Functional connectivity estimated from intracranial EEG predicts surgical outcome in intractable temporal lobe epilepsy

Increased coherence predicts medical refractoriness in patients with temporal lobe epilepsy on monotherapy

Among patients with epilepsy, 30-40% experience recurrent seizures even after adequate antiseizure medications therapies, making them refractory. The early identification of refractory epilepsy is important to provide timely surgical treatment for these patients. In this study, we analyze interictal electroencephalography (EEG) data to predict drug refractoriness in patients with temporal lobe epilepsy (TLE) who were treated with monotherapy at the time of the first EEG acquisition. Various EEG features were extracted, including statistical measurements and interchannel coherence. Feature selection was performed to identify the optimal features, and classification was conducted using different classifiers. Functional connectivity and graph theory measurements were calculated to identify characteristics of refractory TLE. Among the 48 participants, 34 (70.8%) were responsive, while 14 (29.2%) were refractory over a mean follow-up duration of 38.5 months. Coherence feature within the gamma frequency band exhibited the most favorable performance. The light gradient boosting model, employing the mutual information filter-based feature selection method, demonstrated the highest performance (AUROC = 0.821). Compared to the responsive group, interchannel coherence displayed higher values in the refractory group. Interestingly, graph theory measurements using EEG coherence exhibited higher values in the refractory group than in the responsive group. Our study has demonstrated a promising method for the early identification of refractory TLE utilizing machine learning algorithms.

The interictal suppression hypothesis is the dominant differentiator of seizure onset zones in focal epilepsy

Successful surgical treatment of drug-resistant epilepsy traditionally relies on the identification of seizure onset zones (SOZs). Connectome-based analyses of electrographic data from stereo electroencephalography (SEEG) may empower improved detection of SOZs. Specifically, connectome-based analyses based on the interictal suppression hypothesis posit that when the patient is not having a seizure, SOZs are inhibited by non-SOZs through high inward connectivity and low outward connectivity. However, it is not clear whether there are other motifs that can better identify potential SOZs. Thus, we sought to use unsupervised machine learning to identify network motifs that elucidate SOZs and investigate if there is another motif that outperforms the ISH. Resting-state SEEG data from 81 patients with drug-resistant epilepsy undergoing a pre-surgical evaluation at Vanderbilt University Medical Center were collected. Directed connectivity matrices were computed using the alpha band (8-13 Hz). Principal component analysis (PCA) was performed on each patient's connectivity matrix. Each patient's components were analysed qualitatively to identify common patterns across patients. A quantitative definition was then used to identify the component that most closely matched the observed pattern in each patient. A motif characteristic of the interictal suppression hypothesis (high-inward and low-outward connectivity) was present in all individuals and found to be the most robust motif for identification of SOZs in 64/81 (79%) patients. This principal component demonstrated significant differences in SOZs compared to non-SOZs. While other motifs for identifying SOZs were present in other patients, they differed for each patient, suggesting that seizure networks are patient specific, but the ISH is present in nearly all networks. We discovered that a potentially suppressive motif based on the interictal suppression hypothesis was present in all patients, and it was the most robust motif for SOZs in 79% of patients. Each patient had additional motifs that further characterized SOZs, but these motifs were not common across all patients. This work has the potential to augment clinical identification of SOZs to improve epilepsy treatment.

Gamma amplitude-envelope correlations are strongly elevated within hyperexcitable networks in focal epilepsy

Methods to quantify cortical hyperexcitability are of enormous interest for mapping epileptic networks in patients with focal epilepsy. We hypothesize that, in the resting state, cortical hyperexcitability increases firing-rate correlations between neuronal populations within seizure onset zones (SOZs). This hypothesis predicts that in the gamma frequency band (40-200 Hz), amplitude envelope correlations (AECs), a relatively straightforward measure of functional connectivity, should be elevated within SOZs compared to other areas. To test this prediction, we analyzed archived samples of interictal electrocorticographic (ECoG) signals recorded from patients who became seizure-free after surgery targeting SOZs identified by multiday intracranial recordings. We show that in the gamma band, AECs between nodes within SOZs are markedly elevated relative to those elsewhere. AEC-based node strength, eigencentrality, and clustering coefficient are also robustly increased within the SOZ with maxima in the low-gamma band (permutation test Z-scores > 8) and yield moderate discriminability of the SOZ using ROC analysis (maximal mean AUC ~ 0.73). By contrast to AECs, phase locking values (PLVs), a measure of narrow-band phase coupling across sites, and PLV-based graph metrics discriminate the seizure onset nodes weakly. Our results suggest that gamma band AECs may provide a clinically useful marker of cortical hyperexcitability in focal epilepsy.

Centromedian region thalamic responsive neurostimulation mitigates idiopathic generalized and multifocal epilepsy with focal to bilateral tonic-clonic seizures

OBJECTIVE

Although >30% of epilepsy patients have drug-resistant epilepsy (DRE), typically those with generalized or multifocal disease have not traditionally been considered surgical candidates. Responsive neurostimulation (RNS) of the centromedian (CM) region of the thalamus now appears to be a promising therapeutic option for this patient population. We present outcomes following CM RNS for 13 patients with idiopathic generalized epilepsy (IGE) and eight with multifocal onsets that rapidly generalize to bilateral tonic-clonic (focal to bilateral tonic-clonic [FBTC]) seizures.

METHODS

A retrospective review of all patients undergoing bilateral CM RNS by the senior author through July 2022 were reviewed. Electrodes were localized and volumes of tissue activation were modeled in Lead-DBS. Changes in patient seizure frequency were extracted from electronic medical records.

RESULTS

Twenty-one patients with DRE underwent bilateral CM RNS implantation. For 17 patients with at least 1 year of postimplantation follow-up, average seizure reduction from preoperative baseline was 82.6% (SD = 19.0%, median = 91.7%), with 18% of patients Engel class 1, 29% Engel class 2, 53% Engel class 3, and 0% Engel class 4. There was a trend for average seizure reduction to be greater for patients with nonlesional FBTC seizures than for other patients. For patients achieving at least Engel class 3 outcome, median time to worthwhile seizure reduction was 203.5 days (interquartile range = 110.5-343.75 days). Patients with IGE with myoclonic seizures had a significantly shorter time to worthwhile seizure reduction than other patients. The surgical targeting strategy evolved after the first four subjects to achieve greater anatomic accuracy.

SIGNIFICANCE

Patients with both primary and rapidly generalized epilepsy who underwent CM RNS experienced substantial seizure relief. Subsets of these patient populations may particularly benefit from CM RNS. The refinement of lead targeting, tuning of RNS system parameters, and patient selection are ongoing areas of investigation.

Epilepsy lesion localization method based on brain function network

Objective

In the past, the localization of seizure onset zone (SOZ) primarily relied on traditional EEG signal analysis methods. However, due to their limited spatial and temporal resolution, accurately pinpointing neural activity was challenging, thereby restricting their clinical applicability. Compared with traditional EEG signals, SEEG signals have superior spatial and temporal resolution, and can more accurately record neural activity near epileptic foci, making them better suited for studying SOZ. In addition, the traditional EEG signal analysis methods still have limitations, mainly focusing on the analysis of local signal features, while ignoring the complexity and interconnection of the overall brain network. How to more accurately locate SOZ is still not well resolved. The purpose of this study is to develop an effective positioning method for more accurate positioning.

Method

To overcome these limitations, this study proposed a model integrating brain functional network analysis with nonlinear dynamics. We utilized weighted phase lag index (WPLI) to construct brain functional network, epilepic network connectivity strength (ENCS) as the feature, and introduced persistence entropy (PE) for feature fusion, subsequently employing support vector machine (SVM) classification.

Results

The proposed method was verified on the HUP-iEEG dataset, our solution identified the SOZ with 0.9440 accuracy, 0.9848 precision, 0.8974 recall rate, 0.9340 F1 score and 0.9697 area under the ROC curve across patients, which outperforms the existing approaches. It exhibits a 2.30 percentage point enhancement in localisation accuracy along with a 2.97 percentage points in AUC compared to others.

Conclusion

Our method consider the interactions between nodes in brain network connections, as well as the inherent nonlinear and non-stationary properties of neural signals, to be more robust.

Evaluation of Utility of Invasive Electroencephalography for Definitive Surgery in Patients with Drug-Resistant Epilepsy: A Systematic Review and Meta-Analysis

When noninvasive tests are unable to define the epileptogenic zone in patients, intracranial electroencephalography (iEEG) is a method of localizing the epileptogenic zone. Compared with noninvasive evaluations, it offers more precise information about patterns of epileptiform activity, which results in useful diagnostic information that supports surgical decision-making. The primary aim of the present study was to assess the utility of iEEG for definitive surgery for patients with drug-resistant epilepsy. Online databases such as PubMed, Medline, Embase, Scopus, Cochrane Library, Web of Science, and IEEE Xplore were searched for MeSH terms and free-text keywords. The ROBINS I (risk of bias in non-randomized studies - of interventions) critical appraisal tool was used for quality assessment. The prevalence from different studies was pooled together using the inverse variance heterogeneity method. Egger's regression analysis and funnel plot were used to evaluate publication bias. The systematic review included 18 studies, and the meta-analysis included 10 studies to estimate the prevalence of seizure freedom (Engel class I) in patients undergoing surgery after iEEG. A total of 526 patients were included in the meta-analysis. The follow-up period ranged from 1 to 10 years. The overall pooled estimate of the prevalence of seizure freedom (Engel class I) for patients undergoing surgery after iEEG was 53% (95% confidence interval, 44%-62%). The results additionally demonstrated that 12 studies had a moderate risk of bias and 6 had a low risk. Future studies are crucial to enhance our understanding of iEEG to guide patient choices and unravel their implications.

Predictive models of epilepsy outcomes

PURPOSE OF REVIEW

Multiple complex medical decisions are necessary in the course of a chronic disease like epilepsy. Predictive tools to assist physicians and patients in navigating this complexity have emerged as a necessity and are summarized in this review.

RECENT FINDINGS

Nomograms and online risk calculators are user-friendly and offer individualized predictions for outcomes ranging from safety of antiseizure medication withdrawal (accuracy 65-73%) to seizure-freedom, naming, mood, and language outcomes of resective epilepsy surgery (accuracy 72-81%). Improving their predictive performance is limited by the nomograms' inability to ingest complex data inputs. Conversely, machine learning offers the potential of multimodal and expansive model inputs achieving human-expert level accuracy in automated scalp electroencephalogram (EEG) interpretation but lagging in predictive performance or requiring validation for other applications.

SUMMARY

Good to excellent predictive models are now available to guide medical and surgical epilepsy decision-making with nomograms offering individualized predictions and user-friendly tools, and machine learning approaches offering the potential of improved performance. Future research is necessary to bridge the two approaches for optimal translation to clinical care.

High-performance prediction of epilepsy surgical outcomes based on the genetic neural networks and hybrid iEEG marker

Accurately identification of the seizure onset zone (SOZ) is pivotal for successful surgery in patients with medically refractory epilepsy. The purpose of this study is to improve the performance of model predicting the epilepsy surgery outcomes using genetic neural network (GNN) model based on a hybrid intracranial electroencephalography (iEEG) marker. We extracted 21 SOZ related markers based on iEEG data from 79 epilepsy patients. The least absolute shrinkage and selection operator (LASSO) regression was employed to integrated seven markers, selected after testing in pairs with all 21 biomarkers and 7 machine learning models, into a hybrid marker. Based on the hybrid marker, we devised a GNN model and compared its predictive performance for surgical outcomes with six other mainstream machine-learning models. Compared to the mainstream models, underpinning the GNN with the hybrid iEEG marker resulted in a better prediction of surgical outcomes, showing a significant increase of the prediction accuracy from approximately 87% to 94.3% (P = 0.0412). This study suggests that the hybrid iEEG marker can improve the performance of model predicting the epilepsy surgical outcomes, and validates the effectiveness of the GNN in characterizing and analyzing complex relationships between clinical data variables.

Metabolic connectivity as a predictor of surgical outcome in mesial temporal lobe epilepsy

OBJECTIVE

The study investigated metabolic connectivity (MC) differences between patients with unilateral drug-resistant mesial temporal lobe epilepsy (MTLE) with hippocampal sclerosis (HS) and healthy controls (HCs), based on [18 F]-fluorodeoxyglucose (FDG)-PET data. We focused on the MC changes dependent on the lateralization of the epileptogenic lobe and on correlations with postoperative outcomes.

METHODS

FDG-PET scans of 47 patients with unilateral MTLE with histopathologically proven HS and 25 HC were included in the study. All the patients underwent a standard anterior temporal lobectomy and were more than 2 years after the surgery. MC changes were compared between the two HS groups (left HS, right HS) and HC. Differences between the metabolic network of seizure-free and non-seizure-free patients after surgery were depicted afterward. Network changes were correlated with clinical characteristics.

RESULTS

The study showed widespread metabolic network changes in the HS patients as compared to HC. The changes were more extensive in the right HS than in the left HS. Unfavorable surgical outcomes were found in patients with decreased MC within the network including both the lesional and contralesional hippocampus, ipsilesional frontal operculum, and contralesional insula. Favorable outcomes correlated with decreased MC within the network involving both orbitofrontal cortices and the ipsilesional temporal lobe.

SIGNIFICANCE

There are major differences in the metabolic networks of left and right HS, with more extensive changes in right HS. The changes within the metabolic network could help predict surgical outcomes in patients with HS. MC may identify patients with potentially unfavorable outcomes and direct them to a more detailed presurgical evaluation.

PLAIN LANGUAGE SUMMARY

Metabolic connectivity is a promising method for metabolic network mapping. Metabolic networks in mesial temporal lobe epilepsy are dependent on lateralization of the epileptogenic lobe and could predict surgical outcomes.

Profiles of resting state functional connectivity in temporal lobe epilepsy associated with post-laser interstitial thermal therapy seizure outcomes and semiologies

Introduction

It is now understood that in focal epilepsy, impacted neural regions are not limited to the epileptogenic zone. As such, further investigation into the underlying functional connectivity (FC) patterns in those enduring Temporal Lobe Epilepsy (TLE) with Mesial Temporal Sclerosis (MTS) is imperative to understanding the intricacies of the disease.

Methods

The rsfMRIs of 17 healthy participants, 10 left-sided TLE-MTS patients with a pre-operative history of focal impaired awareness seizures (FIA), and 13 left-sided TLE-MTS patients with a pre-operative history of focal aware seizures (FA) were compared to determine the existence of distinct FC patterns with respect to seizure types. Similarly, the rsfMRIs of the above-mentioned healthy participants, 16 left-sided TLE-MTS individuals who were seizure-free (SF) 12 months postoperatively, and 16 left-sided TLE-MTS persons without seizure freedom (nSF) were interrogated. The ROI-to-ROI connectivity analysis included a total of 175 regions of interest (ROIs) and accounted for both age and duration of epileptic activity. Significant correlations were determined via two-sample t-tests and Bonferroni correction (α = 0.05).

Results

Comparisons of FA and FIA groups depicted significant correlations between the contralateral anterior cingulate gyrus, subgenual region, and the contralateral cerebellum, lobule III (p-value = 2.26e-4, mean z-score = -0.05 ± 0.28, T = -4.23). Comparisons of SF with nSF depicted two significantly paired-ROIs; the contralateral amygdala and the contralateral precuneus (p-value = 2.9e-5, mean z-score = -0.12 ± 0.19, T = 4.98), as well as the contralateral locus coeruleus and the ipsilateral intralaminar nucleus (p-value= 1.37e-4, mean z-score = 0.06 ± 0.17, T = -4.41).

Significance

FC analysis proves to be a lucrative modality for exploring unique signatures with respect to seizure types and postoperative outcomes. By furthering our understanding of the differences between epileptic phenotypes, we can achieve improvement in future treatment modalities not limited to targeting advancements.

Spike propagation mapping reveals effective connectivity and predicts surgical outcome in epilepsy

Neurosurgical intervention is the best available treatment for selected patients with drug resistant epilepsy. For these patients, surgical planning requires biomarkers that delineate the epileptogenic zone, the brain area that is indispensable for the generation of seizures. Interictal spikes recorded with electrophysiological techniques are considered key biomarkers of epilepsy. Yet, they lack specificity, mostly because they propagate across brain areas forming networks. Understanding the relationship between interictal spike propagation and functional connections among the involved brain areas may help develop novel biomarkers that can delineate the epileptogenic zone with high precision. Here, we reveal the relationship between spike propagation and effective connectivity among onset and areas of spread and assess the prognostic value of resecting these areas. We analysed intracranial EEG data from 43 children with drug resistant epilepsy who underwent invasive monitoring for neurosurgical planning. Using electric source imaging, we mapped spike propagation in the source domain and identified three zones: onset, early-spread and late-spread. For each zone, we calculated the overlap and distance from surgical resection. We then estimated a virtual sensor for each zone and the direction of information flow among them via Granger causality. Finally, we compared the prognostic value of resecting these zones, the clinically-defined seizure onset zone and the spike onset on intracranial EEG channels by estimating their overlap with resection. We observed a spike propagation in source space for 37 patients with a median duration of 95 ms (interquartile range: 34-206), a spatial displacement of 14 cm (7.5-22 cm) and a velocity of 0.5 m/s (0.3-0.8 m/s). In patients with good surgical outcome (25 patients, Engel I), the onset had higher overlap with resection [96% (40-100%)] than early-spread [86% (34-100%), P = 0.01] and late-spread [59% (12-100%), P = 0.002], and it was also closer to resection than late-spread [5 mm versus 9 mm, P = 0.007]. We found an information flow from onset to early-spread in 66% of patients with good outcomes, and from early-spread to onset in 50% of patients with poor outcome. Finally, resection of spike onset, but not area of spike spread or the seizure onset zone, predicted outcome with positive predictive value of 79% and negative predictive value of 56% (P = 0.04). Spatiotemporal mapping of spike propagation reveals information flow from onset to areas of spread in epilepsy brain. Surgical resection of the spike onset disrupts the epileptogenic network and may render patients with drug resistant epilepsy seizure-free without having to wait for a seizure to occur during intracranial monitoring.

Predicting seizure outcome after epilepsy surgery: Do we need more complex models, larger samples, or better data?

OBJECTIVE

The accurate prediction of seizure freedom after epilepsy surgery remains challenging. We investigated if (1) training more complex models, (2) recruiting larger sample sizes, or (3) using data-driven selection of clinical predictors would improve our ability to predict postoperative seizure outcome using clinical features. We also conducted the first substantial external validation of a machine learning model trained to predict postoperative seizure outcome.

METHODS

We performed a retrospective cohort study of 797 children who had undergone resective or disconnective epilepsy surgery at a tertiary center. We extracted patient information from medical records and trained three models-a logistic regression, a multilayer perceptron, and an XGBoost model-to predict 1-year postoperative seizure outcome on our data set. We evaluated the performance of a recently published XGBoost model on the same patients. We further investigated the impact of sample size on model performance, using learning curve analysis to estimate performance at samples up to N = 2000. Finally, we examined the impact of predictor selection on model performance.

RESULTS

Our logistic regression achieved an accuracy of 72% (95% confidence interval [CI] = 68%-75%, area under the curve [AUC] = .72), whereas our multilayer perceptron and XGBoost both achieved accuracies of 71% (95% CIMLP  = 67%-74%, AUCMLP  = .70; 95% CIXGBoost own  = 68%-75%, AUCXGBoost own  = .70). There was no significant difference in performance between our three models (all p > .4) and they all performed better than the external XGBoost, which achieved an accuracy of 63% (95% CI = 59%-67%, AUC = .62; pLR  = .005, pMLP  = .01, pXGBoost own  = .01) on our data. All models showed improved performance with increasing sample size, but limited improvements beyond our current sample. The best model performance was achieved with data-driven feature selection.

SIGNIFICANCE

We show that neither the deployment of complex machine learning models nor the assembly of thousands of patients alone is likely to generate significant improvements in our ability to predict postoperative seizure freedom. We instead propose that improved feature selection alongside collaboration, data standardization, and model sharing is required to advance the field.

Artificial Intelligence in Neurosurgery: A State-of-the-Art Review from Past to Future

In recent years, there has been a significant surge in discussions surrounding artificial intelligence (AI), along with a corresponding increase in its practical applications in various facets of everyday life, including the medical industry. Notably, even in the highly specialized realm of neurosurgery, AI has been utilized for differential diagnosis, pre-operative evaluation, and improving surgical precision. Many of these applications have begun to mitigate risks of intraoperative and postoperative complications and post-operative care. This article aims to present an overview of the principal published papers on the significant themes of tumor, spine, epilepsy, and vascular issues, wherein AI has been applied to assess its potential applications within neurosurgery. The method involved identifying high-cited seminal papers using PubMed and Google Scholar, conducting a comprehensive review of various study types, and summarizing machine learning applications to enhance understanding among clinicians for future utilization. Recent studies demonstrate that machine learning (ML) holds significant potential in neuro-oncological care, spine surgery, epilepsy management, and other neurosurgical applications. ML techniques have proven effective in tumor identification, surgical outcomes prediction, seizure outcome prediction, aneurysm prediction, and more, highlighting its broad impact and potential in improving patient management and outcomes in neurosurgery. This review will encompass the current state of research, as well as predictions for the future of AI within neurosurgery.

New insights for predicting surgery outcome in patients with temporal lobe epilepsy. A systematic review

Resective surgery is the treatment of choice for one-third of adult patients with focal, drug-resistant epilepsy. This procedure is associated with substantial clinical and cognitive risks. In clinical practice, there is no validated model for epilepsy surgery outcome prediction (ESOP). Meta-analyses on ESOP studies assessing prognostic factors report discrepancies in terms of study design. Our review aims to systematically investigate methodological and analytical aspects of studies predicting clinical and cognitive outcomes after temporal lobe epilepsy surgery. A systematic review of ESOP studies published between 2000 and 2022 from three databases (MEDLINE, Web of Science, and PsycINFO) was completed by following the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines. It yielded 4867 articles. Among them, 21 corresponded to our inclusion criteria and were therefore retained in the final review. The risk of bias was assessed using A Tool to Assess Risk of Bias and Applicability of Prediction Model Studies (PROBAST). Data extracted from the 21 studies were analyzed using narrative synthesis and descriptive statistics. Our findings show an increase in the use of multimodal datasets and machine learning analyses in recent ESOP studies, although regression remained the most frequently used approach. We also identified a more frequent use of network notions in recent ESOP studies. Nevertheless, several methodological issues were noted, such as small sample sizes, lack of information on the follow-up period, variability in seizure outcome, and the definition of neuropsychological postoperative change. Of 21 studies, only one provided a clinical tool to anticipate the cognitive outcome after epilepsy surgery. We conclude that methodological issues should be overcome before we move towards more complete models to better predict clinical and cognitive outcomes after epilepsy surgery. Recommendations for future studies to harness the possibilities of multimodal datasets and data fusion, are provided. A stronger bridge between fundamental and clinical research may result in developing accessible clinical tools.

Pre-surgical features of intrinsic brain networks predict single and joint epilepsy surgery outcomes

Despite the effectiveness of surgical interventions for the treatment of intractable focal temporal lobe epilepsy (TLE), the substrates that support good outcomes are poorly understood. While algorithms have been developed for the prediction of either seizure or cognitive/psychiatric outcomes alone, no study has reported on the functional and structural architecture that supports joint outcomes. We measured key aspects of pre-surgical whole brain functional/structural network architecture and evaluated their ability to predict post-operative seizure control in combination with cognitive/psychiatric outcomes. Pre-surgically, we identified the intrinsic connectivity networks (ICNs) unique to each person through independent component analysis (ICA), and computed: (1) the spatial-temporal match between each person's ICA components and established, canonical ICNs, (2) the connectivity strength within each identified person-specific ICN, (3) the gray matter (GM) volume underlying the person-specific ICNs, and (4) the amount of variance not explained by the canonical ICNs for each person. Post-surgical seizure control and reliable change indices of change (for language [naming, phonemic fluency], verbal episodic memory, and depression) served as binary outcome responses in random forest (RF) models. The above functional and structural measures served as input predictors. Our empirically derived ICN-based measures customized to the individual showed that good joint seizure and cognitive/psychiatric outcomes depended upon higher levels of brain reserve (GM volume) in specific networks. In contrast, singular outcomes relied on systematic, idiosyncratic variance in the case of seizure control, and the weakened pre-surgical presence of functional ICNs that encompassed the ictal temporal lobe in the case of cognitive/psychiatric outcomes. Our data made clear that the ICNs differed in their propensity to provide reserve for adaptive outcomes, with some providing structural (brain), and others functional (cognitive) reserve. Our customized methodology demonstrated that when substantial unique, patient-specific ICNs are present prior to surgery there is a reliable association with poor post-surgical seizure control. These ICNs are idiosyncratic in that they did not match the canonical, normative ICNs and, therefore, could not be defined functionally, with their location likely varying by patient. This important finding suggested the level of highly individualized ICN's in the epileptic brain may signal the emergence of epileptogenic activity after surgery.

Resting-state SEEG-based brain network analysis for the detection of epileptic area

BACKGROUND

Most epilepsy research is based on interictal or ictal functional connectivity. However, prolonged electrode implantation may affect patients' health and the accuracy of epileptic zone identification. Brief resting-state SEEG recordings reduce the observation of epileptic discharges by reducing electrode implantation and other seizure-inducing interventions.

NEW METHOD

The location coordinates of SEEG in the brain were identified using CT and MRI. Based on undirected brain network connectivity, five functional connectivity measures and data feature vector centrality were calculated. Network connectivity was calculated from multiple perspectives of linear correlation, information theory, phase, and frequency, and the relative influence of nodes on network connectivity was considered. We investigated the potential value of resting-state SEEG for epileptic zone identification by comparing the differences between epileptic and non-epileptic zones, as well as the differences between patients with different surgical outcomes.

RESULTS

By comparing the centrality of brain network connectivity between epileptic and non-epileptic zones, we found significant differences in the distribution of brain networks between the two zones. There was a significant difference in brain network between patients with good surgical outcomes and those with poor surgical outcomes (p < 0.01). By combining support vector machines with static node importance, we predicted an AUC of 0.94 ± 0.08 for the epilepsy zone.

CONCLUSIONS AND SIGNIFICANCE

The results illustrated that nodes in epileptic zones are distinct from those in non-epileptic zones. Analysis of resting-state SEEG data and the importance of nodes in the brain network may contribute to identifying the epileptic zone and predicting the outcome.

Control of epileptic activities in a cortex network of multiple coupled neural populations under electromagnetic induction

Epilepsy is believed to be associated with the abnormal synchronous neuronal activity in the brain, which results from large groups or circuits of neurons. In this paper, we choose to focus on the temporal lobe epilepsy, and establish a cortex network of multiple coupled neural populations to explore the epileptic activities under electromagnetic induction. We demonstrate that the epileptic activities can be controlled and modulated by electromagnetic induction and coupling among regions. In certain regions, these two types of control are observed to show exactly reverse effects. The results show that the strong electromagnetic induction is conducive to eliminating the epileptic seizures. The coupling among regions has a conduction effect that the previous normal background activity of the region gives way to the epileptic discharge, owing to coupling with spike wave discharge regions. Overall, these results highlight the role of electromagnetic induction and coupling among the regions in controlling and modulating epileptic activities, and might provide novel insights into the treatments of epilepsy.

Novel noninvasive identification of patient-specific epileptic networks in focal epilepsies: Linking single-photon emission computed tomography perfusion during seizures with resting-state magnetoencephalography dynamics

Single-photon emission computed tomography (SPECT) during seizures and magnetoencephalography (MEG) during the interictal state are noninvasive modalities employed in the localization of the epileptogenic zone in patients with drug-resistant focal epilepsy (DRFE). The present study aims to investigate whether there exists a preferentially high MEG functional connectivity (FC) among those regions of the brain that exhibit hyperperfusion or hypoperfusion during seizures. We studied MEG and SPECT data in 30 consecutive DRFE patients who had resective epilepsy surgery. We parcellated each ictal perfusion map into 200 regions of interest (ROIs) and generated ROI time series using source modeling of MEG data. FC between ROIs was quantified using coherence and phase-locking value. We defined a generalized linear model to relate the connectivity of each ROI, ictal perfusion z score, and distance between ROIs. We compared the coefficients relating perfusion z score to FC of each ROI and estimated the connectivity within and between resected and unresected ROIs. We found that perfusion z scores were strongly correlated with the FC of hyper-, and separately, hypoperfused ROIs across patients. High interictal connectivity was observed between hyperperfused brain regions inside and outside the resected area. High connectivity was also observed between regions of ictal hypoperfusion. Importantly, the ictally hypoperfused regions had a low interictal connectivity to regions that became hyperperfused during seizures. We conclude that brain regions exhibiting hyperperfusion during seizures highlight a preferentially connected interictal network, whereas regions of ictal hypoperfusion highlight a separate, discrete and interconnected, interictal network.

Epileptic seizure focus detection from interictal electroencephalogram: a survey

Electroencephalogram (EEG) is one of most effective clinical diagnosis modalities for the localization of epileptic focus. Most current AI solutions use this modality to analyze the EEG signals in an automated manner to identify the epileptic seizure focus. To develop AI system for identifying the epileptic focus, there are many recently-published AI solutions based on biomarkers or statistic features that utilize interictal EEGs. In this review, we survey these solutions and find that they can be divided into three main categories: (i) those that use of biomarkers in EEG signals, including high-frequency oscillation, phase-amplitude coupling, and interictal epileptiform discharges, (ii) others that utilize feature-extraction methods, and (iii) solutions based upon neural networks (an end-to-end approach). We provide a detailed description of seizure focus with clinical diagnosis methods, a summary of the public datasets that seek to reduce the research gap in epilepsy, recent novel performance evaluation criteria used to evaluate the AI systems, and guidelines on when and how to use them. This review also suggests a number of future research challenges that must be overcome in order to design more efficient computer-aided solutions to epilepsy focus detection.

Machine Learning for Precision Epilepsy Surgery

Brain surgery offers the best chance of seizure-freedom for patients with focal drug-resistant epilepsy, but only 50% achieve sustained seizure-freedom. With the explosion of data collected during routine presurgical evaluations and recent advances in computational science, we now have a tremendous potential to achieve precision epilepsy surgery: a data-driven tailoring of surgical planning. This review highlights the clinical need, the relevant computational science focusing on machine learning, and discusses some specific applications in epilepsy surgery.

Can Presurgical Interhemispheric EEG Connectivity Predict Outcome in Hemispheric Surgery? A Brain Machine Learning Approach

OBJECTIVES

Hemispherotomy (HT) is a surgical option for treatment of drug-resistant seizures due to hemispheric structural lesions. Factors affecting seizure outcome have not been fully clarified. In our study, we used a brain Machine Learning (ML) approach to evaluate the possible role of Inter-hemispheric EEG Connectivity (IC) in predicting post-surgical seizure outcome.

METHODS

We collected 21 pediatric patients with drug-resistant epilepsy; who underwent HT in our center from 2009 to 2020; with a follow-up of at least two years. We selected 5-s windows of wakefulness and sleep pre-surgical EEG and we trained Artificial Neuronal Network (ANN) to estimate epilepsy outcome. We extracted EEG features as input data and selected the ANN with best accuracy.

RESULTS

Among 21 patients, 15 (71%) were seizure and drug-free at last follow-up. ANN showed 73.3% of accuracy, with 85% of seizure free and 40% of non-seizure free patients appropriately classified.

CONCLUSIONS

The accuracy level that we reached supports the hypothesis that pre-surgical EEG features may have the potential to predict epilepsy outcome after HT.

SIGNIFICANCE

The role of pre-surgical EEG data in influencing seizure outcome after HT is still debated. We proposed a computational predictive model, with an ML approach, with a high accuracy level.

Interictal Functional Connectivity in Focal Refractory Epilepsies Investigated by Intracranial EEG

Introduction: Focal epilepsies are diseases of neuronal excitability affecting macroscopic networks of cortical and subcortical neural structures. These networks ("epileptogenic networks") can generate pathological electrophysiological activities during seizures, and also between seizures (interictal period). Many works attempt to describe these networks by using quantification methods, particularly based on the estimation of statistical relationships between signals produced by brain regions, namely functional connectivity (FC). Results: FC has been shown to be greatly altered during seizures and in the immediate peri-ictal period. An increasing number of studies have shown that FC is also altered during the interictal period depending on the degree of epileptogenicity of the structures. Furthermore, connectivity values could be correlated with other clinical variables including surgical outcome. Significance: This leads to a conceptual change and to consider epileptic areas as both hyperexcitable and abnormally connected. These data open the door to the use of interictal FC as a marker of epileptogenicity and as a complementary tool for predicting the effect of surgery. Aim: In this article, we review the available data concerning interictal FC estimated from intracranial electroencephalograhy (EEG) in focal epilepsies and discuss it in the light of data obtained from other modalities (EEG imaging) and modeling studies.

Interictal Connectivity Revealed by Granger Analysis of Stereoelectroencephalography: Association With Ictal Onset Zone, Resection, and Outcome

BACKGROUND

Stereoelectroencephalography (sEEG) facilitates electrical sampling and evaluation of complex deep-seated, dispersed, and multifocal locations. Granger causality (GC), previously used to study seizure networks using interictal data from subdural grids, may help identify the seizure-onset zone from interictal sEEG recordings.

OBJECTIVE

To examine whether statistical analysis of interictal sEEG helps identify surgical target sites and whether surgical resection of highly ranked nodes correspond to favorable outcomes.

METHODS

Ten minutes of extraoperative recordings from sequential patients who underwent sEEG evaluation were analyzed (n = 20). GC maps were compared with clinically defined surgical targets using rank order statistics. Outcomes of patients with focal resection/ablation with median follow-up of 3.6 years were classified as favorable (Engel 1, 2) or poor (Engel 3, 4) to assess their relationship with the removal of highly ranked nodes using the Wilcoxon rank-sum test.

RESULTS

In 12 of 20 cases, the rankings of contacts (based on the sum of outward connection weights) mapped to the seizure-onset zone showed higher causal node connectivity than predicted by chance ( P ≤ .02). A very low aggregate probability ( P < 10 -18 , n = 20) suggests that causal node connectivity predicts seizure networks. In 8 of 16 with outcome data, causal connectivity in the resection was significantly greater than in the remaining contacts ( P ≤ .05). We found a significant association between favorable outcome and the presence of highly ranked nodes in the resection ( P < .05).

CONCLUSION

Granger analysis can identify seizure foci from interictal sEEG and correlates highly ranked nodes with favorable outcome, potentially informing surgical decision-making without reliance on ictal recordings.

Normative brain mapping of interictal intracranial EEG to localize epileptogenic tissue

The identification of abnormal electrographic activity is important in a wide range of neurological disorders, including epilepsy for localizing epileptogenic tissue. However, this identification may be challenging during non-seizure (interictal) periods, especially if abnormalities are subtle compared to the repertoire of possible healthy brain dynamics. Here, we investigate if such interictal abnormalities become more salient by quantitatively accounting for the range of healthy brain dynamics in a location-specific manner. To this end, we constructed a normative map of brain dynamics, in terms of relative band power, from interictal intracranial recordings from 234 participants (21 598 electrode contacts). We then compared interictal recordings from 62 patients with epilepsy to the normative map to identify abnormal regions. We proposed that if the most abnormal regions were spared by surgery, then patients would be more likely to experience continued seizures postoperatively. We first confirmed that the spatial variations of band power in the normative map across brain regions were consistent with healthy variations reported in the literature. Second, when accounting for the normative variations, regions that were spared by surgery were more abnormal than those resected only in patients with persistent postoperative seizures (t = -3.6, P = 0.0003), confirming our hypothesis. Third, we found that this effect discriminated patient outcomes (area under curve 0.75 P = 0.0003). Normative mapping is a well-established practice in neuroscientific research. Our study suggests that this approach is feasible to detect interictal abnormalities in intracranial EEG, and of potential clinical value to identify pathological tissue in epilepsy. Finally, we make our normative intracranial map publicly available to facilitate future investigations in epilepsy and beyond.

Neuroimaging gradient alterations and epileptogenic prediction in focal cortical dysplasia Ⅲa

INTRODUCTION

Focal cortical dysplasia Type Ⅲa (FCD Ⅲa) is a highly prevalent temporal lobe epilepsy but the seizure outcomes are not satisfactory after epilepsy surgery. Hence, quantitative neuroimaging, epileptogenic alterations, as well as their values in guiding surgery are worth exploring.

METHODS

We examined 69 patients with pathologically verified FCD Ⅲa using multimodal neuroimaging and stereoelectroencephalography (SEEG). Among them, 18 received postoperative imaging which showed the extent of surgical resection and 9 underwent SEEG implantation. We also explored neuroimaging gradient alterations along with the distance to the temporal pole. Subsequently, the machine learning regression model was employed to predict whole-brain epileptogenicity. Lastly, the correlation between neuroimaging or epileptogenicity and surgical cavities was assessed.

RESULTS

FCD Ⅲa displayed neuroimaging gradient alterations on the temporal neocortex, morphology-signal intensity decoupling, low similarity of intra-morphological features and high similarity of intra-signal intensity features. The support vector regression model was successfully applied at the whole-brain level to calculate the continuous epileptogenic value at each vertex (mean-squared error = 13.8 ± 9.8).

CONCLUSION

Our study investigated the neuroimaging gradient alterations and epileptogenicity of FCD Ⅲa, along with their potential values in guiding suitable resection range and in predicting postoperative seizure outcomes. The conclusions from this study may facilitate an accurate presurgical examination of FCD Ⅲa. However, further investigation including a larger cohort is necessary to confirm the results.

Spatio-temporal spike dynamics predict surgical outcome in adult focal epilepsy

OBJECTIVE

We hypothesized that spatio-temporal dynamics of interictal spikes reflect the extent and stability of epileptic sources and determine surgical outcome.

METHODS

We studied 30 consecutive patients (14 good outcome). Spikes were detected in prolonged stereo-electroencephalography recordings. We quantified the spatio-temporal dynamics of spikes using the variance of the spike rate, line length and skewness of the spike distribution, and related these features to outcome. We built a logistic regression model, and compared its performance to traditional markers.

RESULTS

Good outcome patients had more dominant and stable sources than poor outcome patients as expressed by a higher variance of spike rates, a lower variance of line length, and a lower variance of positive skewness (ps < 0.05). The outcome was correctly predicted in 80% of patients. This was better or non-inferior to predictions based on a focal lesion (p = 0.016), focal seizure-onset zone, or complete resection (ps > 0.05). In the five patients where traditional markers failed, spike distribution predicted the outcome correctly. The best results were achieved by 18-h periods or longer.

CONCLUSIONS

Analysis of spike dynamics shows that surgery outcome depends on strong, single and stable sources.

SIGNIFICANCE

Our quantitative method has the potential to be a reliable predictor of surgical outcome.

Machine learning models for decision support in epilepsy management: A critical review

PURPOSE

There remain major challenges for the clinician in managing patients with epilepsy effectively. Choosing anti-seizure medications (ASMs) is subject to trial and error. About one-third of patients have drug-resistant epilepsy (DRE). Surgery may be considered for selected patients, but time from diagnosis to surgery averages 20 years. We reviewed the potential use of machine learning (ML) predictive models as clinical decision support tools to help address some of these issues.

METHODS

We conducted a comprehensive search of Medline and Embase of studies that investigated the application of ML in epilepsy management in terms of predicting ASM responsiveness, predicting DRE, identifying surgical candidates, and predicting epilepsy surgery outcomes. Original articles addressing these 4 areas published in English between 2000 and 2020 were included.

RESULTS

We identified 24 relevant articles: 6 on ASM responsiveness, 3 on DRE prediction, 2 on identifying surgical candidates, and 13 on predicting surgical outcomes. A variety of potential predictors were used including clinical, neuropsychological, imaging, electroencephalography, and health system claims data. A number of different ML algorithms and approaches were used for prediction, but only one study utilized deep learning methods. Some models show promising performance with areas under the curve above 0.9. However, most were single setting studies (18 of 24) with small sample sizes (median number of patients 55), with the exception of 3 studies that utilized large databases and 3 studies that performed external validation. There was a lack of standardization in reporting model performance. None of the models reviewed have been prospectively evaluated for their clinical benefits.

CONCLUSION

The utility of ML models for clinical decision support in epilepsy management remains to be determined. Future research should be directed toward conducting larger studies with external validation, standardization of reporting, and prospective evaluation of the ML model on patient outcomes.

Automated Identification of Surgical Candidates and Estimation of Postoperative Seizure Freedom in Children - A Focused Review

Surgery is an effective but underused treatment for drug-resistant epilepsy in children. Algorithms to identify surgical candidates and estimate the likelihood of postoperative clinical improvement may be valuable to improve access to epilepsy surgery. We provide a focused review of these approaches. For adults with epilepsy, tools to identify surgical candidates and predict seizure and cognitive outcomes (Ie, Cases for Epilepsy (toolsforepilepsy.com) and Epilepsy Surgery Grading Scale) have been validated and are in use. Analogous tools for children need development. A promising approach is to apply statistical learning tools to clinical datasets, such as electroencephalogram tracings, imaging studies, and the text of clinician notes. Demonstration projects suggest these techniques have the potential to be highly accurate, and await further validation and clinical application.

Highly consistent temporal lobe interictal spike networks revealed from foramen ovale electrodes

OBJECTIVE

A major challenge that limits understanding and treatment of epileptic events from mesial temporal structures comes from our inability to detect and map interictal networks reproducibly using scalp electrodes. Here, we developed a novel approach to map interictal spike networks and demonstrate their relationships to seizure onset and lesions in patients with foramen ovale electrode implantations.

METHODS

We applied the direct Directed Transfer Function to reveal interictal spike propagation from bilateral foramen ovale electrodes on 10 consecutive patients and co-registered spatially with both seizure onset zones and temporal lobe lesions.

RESULTS

Highly reproducible, yet unique interictal spike networks were seen for each patient (correlation: 0.93 ± 0.13). Interictal spikes spread in both anterior and posterior directions within each temporal lobe, often reverberating between sites. Spikes propagated to the opposite temporal lobe predominantly through posterior pathways. Patients with structural lesions (N = 4), including tumors and sclerosis, developed reproducible spike networks adjacent to their lesions that were highly lateralized compared to patients without lesions. Only 5% of mesial temporal lobe spikes were time-locked with scalp electrode spikes. Our preliminary observation on two lesional patients suggested that along with lesion location, Interictal spike networks also partially co-registered with seizure onset zones suggesting interrelationship between seizure onset and a subset of spike networks.

CONCLUSIONS

This is the first demonstration of patient-specific, reproducible interictal spike networks in mesial temporal structures that are closely linked to both temporal lobe lesions and seizure onset zones.

SIGNIFICANCE

Interictal spike connectivity is a novel approach to map epileptic networks that could help advance invasive and non-invasive epilepsy treatments.

Parieto-premotor functional connectivity changes during parietal lobe seizures are associated with motor semiology

OBJECTIVES

Parietal lobe seizures (PLS) are characterized by multiple clinical manifestations including motor signs. The mechanisms underlying the occurrence of motor signs are poorly understood. The main objective of this work was to estimate the functional coupling of brain regions associated with this clinical presentation.

METHODS

We retrospectively selected patients affected by drug-resistant epilepsy who underwent Stereoelectroencephalography (SEEG) for pre-surgical evaluation and in whom the seizure onset zone (SOZ) was located in the parietal cortex. The SOZ was defined visually and quantitatively by the epileptogenicity index (EI) method. Two groups of seizures were defined according to the presence ("motor seizures") or the absence ("non-motor seizures") of motor signs. Functional connectivity (FC) estimation was based on pairwise nonlinear regression analysis (h² coefficient). To study FC changes between parietal, frontal and temporal regions, for each patient, z-score values of 16 cortico-cortical interactions were obtained comparing h² coefficients of pre-ictal, seizure onset and seizure propagation periods.

RESULTS

We included 22 patients, 13 with "motor seizures" and 9 with "non-motor seizures". Resective surgery was performed in 14 patients, 8 patients had a positive surgical outcome (Engel's class I and II). During seizure onset period, a decrease of FC was observed and was significantly more important (in comparison with background period) in "motor" seizures. This was particularly observed between parietal operculum/post-central gyrus (OP/PoCg) and mesial temporal areas. During seizure propagation, a FC increase was significantly more important (in comparison with seizure onset) in "motor seizures", in particular between lateral pre-motor (pmL) area and precuneus, pmL and superior parietal lobule (SPL) and between inferior parietal lobule (IPL) and supplementary motor area (SMA).

CONCLUSIONS

Our study shows that motor semiology in PLS is accompanied by an increase of FC between parietal and premotor cortices, significantly different than what is observed in PLS without motor semiology.

SIGNIFICANCE

Our results indicate that preferential routes of coupling between parietal and premotor cortices are responsible for the prominent motor presentation during PLS.

Ictal gamma-band interactions localize ictogenic nodes of the epileptic network in focal cortical dysplasia

OBJECTIVE

Epilepsy surgery fails in > 30% of patients with focal cortical dysplasia (FCD). The seizure persistence after surgery can be attributed to the inability to precisely localize the tissue with an endogenous potential to generate seizures. In this study, we aimed to identify the critical components of the epileptic network that were actively involved in seizure genesis.

METHODS

The directed transfer function was applied to intracranial EEG recordings and the effective connectivity was determined with a high temporal and frequency resolution. Pre-ictal network properties were compared with ictal epochs to identify regions actively generating ictal activity and discriminate them from the areas of propagation.

RESULTS

Analysis of 276 seizures from 30 patients revealed the existence of a seizure-related network reconfiguration in the gamma-band (25-170 Hz; p < 0.005) - ictogenic nodes. Unlike seizure onset zone, resecting the majority of ictogenic nodes correlated with favorable outcomes (p < 0.012).

CONCLUSION

The prerequisite to successful epilepsy surgery is the accurate identification of brain areas from which seizures arise. We show that in FCD-related epilepsy, gamma-band network markers can reliably identify and distinguish ictogenic areas in macroelectrode recordings, improve intracranial EEG interpretation and better delineate the epileptogenic zone.

SIGNIFICANCE

Ictogenic nodes localize the critical parts of the epileptogenic tissue and increase the diagnostic yield of intracranial evaluation.

Large-Scale Desynchronization During Interictal Epileptic Discharges Recorded With Intracranial EEG

It is increasingly recognized that deep understanding of epileptic seizures requires both localizing and characterizing the functional network of the region where they are initiated, i. e., the epileptic focus. Previous investigations of the epileptogenic focus' functional connectivity have yielded contrasting results, reporting both pathological increases and decreases during resting periods and seizures. In this study, we shifted paradigm to investigate the time course of connectivity in relation to interictal epileptiform discharges. We recruited 35 epileptic patients undergoing intracranial EEG (iEEG) investigation as part of their presurgical evaluation. For each patient, 50 interictal epileptic discharges (IEDs) were marked and iEEG signals were epoched around those markers. Signals were narrow-band filtered and time resolved phase-locking values were computed to track the dynamics of functional connectivity during IEDs. Results show that IEDs are associated with a transient decrease in global functional connectivity, time-locked to the peak of the discharge and specific to the high range of the gamma frequency band. Disruption of the long-range connectivity between the epileptic focus and other brain areas might be an important process for the generation of epileptic activity. Transient desynchronization could be a potential biomarker of the epileptogenic focus since 1) the functional connectivity involving the focus decreases significantly more than the connectivity outside the focus and 2) patients with good surgical outcome appear to have a significantly more disconnected focus than patients with bad outcomes.

Connectome biomarkers of drug-resistant epilepsy

Drug-resistant epilepsy (DRE) considerably affects patient health, cognition, and well-being, and disproportionally contributes to the overall burden of epilepsy. The most common DRE syndromes are temporal lobe epilepsy related to mesiotemporal sclerosis and extratemporal epilepsy related to cortical malformations. Both syndromes have been traditionally considered as "focal," and most patients benefit from brain surgery for long-term seizure control. However, increasing evidence indicates that many DRE patients also present with widespread structural and functional network disruptions. These anomalies have been suggested to relate to cognitive impairment and prognosis, highlighting their importance for patient management. The advent of multimodal neuroimaging and formal methods to quantify complex systems has offered unprecedented ability to profile structural and functional brain networks in DRE patients. Here, we performed a systematic review on existing DRE network biomarker candidates and their contribution to three key application areas: (1) modeling of cognitive impairments, (2) localization of the surgical target, and (3) prediction of clinical and cognitive outcomes after surgery. Although network biomarkers hold promise for a range of clinical applications, translation of neuroimaging biomarkers to the patient's bedside has been challenged by a lack of clinical and prospective studies. We therefore close by highlighting conceptual and methodological strategies to improve the evaluation and accessibility of network biomarkers, and ultimately guide clinically actionable decisions.

Seizure Onset Zone Lateralization Using a Non-linear Analysis of Micro vs. Macro Electroencephalographic Recordings During Seizure-Free Stages of the Sleep-Wake Cycle From Epilepsy Patients

The application of non-linear signal analysis techniques to biomedical data is key to improve our knowledge about complex physiological and pathological processes. In particular, the use of non-linear techniques to study electroencephalographic (EEG) recordings can provide an advanced characterization of brain dynamics. In epilepsy these dynamics are altered at different spatial scales of neuronal organization. We therefore apply non-linear signal analysis to EEG recordings from epilepsy patients derived with intracranial hybrid electrodes, which are composed of classical macro contacts and micro wires. Thereby, these electrodes record EEG at two different spatial scales. Our aim is to test the degree to which the analysis of the EEG recorded at these different scales allows us to characterize the neuronal dynamics affected by epilepsy. For this purpose, we retrospectively analyzed long-term recordings performed during five nights in three patients during which no seizures took place. As a benchmark we used the accuracy with which this analysis allows determining the hemisphere that contains the seizure onset zone, which is the brain area where clinical seizures originate. We applied the surrogate-corrected non-linear predictability score (ψ), a non-linear signal analysis technique which was shown previously to be useful for the lateralization of the seizure onset zone from classical intracranial EEG macro contact recordings. Higher values of ψ were found predominantly for signals recorded from the hemisphere containing the seizure onset zone as compared to signals recorded from the opposite hemisphere. These differences were found not only for the EEG signals recorded with macro contacts, but also for those recorded with micro wires. In conclusion, the information obtained from the analysis of classical macro EEG contacts can be complemented by the one of micro wire EEG recordings. This combined approach may therefore help to further improve the degree to which quantitative EEG analysis can contribute to the diagnostics in epilepsy patients.

Differentiation of epileptic regions from voluntary high-gamma activation via interictal cross-frequency windowed power-power correlation

OBJECTIVE

Electrocorticography is an indispensable tool in identifying the epileptogenic zone in the presurgical evaluation of many epilepsy patients. Traditional electrocorticographic features (spikes, ictal onset changes, and recently high-frequency oscillations [HFOs]) rely on the presence of transient features that occur within or near epileptogenic cortex. Here the authors report on a novel corticography feature of epileptogenic cortex-covariation of high-gamma and beta frequency band power profiles. Band-limited power was measured from each recording site based on native physiological signal differences without relying on clinical ictal or interictal epileptogenic features. In this preliminary analysis, frequency windowed power correlation appears to be a specific marker of the epileptogenic zone. The authors' overall aim was to validate this observation with the location of the eventual resection and outcome.

METHODS

The authors conducted a retrospective analysis of 13 adult patients who had undergone electrocorticography for surgical planning at their center. They quantified the correlation of high-gamma (70-200 Hz) and beta (12-18 Hz) band frequency power per electrode site during a cognitive task. They used a sliding window method to correlate the power of smoothed, Hilbert-transformed high-gamma and beta bands. They then compared positive and negative correlations between power in the high-gamma and beta bands in the setting of a hand versus a tongue motor task as well as within the resting state. Significant positive correlations were compared to surgically resected areas and outcomes based on reviewed records.

RESULTS

Positive high-gamma and beta correlations appeared to predict the area of eventual resection and, preliminarily, surgical outcome independent of spike detection. In general, patients with the best outcomes had well-localized positive correlations (high-gamma and beta activities) to areas of eventual resection, while those with poorer outcomes displayed more diffuse patterns.

CONCLUSIONS

Data in this study suggest that positive high-gamma and beta correlations independent of any behavioral metric may have clinical applicability in surgical decision-making. Further studies are needed to evaluate the clinical potential of this methodology. Additional work is also needed to relate these results to other methods, such as HFO detection or connectivity with other cortical areas.

Machine learning applications in epilepsy

Machine learning leverages statistical and computer science principles to develop algorithms capable of improving performance through interpretation of data rather than through explicit instructions. Alongside widespread use in image recognition, language processing, and data mining, machine learning techniques have received increasing attention in medical applications, ranging from automated imaging analysis to disease forecasting. This review examines the parallel progress made in epilepsy, highlighting applications in automated seizure detection from electroencephalography (EEG), video, and kinetic data, automated imaging analysis and pre-surgical planning, prediction of medication response, and prediction of medical and surgical outcomes using a wide variety of data sources. A brief overview of commonly used machine learning approaches, as well as challenges in further application of machine learning techniques in epilepsy, is also presented. With increasing computational capabilities, availability of effective machine learning algorithms, and accumulation of larger datasets, clinicians and researchers will increasingly benefit from familiarity with these techniques and the significant progress already made in their application in 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.

Interictal stereotactic-EEG functional connectivity in refractory focal epilepsies

Drug-refractory focal epilepsies are network diseases associated with functional connectivity alterations both during ictal and interictal periods. A large majority of studies on the interictal/resting state have focused on functional MRI-based functional connectivity. Few studies have used electrophysiology, despite its high temporal capacities. In particular, stereotactic-EEG is highly suitable to study functional connectivity because it permits direct intracranial electrophysiological recordings with relative large-scale sampling. Most previous studies in stereotactic-EEG have been directed towards temporal lobe epilepsy, which does not represent the whole spectrum of drug-refractory epilepsies. The present study aims at filling this gap, investigating interictal functional connectivity alterations behind cortical epileptic organization and its association with post-surgical prognosis. To this purpose, we studied a large cohort of 59 patients with malformation of cortical development explored by stereotactic-EEG with a wide spatial sampling (76 distinct brain areas were recorded, median of 13.2 per patient). We computed functional connectivity using non-linear correlation. We focused on three zones defined by stereotactic-EEG ictal activity: the epileptogenic zone, the propagation zone and the non-involved zone. First, we compared within-zone and between-zones functional connectivity. Second, we analysed the directionality of functional connectivity between these zones. Third, we measured the associations between functional connectivity measures and clinical variables, especially post-surgical prognosis. Our study confirms that functional connectivity differs according to the zone under investigation. We found: (i) a gradual decrease of the within-zone functional connectivity with higher values for epileptogenic zone and propagation zone, and lower for non-involved zones; (ii) preferential coupling between structures of the epileptogenic zone; (iii) preferential coupling between epileptogenic zone and propagation zone; and (iv) poorer post-surgical outcome in patients with higher functional connectivity of non-involved zone (within- non-involved zone, between non-involved zone and propagation zone functional connectivity). Our work suggests that, even during the interictal state, functional connectivity is reinforced within epileptic cortices (epileptogenic zone and propagation zone) with a gradual organization. Moreover, larger functional connectivity alterations, suggesting more diffuse disease, are associated with poorer post-surgical prognosis. This is consistent with computational studies suggesting that connectivity is crucial in order to model the spatiotemporal dynamics of seizures.10.1093/brain/awy214_video1awy214media15833456182001.

Human non-olfactory cognition phase-locked with inhalation

Olfactory stimulus acquisition is perfectly synchronized with inhalation, which tunes neuronal ensembles for incoming information. Because olfaction is an ancient sensory system that provided a template for brain evolution, we hypothesized that this link persisted, and therefore nasal inhalations may also tune the brain for acquisition of non-olfactory information. To test this, we measured nasal airflow and electroencephalography during various non-olfactory cognitive tasks. We observed that participants spontaneously inhale at non-olfactory cognitive task onset and that such inhalations shift brain functional network architecture. Concentrating on visuospatial perception, we observed that nasal inhalation drove increased task-related brain activity in specific task-related brain regions and resulted in improved performance accuracy in the visuospatial task. Thus, mental processes with no link to olfaction are nevertheless phase-locked with nasal inhalation, consistent with the notion of an olfaction-based template in the evolution of human brain function.

Resting state connectivity in neocortical epilepsy: The epilepsy network as a patient-specific biomarker

OBJECTIVE

Localization related epilepsy (LRE) is increasingly accepted as a network disorder. To better understand the network specific characteristics of LRE, we defined individual epilepsy networks and compared them across patients.

METHODS

The epilepsy network was defined in the slow cortical potential frequency band in 10 patients using intracranial EEG data obtained during interictal periods. Cortical regions were included in the epilepsy network if their connectivity pattern was similar to the connectivity pattern of the seizure onset electrode contact. Patients were subdivided into frontal, temporal, and posterior quadrant cohorts according to the anatomic location of seizure onset. Jaccard similarity was calculated within each cohort to assess for similarity of the epilepsy network between patients within each cohort.

RESULTS

All patients exhibited an epilepsy network in the slow cortical potential frequency band. The topographic distribution of this correlated network activity was found to be unique at the single subject level.

CONCLUSIONS

The epilepsy network was unique at the single patient level, even between patients with similar seizure onset locations.

SIGNIFICANCE

We demonstrated that the epilepsy network is patient-specific. This is in keeping with our current understanding of brain networks and identifies the patient-specific epilepsy network as a possible biomarker in LRE.

Evoked directional network characteristics of epileptogenic tissue derived from single pulse electrical stimulation

We investigated effective networks constructed from single pulse electrical stimulation (SPES) in epilepsy patients who underwent intracranial electrocorticography. Using graph analysis, we compared network characteristics of tissue within and outside the epileptogenic area. In 21 patients with subdural electrode grids (1 cm interelectrode distance), we constructed a binary, directional network derived from SPES early responses (<100 ms). We calculated in‐degree, out‐degree, betweenness centrality, the percentage of bidirectional, receiving and activating connections, and the percentage of connections toward the (non‐)epileptogenic tissue for each node in the network. We analyzed whether these network measures were significantly different in seizure onset zone (SOZ)‐electrodes compared to non‐SOZ electrodes, in resected area (RA)‐electrodes compared to non‐RA electrodes, and in seizure free compared to not seizure‐free patients. Electrodes in the SOZ/RA showed significantly higher values for in‐degree and out‐degree, both at group level, and at patient level, and more so in seizure‐free patients. These differences were not observed for betweenness centrality. There were also more bidirectional and fewer receiving connections in the SOZ/RA in seizure‐free patients. It appears that the SOZ/RA is densely connected with itself, with only little input arriving from non‐SOZ/non‐RA electrodes. These results suggest that meso‐scale effective network measures are different in epileptogenic compared to normal brain tissue. Local connections within the SOZ/RA are increased and the SOZ/RA is relatively isolated from the surrounding cortex. This offers the prospect of enhanced prediction of epilepsy‐prone brain areas using SPES.

Morphometric MRI features and surgical outcome in patients with epilepsy related to hippocampal sclerosis and low intellectual quotient

OBJECTIVE

The objectives of this study were to verify in a series of patients with mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE-HS) if those with low intellectual quotient (IQ) levels have more extended areas of atrophy compared with those with higher IQ levels and to analyze whether IQ could be a variable implicated on a surgical outcome.

MATERIAL AND METHODS

Patients (n=106) with refractory MTLE-HS submitted to corticoamygdalohippocampectomy (CAH) (57 left mesial temporal lobe epilepsy (MTLE); 45 males) were enrolled. To determine if the IQ was a predictor of seizure outcome, totally seizure-free (SF) versus nonseizure-free (NSF) patients were evaluated. FreeSurfer was used for cortical thickness and volume estimation, comparing groups with lower (<80) and higher IQ (90-109) levels.

RESULTS

In the whole series, 42.45% of patients were SF (Engel Class 1a; n=45), and 57.54% were NSF (n=61). Total cortical volume was significantly reduced in the group with lower IQ (p=0.01). Significant reductions in the left hemisphere included the following: rostral middle frontal (p=0.001), insula (p=0.002), superior temporal gyrus (p=0.003), thalamus (p=0.004), and precentral gyrus (p=0.02); and those in the right hemisphere included the following: rostral middle frontal (p=0.003), pars orbitalis (p=0.01), and insula (p=0.02). Cortical thickness analysis also showed reductions in the right superior parietal gyrus in patients with lower IQ. No significant relationship between IQ and seizure outcome was found.

CONCLUSIONS

This is the first study of a series of patients with pure MTLE-HS, including those with low IQ and their morphometric magnetic resonance imaging (MRI) features using FreeSurfer. Although patients with lower intellectual scores presented more areas of brain atrophy, IQ was not a predictor of surgical outcome. Therefore, when evaluating seizure follow-up, low IQ in patients with MTLE-HS might not contraindicate resective surgery.

Resection of high frequency oscillations predicts seizure outcome in the individual patient

High frequency oscillations (HFOs) are recognized as biomarkers for epileptogenic brain tissue. A remaining challenge for epilepsy surgery is the prospective classification of tissue sampled by individual electrode contacts. We analysed long-term invasive recordings of 20 consecutive patients who subsequently underwent epilepsy surgery. HFOs were defined prospectively by a previously validated, automated algorithm in the ripple (80–250 Hz) and the fast ripple (FR, 250–500 Hz) frequency band. Contacts with the highest rate of ripples co-occurring with FR over several five-minute time intervals designated the HFO area. The HFO area was fully included in the resected area in all 13 patients who achieved seizure freedom (specificity 100%) and in 3 patients where seizures reoccurred (negative predictive value 81%). The HFO area was only partially resected in 4 patients suffering from recurrent seizures (positive predictive value 100%, sensitivity 57%). Thus, the resection of the prospectively defined HFO area proved to be highly specific and reproducible in 13/13 patients with seizure freedom, while it may have improved the outcome in 4/7 patients with recurrent seizures. We thus validated the clinical relevance of the HFO area in the individual patient with an automated procedure. This is a prerequisite before HFOs can guide surgical treatment in multicentre studies.

Task-based dynamic functional connectivity: Recent findings and open questions

The temporal evolution of functional connectivity (FC) within the confines of individual scans is nowadays often explored with functional neuroimaging. This is particularly true for resting-state; yet, FC-dynamics have also been investigated as subjects engage on numerous tasks. It is these research efforts that constitute the core of this survey. First, empirical observations on how FC differs between task and rest-independent of temporal scale-are reviewed, as they underscore how, despite overall preservation of network topography, the brain's FC does reconfigure in systematic ways to accommodate task demands. Next, reports on the relationships between instantaneous FC and perception/performance in subsequent trials are discussed. Similarly, research where different aspects of task-concurrent FC-dynamics are explored or utilized to predict ongoing mental states are also examined. The manuscript finishes with an incomplete list of challenges that hopefully fuels future work in this vibrant area of neuroscientific research. Overall, this review concludes that task-concurrent FC-dynamics, when properly characterized, are relevant to behavior, and that their translational value holds considerable promise.

Identification of the epileptogenic zone of temporal lobe epilepsy from stereo-electroencephalography signals: A phase transfer entropy and graph theory approach

The aim of this research is to apply an approach based on phase transfer entropy (PTE) and graph theory to study the interactions between the stereo-electroencephalography (SEEG) activities recorded in multilobar origin, in order to evaluate their ability to detect the epileptogenic zone (EZ) of temporal lobe epilepsies (TLE). Forty-three patients were included in this retrospective study. Five to sixteen (median = 12) multilead electrodes were implanted per patient, and, for each patient, a sub-set of between 10 and 32 (median = 22) bipolar derivations was selected for analysis. The leads were classified into the onset leads (OLs), the early propagation leads (EPLs), and the rest of the leads (RLs). The results showed that a significantly different dynamic trend of the out/in ratio (more obvious in the gamma band) distinguishes the OLs from RLs in the 23 patients who were seizure-free not only during the ictal event (significant elevation), but also during the inter-,pre-, late-ictal periods, and especially in the post-ictal (sharp decline) state. However, in the 20 patients who were not-seizure-free, the differences between the OLs and RLs during the post-ictal period were not found in any frequency band. The dynamic trend was used to predict surgical outcome, and the results showed that the sensitivity was 91% and the specificity was 70%. In brief, this study indicates that our approach may add new and valuable information, providing efficient quantitative measures useful for localizing the EZ.

Multimodal investigation of epileptic networks: The case of insular cortex epilepsy

The insula is a deep cortical structure sharing extensive synaptic connections with a variety of brain regions, including several frontal, temporal, and parietal structures. The identification of the insular connectivity network is obviously valuable for understanding a number of cognitive processes, but also for understanding epilepsy since insular seizures involve a number of remote brain regions. Ultimately, knowledge of the structure and causal relationships within the epileptic networks associated with insular cortex epilepsy can offer deeper insights into this relatively neglected type of epilepsy enabling the refining of the clinical approach in managing patients affected by it. In the present chapter, we first review the multimodal noninvasive tests performed during the presurgical evaluation of epileptic patients with drug refractory focal epilepsy, with particular emphasis on their value for the detection of insular cortex epilepsy. Second, we review the emerging multimodal investigation techniques in the field of epilepsy, that aim to (1) enhance the detection of insular cortex epilepsy and (2) unveil the architecture and causal relationships within epileptic networks. We summarize the results of these approaches with emphasis on the specific case of insular cortex epilepsy.

Interictal network synchrony and local heterogeneity predict epilepsy surgery outcome among pediatric patients

OBJECTIVE

Epilepsy is a disorder of aberrant cortical networks. Researchers have proposed that characterizing presurgical network connectivity may improve the surgical management of intractable seizures, but few studies have rigorously examined the relationship between network activity and surgical outcome. In this study, we assessed whether local and global measures of network activity differentiated patients with favorable (seizure-free) versus unfavorable (seizure-persistent) surgical outcomes.

METHODS

Seventeen pediatric intracranial electroencephalography (IEEG) patients were retrospectively examined. For each patient, 1,200 random interictal epochs of 1-s duration were analyzed. Functional connectivity networks were constructed using an amplitude-based correlation technique (Spearman correlation). Global network synchrony was computed as the average pairwise connectivity strength. Local signal heterogeneity was defined for each channel as the variability of EEG amplitude (root mean square) and absolute delta power (μV² /Hz) across epochs. A support vector machine learning algorithm used global and local measures to classify patients by surgical outcome. Classification was assessed using the Leave-One-Out (LOO) permutation test.

RESULTS

Global synchrony was increased in the seizure-persistent group compared to seizure-free patients (Student's t-test, p = 0.006). Seizure-onset zone (SOZ) electrodes exhibited increased signal heterogeneity compared to non-SOZ electrodes, primarily in seizure-persistent patients. Global synchrony and local heterogeneity measures were used to accurately classify 16 (94.1%) of 17 patients by surgical outcome (LOO test, iterations = 10,000, p < 0.001).

SIGNIFICANCE

Measures of global network synchrony and local signal heterogeneity represent promising biomarkers for assessing patient candidacy in pediatric epilepsy surgery.