 35.6 years) who underwent intracranial electroencephalography. A morphology detector was designed and used to detect candidate HFO events, lower-threshold events, and distractor events. These events were subsequently presented to six expert reviewers, who visually evaluated events for the presence of HFOs. Generalizability theory was used to characterize the epoch generalizability (interrater reliability) and reviewer generalizability (internal threshold consistency) of visual evaluations, as well as to project the numbers of epochs, reviewers, and datasets required to achieve strong generalizability (threshold of 0.8).

Results: The reviewer generalizability was almost perfect (0.983), indicating there were sufficient evaluations to determine the internal threshold of each reviewer. However, the interrater reliability for 6 reviewers (0.588) and pairwise interrater reliability (0.322) were both poor, indicating that the agreement of 6 reviewers is insufficient to reliably establish the presence or absence of individual HFOs. Strong interrater reliability (≥0.8) was projected as requiring a minimum of 17 reviewers, while strong reviewer generalizability could be achieved with <30 epoch evaluations per reviewer.

Significance: This study reaffirms the poor reliability of using small numbers of reviewers to identify HFOs, and projects the number of reviewers required to overcome this limitation. It also provides a set of tools which may be used for training reviewers, tracking changes to interrater reliability, and for constructing a benchmark set of epochs that can serve as a generalizable gold standard, against which other HFO detection algorithms may be compared. This study represents an important step toward the reconciliation of important but discordant findings from HFO studies undertaken with different sets of HFOs, and ultimately toward transitioning HFO analysis into a meaningful part of the clinical epilepsy workup.

Using network analysis to localize the epileptogenic zone from invasive EEG recordings in intractable focal epilepsy

Treatment of medically intractable focal epilepsy (MIFE) by surgical resection of the epileptogenic zone (EZ) is often effective provided the EZ can be reliably identified. Even with the use of invasive recordings, the clinical differentiation between the EZ and normal brain areas can be quite challenging, mainly in patients without MRI detectable lesions. Consequently, despite relatively large brain regions being removed, surgical success rates barely reach 60–65%. Such variable and unfavorable outcomes associated with high morbidity rates are often caused by imprecise and/or inaccurate EZ localization. We developed a localization algorithm that uses network-based data analytics to process invasive EEG recordings. This network algorithm analyzes the centrality signatures of every contact electrode within the recording network and characterizes contacts into susceptible EZ based on the centrality trends over time. The algorithm was tested in a retrospective study that included 42 patients from four epilepsy centers. Our algorithm had higher agreement with EZ regions identified by clinicians for patients with successful surgical outcomes and less agreement for patients with failed outcomes. These findings suggest that network analytics and a network systems perspective of epilepsy may be useful in assisting clinicians in more accurately localizing the EZ.

Epilepsy is a disease that results in abnormal firing patterns in parts of the brain that comprise the epileptogenic network, known as the epileptogenic zone (EZ). Current methods to localize the EZ for surgical treatment often require observations of hundreds of thousands of EEG data points measured from many electrodes implanted in a patient’s brain. In this paper, we used network science to show that EZ regions may exhibit specific network signatures before, during, and after seizure events. Our algorithm computes the likelihood of each electrode being in the EZ and tends to agree more with clinicians during successful resections and less during failed surgeries. These results suggest that a networked analysis approach to EZ localization may be valuable in a clinical setting.

The role of high-frequency oscillations in epilepsy surgery planning

BACKGROUND

Epilepsy is a serious brain disorder characterized by recurrent unprovoked seizures. Approximately two-thirds of seizures can be controlled with antiepileptic medications (Kwan 2000). For some of the others, surgery can completely eliminate or significantly reduce the occurrence of disabling seizures. Localization of epileptogenic areas for resective surgery is far from perfect, and new tools are being investigated to more accurately localize the epileptogenic zone (the zone of the brain where the seizures begin) and improve the likelihood of freedom from postsurgical seizures. Recordings of pathological high-frequency oscillations (HFOs) may be one such tool.

OBJECTIVES

To assess the ability of HFOs to improve the outcomes of epilepsy surgery by helping to identify more accurately the epileptogenic areas of the brain.

SEARCH METHODS

For the latest update, we searched the Cochrane Epilepsy Group Specialized Register (25 July 2016), the Cochrane Central Register of Controlled Trials (CENTRAL) via the Cochrane Register of Studies Online (CRSO, 25 July 2016), MEDLINE (Ovid, 1946 to 25 July 2016), CINAHL Plus (EBSCOhost, 25 July 2016), Web of Science (Thomson Reuters, 25 July 2016), ClinicalTrials.gov (25 July 2016), and the World Health Organization International Clinical Trials Registry Platform ICTRP (25 July 2016).

SELECTION CRITERIA

We included studies that provided information on the outcomes of epilepsy surgery for at least six months and which used high-frequency oscillations in making decisions about epilepsy surgery.

DATA COLLECTION AND ANALYSIS

The primary outcome of the review was the Engel Class Outcome System (class I = no disabling seizures, II = rare disabling seizures, III = worthwhile improvement, IV = no worthwhile improvement). Secondary outcomes were responder rate, International League Against Epilepsy (ILAE) epilepsy surgery outcome, frequency of adverse events from any source and quality of life outcomes. We intended to analyse outcomes via an aggregated data fixed-effect model meta-analysis.

MAIN RESULTS

Two studies representing 11 participants met the inclusion criteria. Both studies were small non-randomised trials, with no control group and no blinding. The quality of evidence for all outcomes was very low. The combination of these two studies resulted in 11 participants who prospectively used ictal HFOs for epilepsy surgery decision making. Results of the postsurgical seizure freedom Engel class I to IV outcome were determined over a period of 12 to 38 months (average 23.4 months) and indicated that six participants had an Engel class I outcome (seizure freedom), two had class II (rare disabling seizures), three had class III (worthwhile improvement). No adverse effects were reported. Neither study compared surgical results guided by HFOs versus surgical results guided without HFOs.

AUTHORS' CONCLUSIONS

No reliable conclusions can be drawn regarding the efficacy of using HFOs in epilepsy surgery decision making at present.

Tailoring epilepsy surgery with fast ripples in the intraoperative electrocorticogram

OBJECTIVE

Intraoperative electrocorticography (ECoG) can be used to delineate the resection area in epilepsy surgery. High-frequency oscillations (HFOs; 80-500 Hz) seem better biomarkers for epileptogenic tissue than spikes. We studied how HFOs and spikes in combined pre- and postresection ECoG predict surgical outcome in different tailoring approaches.

METHODS

We, retrospectively, marked HFOs, divided into fast ripples (FRs; 250-500 Hz) and ripples (80-250 Hz), and spikes in pre- and postresection ECoG sampled at 2,048 Hz in people with refractory focal epilepsy. We defined four groups of electroencephalography (EEG) event occurrence: pre+post- (+/-), pre+post+ (+/+), pre-post+ (-/+) and pre-post- (-/-). We subcategorized three tailoring approaches: hippocampectomy with tailoring for neocortical involvement; lesionectomy of temporal lesions with tailoring for mesiotemporal involvement; and lesionectomy with tailoring for surrounding neocortical involvement. We compared the percentage of resected pre-EEG events, time to recurrence, and the different tailoring approaches to outcome (seizure-free vs recurrence).

RESULTS

We included 54 patients (median age, 15.5 years; 25 months of follow-up; 30 seizure free). The percentage of resected FRs, ripples, or spikes in pre-ECoG did not predict outcome. The occurrence of FRs in post-ECoG, given FRs in pre-ECoG (+/-, +/+), predicted outcome (hazard ratio, 3.13; confidence interval = 1.22-6.25; p = 0.01). Seven of 8 patients without spikes in pre-ECoG were seizure free. The highest predictive value for seizure recurrence was presence of FRs in post-ECoG for all tailoring approaches.

INTERPRETATION

FRs that persist before and after resection predict poor postsurgical outcome. These findings hold for different tailoring approaches. FRs can thus be used for tailoring epilepsy surgery with repeated intraoperative ECoG measurements. Ann Neurol 2017;81:664-676.

High-frequency oscillations: The state of clinical research

Modern electroencephalographic (EEG) technology contributed to the appreciation that the EEG signal outside the classical Berger frequency band contains important information. In epilepsy, research of the past decade focused particularly on interictal high-frequency oscillations (HFOs) > 80 Hz. The first large application of HFOs was in the context of epilepsy surgery. This is now followed by other applications such as assessment of epilepsy severity and monitoring of antiepileptic therapy. This article reviews the evidence on the clinical use of HFOs in epilepsy with an emphasis on the latest developments. It highlights the growing literature on the association between HFOs and postsurgical seizure outcome. A recent meta-analysis confirmed a higher resection ratio for HFOs in seizure-free versus non-seizure-free patients. Residual HFOs in the postoperative electrocorticogram were shown to predict epilepsy surgery outcome better than preoperative HFO rates. The review further discusses the different attempts to separate physiological from epileptic HFOs, as this might increase the specificity of HFOs. As an example, analysis of sleep microstructure demonstrated a different coupling between HFOs inside and outside the epileptogenic zone. Moreover, there is increasing evidence that HFOs are useful to measure disease activity and assess treatment response using noninvasive EEG and magnetoencephalography. This approach is particularly promising in children, because they show high scalp HFO rates. HFO rates in West syndrome decrease after adrenocorticotropic hormone treatment. Presence of HFOs at the time of rolandic spikes correlates with seizure frequency. The time-consuming visual assessment of HFOs, which prevented their clinical application in the past, is now overcome by validated computer-assisted algorithms. HFO research has considerably advanced over the past decade, and use of noninvasive methods will make HFOs accessible to large numbers of patients. Prospective multicenter trials are awaited to gather information over long recording periods in large patient samples.

The glutamate receptor antagonists CNQX and MPEP decrease fast ripple events in rats treated with kainic acid

Fast ripples (FR) are high frequency oscillations (250-600Hz) that have been associated with epilepsy. FR are assumed to be generated in small areas of the hippocampus (1mm³) that contain pathologically interconnected glutamate pyramidal cell clusters. Additionally, a relation between glutamate neurotransmission and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid/kainite (AMPA/KA) and metabotropic mGluR5 receptors is well established. Therefore, we hypothesized that antagonism of these glutamate receptors would decrease FR activity. For this propose, we induced status epilepticus with a kainic acid injection in the posterior right hippocampus and performed intracranial EEG recordings to detect and evaluate the presence of FR 15days after the injection. The glutamate AMPA/KA receptor antagonist CNQX (10mg/kg) and the mGluR5 antagonist MPEP (20mg/kg) were administered intraperitoneally, and the effects of the drugs were evaluated for a period of three hours after their administration. The results show a decrease in the number of FR in the first hour after drug administration in both cases (CNQX, p=0.0125; MPEP, p=0.0132) and a return to basal values in the third hour of the experiment, but not significant differences in the number of oscillations per event of FR, and the frequency and duration of each event of FR. We therefore conclude that blockade of AMPA/KA and mGluR5 receptors transiently decreases the generation of FR; however, the mechanisms by which this effect is achieved are to be further analyzed in future experiments.

How to record high-frequency oscillations in epilepsy: A practical guideline

OBJECTIVE

Technology for localizing epileptogenic brain regions plays a central role in surgical planning. Recent improvements in acquisition and electrode technology have revealed that high-frequency oscillations (HFOs) within the 80-500 Hz frequency range provide the neurophysiologist with new information about the extent of the epileptogenic tissue in addition to ictal and interictal lower frequency events. Nevertheless, two decades after their discovery there remain questions about HFOs as biomarkers of epileptogenic brain and there use in clinical practice.

METHODS

In this review, we provide practical, technical guidance for epileptologists and clinical researchers on recording, evaluation, and interpretation of ripples, fast ripples, and very high-frequency oscillations.

RESULTS

We emphasize the importance of low noise recording to minimize artifacts. HFO analysis, either visual or with automatic detection methods, of high fidelity recordings can still be challenging because of various artifacts including muscle, movement, and filtering. Magnetoencephalography and intracranial electroencephalography (iEEG) recordings are subject to the same artifacts.

SIGNIFICANCE

High-frequency oscillations are promising new biomarkers in epilepsy. This review provides interested researchers and clinicians with a review of current state of the art of recording and identification and potential challenges to clinical translation.

Intraoperative fast ripples independently predict postsurgical epilepsy outcome: Comparison with other electrocorticographic phenomena

In the surgical management of epilepsy, the resection of cortex exhibiting interictal fast ripples (250-500Hz) on electrocorticography has been linked to postoperative seizure-freedom. Although fast ripples appear to accurately identify the epileptogenic zone-the minimum tissue that must be removed at surgery to achieve seizure-freedom-it has not been established that fast ripples are a superior biomarker in comparison with multimodal presurgical neuroimaging and other electrocorticography abnormalities. Hence, in the prediction of postoperative seizure-freedom, we compared the value of fast ripples with other intraoperative electocorticography abnormalities including focal slowing, paroxysmal fast activity, intermittent spike discharges, continuous epileptiform discharges, focal attenuation, and intraoperative seizures, as well as complete resection of the lesion defined by MRI and other neuroimaging. In a cohort of 60 children with lesional epilepsy and median postsurgical follow-up exceeding 4 years, who underwent resective epilepsy surgery with intraoperative electrocorticography, we evaluated the extent to which removal of each intraoperative electrocorticography abnormality impacts time to first postoperative seizure using the Kaplan-Meier method and Cox proportional hazards regression. Secondly, we contrasted the predictive value of resection of each competing electrocorticography abnormality using standard test metrics (sensitivity, specificity, positive predictive value, and negative predictive value). In contrast with all other intraoperative electrocorticography abnormalities, fast ripples demonstrated the most favorable combination of positive predictive value (100%) and negative predictive value (76%) in the prediction of postoperative seizures. Among all candidate electrocorticography features, time to first postoperative seizure was most strongly associated with incomplete resection of fast ripples (hazard ratio=19.8, p<0.001). In multivariate survival analyses, postoperative seizures were independently predicted by incomplete resection of cortex generating fast ripples (hazard ratio=25.4, 95%CI 6.71-96.0, p<0.001) and focal slowing (hazard ratio=5.79, 95%CI 1.76-19.0, p=0.004), even after adjustment for the impact of an otherwise complete resection. All children with incomplete resection of interictal FR-generating cortex exhibited postoperative seizures within six months. Notably, this cohort included many patients with large resections and thus limited opportunity to exhibit unresected fast ripples. Future study in a cohort with small resection volume, or a clinical trial in which resection margins are guided by fast ripple distribution, would likely yield a more precise estimate of the risk posed by unresected fast ripples. With a high detection rate during brief intraoperative electrocorticography and favorable positive and negative predictive value, interictal fast ripple characterization during surgery is a feasible and useful adjunct to standard methods for epilepsy surgery planning, and represents a valuable spatially-localizing biomarker of the epileptogenic zone, without the need for prolonged extraoperative electrocorticography.

Prediction of seizure outcome improved by fast ripples detected in low-noise intraoperative corticogram

OBJECTIVE

Fast ripples (FR, 250-500Hz) in the intraoperative corticogram have recently been proposed as specific predictors of surgical outcome in epilepsy patients. However, online FR detection is restricted by their low signal-to-noise ratio. Here we propose the integration of low-noise EEG with unsupervised FR detection.

METHODS

Pre- and post-resection ECoG (N=9 patients) was simultaneously recorded by a commercial device (CD) and by a custom-made low-noise amplifier (LNA). FR were analyzed by an automated detector previously validated on visual markings in a different dataset.

RESULTS

Across all recordings, in the FR band the background noise was lower in LNA than in CD (p<0.001). FR rates were higher in LNA than CD recordings (0.9±1.4 vs 0.4±0.9, p<0.001). Comparison between FR rates in post-resection ECoG and surgery outcome resulted in positive predictive value PPV=100% in CD and LNA, and negative predictive value NPV=38% in CD and NPV=50% for LNA. Prediction accuracy was 44% for CD and 67% for LNA.

CONCLUSIONS

Prediction of seizure outcome was improved by the optimal integration of low-noise EEG and unsupervised FR detection.

SIGNIFICANCE

Accurate, automated and fast FR rating is essential for consideration of FR in the intraoperative setting.