Improving the prediction of epilepsy surgery outcomes using basic scalp EEG findings

EEG Ictal Power Dynamics, Function-Structure Associations, and Epilepsy Surgical Outcomes

BACKGROUND AND OBJECTIVES

Postoperative seizure control in drug-resistant temporal lobe epilepsy (TLE) remains variable, and the causes for this variability are not well understood. One contributing factor could be the extensive spread of synchronized ictal activity across networks. Our study used novel quantifiable assessments from intracranial EEG (iEEG) to test this hypothesis and investigated how the spread of seizures is determined by underlying structural network topological properties.

METHODS

We evaluated iEEG data from 157 seizures in 27 patients with TLE: 100 seizures from 17 patients with postoperative seizure control (Engel score I) vs 57 seizures from 10 patients with unfavorable surgical outcomes (Engel score II-IV). We introduced a quantifiable method to measure seizure power dynamics within anatomical regions, refining existing seizure imaging frameworks and minimizing reliance on subjective human decision-making. Time-frequency power representations were obtained in 6 frequency bands ranging from theta to gamma. Ictal power spectrums were normalized against a baseline clip taken at least 6 hours away from ictal events. Electrodes' time-frequency power spectrums were then mapped onto individual T1-weighted MRIs and grouped based on a standard brain atlas. We compared spatiotemporal dynamics for seizures between groups with favorable and unfavorable surgical outcomes. This comparison included examining the range of activated brain regions and the spreading rate of ictal activities. We then evaluated whether regional iEEG power values were a function of fractional anisotropy (FA) from diffusion tensor imaging across regions over time.

RESULTS

Seizures from patients with unfavorable outcomes exhibited significantly higher maximum activation sizes in various frequency bands. Notably, we provided quantifiable evidence that in seizures associated with unfavorable surgical outcomes, the spread of beta-band power across brain regions is significantly faster, detectable as early as the first second after seizure onset. There was a significant correlation between beta power during seizures and FA in the corresponding areas, particularly in the unfavorable outcome group. Our findings further suggest that integrating structural and functional features could improve the prediction of epilepsy surgical outcomes.

DISCUSSION

Our findings suggest that ictal iEEG power dynamics and the structural-functional relationship are mechanistic factors associated with surgical outcomes in TLE.

The value of ictal scalp EEG in focal epilepsies surgery: a retrospective analysis

OBJECTIVE

To describe the association between preoperative ictal scalp electroencephalogram (EEG) results and surgical outcomes in patients with focal epilepsies.

METHODS

The data of consecutive patients with focal epilepsies who received surgical treatments at our center from January 2012 to December 2021 were retrospectively analyzed.

RESULTS

Our data showed that 44.2% (322/729) of patients had ictal EEG recorded on video EEG monitoring during preoperative evaluation, of which 60.6% (195/322) had a concordant ictal EEG results. No significant difference of surgery outcomes between patients with and without ictal EEG was discovered. Among MRI-negative patients, those with concordant ictal EEG had a significantly better outcome than those without ictal EEG (75.7% vs. 43.8%, p = 0.024). Further logistic regression analysis showed that concordant ictal EEG was an independent predictor for a favorable outcome (OR = 4.430, 95%CI 1.175-16.694, p = 0.028). Among MRI-positive patients, those with extra-temporal lesions and discordant ictal EEG results had a worse outcome compared to those without an ictal EEG result (44.7% vs. 68.8%, p = 0.005). Further logistic regression analysis showed that discordant ictal EEG was an independent predictor of worse outcome (OR = 0.387, 95%CI 0.186-0.807, p = 0.011) in these patients. Furthermore, our data indicated that the number of seizures was not associated with the concordance rates of the ictal EEG, nor the surgical outcomes.

CONCLUSIONS

The value of ictal scalp EEG for epilepsy surgery varies widely among patients. A concordant ictal EEG predicts a good surgical outcome in MRI-negative patients, whereas a discordant ictal EEG predicts a poor postoperative outcome in lesional extratemporal lobe epilepsy.

The value of linear and non-linear quantitative EEG analysis in paediatric epilepsy surgery: a machine learning approach

Epilepsy surgery is effective for patients with medication-resistant seizures, however 20-40% of them are not seizure free after surgery. Aim of this study is to evaluate the role of linear and non-linear EEG features to predict post-surgical outcome. We included 123 paediatric patients who underwent epilepsy surgery at Bambino Gesù Children Hospital (January 2009-April 2020). All patients had long term video-EEG monitoring. We analysed 1-min scalp interictal EEG (wakefulness and sleep) and extracted 13 linear and non-linear EEG features (power spectral density (PSD), Hjorth, approximate entropy, permutation entropy, Lyapunov and Hurst value). We used a logistic regression (LR) as feature selection process. To quantify the correlation between EEG features and surgical outcome we used an artificial neural network (ANN) model with 18 architectures. LR revealed a significant correlation between PSD of alpha band (sleep), Mobility index (sleep) and the Hurst value (sleep and awake) with outcome. The fifty-four ANN models gave a range of accuracy (46-65%) in predicting outcome. Within the fifty-four ANN models, we found a higher accuracy (64.8% ± 7.6%) in seizure outcome prediction, using features selected by LR. The combination of PSD of alpha band, mobility and the Hurst value positively correlate with good surgical outcome.

Fully automatic segmentation of brain lacunas resulting from resective surgery using a 3D deep learning model

The rapid and constant development of deep learning (DL) strategies is pushing forward the quality of object segmentation in images from diverse fields of interest. In particular, these algorithms can be very helpful in delineating brain abnormalities (lesions, tumors, lacunas, etc), enabling the extraction of information such as volume and location, that can inform doctors or feed predictive models. Here, we describe ResectVol DL, a fully automatic tool developed to segment resective lacunas in brain images of patients with epilepsy. ResectVol DL relies on the nnU-Net framework that leverages the 3D U-Net deep learning architecture. T1-weighted MRI datasets from 120 patients (57 women; 31.5 ± 15.9 years old at surgery) were used to train (n=78) and test (n=48) our tool. Manual segmentations were carried out by five different raters and were considered as ground truth for performance assessment. We compared ResectVol DL with two other fully automatic methods: ResectVol 1.1.2 and DeepResection, using the Dice similarity coefficient (DSC), Pearson's correlation coefficient, and relative difference to manual segmentation. ResectVol DL presented the highest median DSC (0.92 vs. 0.78 and 0.90), the highest correlation coefficient (0.99 vs. 0.63 and 0.94), and the lowest median relative difference (9 vs. 44 and 12 %). Overall, we demonstrate that ResectVol DL accurately segments brain lacunas, which has the potential to assist in the development of predictive models for postoperative cognitive and seizure outcomes.

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.

Hemimegalencephaly: A Systematic Comparison of Functional and Anatomic Hemispherectomy for Drug-Resistant Epilepsy

BACKGROUND AND OBJECTIVES

Hemimegalencephaly (HME) is a rare diffuse malformation of cortical development characterized by unihemispheric hypertrophy, drug-resistant epilepsy (DRE), hemiparesis, and developmental delay. Definitive treatment for HME-related DRE is hemispheric surgery through either anatomic (AH) or functional hemispherectomy (FH). This individual patient data meta-analysis assessed seizure outcomes of AH and FH for HME with pharmacoresistant epilepsy, predictors of Engel I, and efficacy of different FH approaches.

METHODS

PubMed, Web of Science, and Cumulative Index to Nursing and Allied Health Literature were searched from inception to Jan 13th, 2023, for primary literature reporting seizure outcomes in >3 patients with HME receiving AH or FH. Demographics, neurophysiology findings, and Engel outcome at the last follow-up were extracted. Postsurgical seizure outcomes were compared through 2-tailed t -test and Fisher exact test. Univariate and multivariate Cox regression analyses were performed to identify independent predictors of Engel I outcome.

RESULTS

Data from 145 patients were extracted from 26 studies, of which 89 underwent FH (22 vertical, 33 lateral), 47 underwent AH, and 9 received an unspecified hemispherectomy with a median last follow-up of 44.0 months (FH cohort) and 45.0 months (AH cohort). Cohorts were similar in preoperative characteristics and at the last follow-up; 77% (n = 66) of the FH cohort and 81% (n = 38) and of the AH cohort were Engel I. On multivariate analysis, only the presence of bilateral ictal electroencephalography abnormalities (hazard ratio = 11.5; P = .002) was significantly associated with faster time-to-seizure recurrence. A number-needed-to-treat analysis to prevent 1 additional case of posthemispherectomy hydrocephalus reveals that FH, compared with AH, was 3. There was no statistical significance for any differences in time-to-seizure recurrence between lateral and vertical FH approaches (hazard ratio = 2.59; P = .101).

CONCLUSION

We show that hemispheric surgery is a highly effective treatment for HME-related DRE. Unilateral ictal electroencephalography changes and using the FH approach as initial surgical management may result in better outcomes due to significantly lower posthemispherectomy hydrocephalus probability. However, larger HME registries are needed to further delineate the predictors of seizure outcomes.

Prediction tools and risk stratification in epilepsy surgery

OBJECTIVE

This study was undertaken to conduct external validation of previously published epilepsy surgery prediction tools using a large independent multicenter dataset and to assess whether these tools can stratify patients for being operated on and for becoming free of disabling seizures (International League Against Epilepsy stage 1 and 2).

METHODS

We analyzed a dataset of 1562 patients, not used for tool development. We applied two scales: Epilepsy Surgery Grading Scale (ESGS) and Seizure Freedom Score (SFS); and two versions of Epilepsy Surgery Nomogram (ESN): the original version and the modified version, which included electroencephalographic data. For the ESNs, we used calibration curves and concordance indexes. We stratified the patients into three tiers for assessing the chances of attaining freedom from disabling seizures after surgery: high (ESGS = 1, SFS = 3-4, ESNs > 70%), moderate (ESGS = 2, SFS = 2, ESNs = 40%-70%), and low (ESGS = 2, SFS = 0-1, ESNs < 40%). We compared the three tiers as stratified by these tools, concerning the proportion of patients who were operated on, and for the proportion of patients who became free of disabling seizures.

RESULTS

The concordance indexes for the various versions of the nomograms were between .56 and .69. Both scales (ESGS, SFS) and nomograms accurately stratified the patients for becoming free of disabling seizures, with significant differences among the three tiers (p < .05). In addition, ESGS and the modified ESN accurately stratified the patients for having been offered surgery, with significant difference among the three tiers (p < .05).

SIGNIFICANCE

ESGS and the modified ESN (at thresholds of 40% and 70%) stratify patients undergoing presurgical evaluation into three tiers, with high, moderate, and low chance for favorable outcome, with significant differences between the groups concerning having surgery and becoming free of disabling seizures. Stratifying patients for epilepsy surgery has the potential to help select the optimal candidates in underprivileged areas and better allocate resources in developed countries.

Early Identification of Candidates for Epilepsy Surgery: A Multicenter, Machine Learning, Prospective Validation Study

BACKGROUND AND OBJECTIVES

Epilepsy surgery is often delayed. We previously developed machine learning (ML) models to identify candidates for resective epilepsy surgery earlier in the disease course. In this study, we report the prospective validation.

METHODS

In this multicenter, prospective, longitudinal cohort study, random forest models were validated at a pediatric epilepsy center consisting of 2 hospitals and 14 outpatient neurology clinic sites and an adult epilepsy center with 2 hospitals and 27 outpatient neurology clinic sites. The models used neurology visit notes, EEG and MRI reports, visit patterns, hospitalizations, and medication, laboratory, and procedure orders to identify candidates for surgery. The models were trained on historical data up to May 10, 2019. Patients with an ICD-10 diagnosis of epilepsy who visited from May 11, 2019, to May 10, 2020, were screened by the algorithm and assigned surgical candidacy scores. The primary outcome was area under the curve (AUC), which was calculated by comparing scores from patients who underwent epilepsy surgery before November 10, 2020, against scores from nonsurgical patients. Nonsurgical patients' charts were reviewed to determine whether patients with high scores were more likely to be missed surgical candidates. Delay to surgery was defined as the time between the first visit that a surgical candidate was identified by the algorithm and the date of the surgery.

RESULTS

A total of 5,285 pediatric and 5,782 adult patients were included to train the ML algorithms. During the study period, 41 children and 23 adults underwent resective epilepsy surgery. In the pediatric cohort, AUC was 0.91 (95% CI 0.87-0.94), positive predictive value (PPV) was 0.08 (0.05-0.10), and negative predictive value (NPV) was 1.00 (0.99-1.00). In the adult cohort, AUC was 0.91 (0.86-0.97), PPV was 0.07 (0.04-0.11), and NPV was 1.00 (0.99-1.00). The models first identified patients at a median of 2.1 years (interquartile range [IQR]: 1.2-4.9 years, maximum: 11.1 years) before their surgery and 1.3 years (IQR: 0.3-4.0 years, maximum: 10.1 years) before their presurgical evaluations.

DISCUSSION

ML algorithms can identify surgical candidates earlier in the disease course. Even at specialized epilepsy centers, there is room to shorten the time to surgery.

CLASSIFICATION OF EVIDENCE

This study provides Class II evidence that a machine learning algorithm can accurately distinguish patients with epilepsy who require resective surgery from those who do not.

Association Between Magnetoencephalography-Localized Epileptogenic Zone, Surgical Resection Volume, and Postsurgical Seizure Outcome

PURPOSE

Surgical resection of magnetoencephalography (MEG) dipole clusters, reconstructed from interictal epileptiform discharges, is associated with favorable seizure outcomes. However, the relation of MEG cluster resection to the surgical resection volume is not known nor is it clear whether this association is direct and causal, or it may be mediated by the resection volume or other predictive factors. This study aims to clarify these open questions and assess the diagnostic accuracy of MEG in our center.

METHODS

We performed a retrospective cohort study of 68 patients with drug-resistant epilepsy who underwent MEG followed by resective epilepsy surgery and had at least 12 months of postsurgical follow-up.

RESULTS

Good seizure outcomes were associated with monofocal localization (χ2 = 6.94, P = 0.001; diagnostic odds ratio = 10.2) and complete resection of MEG clusters (χ2 = 22.1, P < 0.001; diagnostic odds ratio = 42.5). Resection volumes in patients with and without removal of MEG clusters were not significantly different (t = 0.18, P = 0.86; removed: M = 20,118 mm3, SD = 10,257; not removed: M = 19,566 mm3, SD = 10,703). Logistic regression showed that removal of MEG clusters predicts seizure-free outcome independent of the resection volume and other prognostic factors (P < 0.001).

CONCLUSIONS

Complete resection of MEG clusters leads to favorable seizure outcomes without affecting the volume of surgical resection and independent of other prognostic factors. MEG can localize the epileptogenic zone with high accuracy. MEG interictal epileptiform discharges mapping should be used whenever feasible to improve postsurgical seizure outcomes.

Predicting seizure recurrence after an initial seizure-like episode from routine clinical notes using large language models: a retrospective cohort study

BACKGROUND

The evaluation and management of first-time seizure-like events in children can be difficult because these episodes are not always directly observed and might be epileptic seizures or other conditions (seizure mimics). We aimed to evaluate whether machine learning models using real-world data could predict seizure recurrence after an initial seizure-like event.

METHODS

This retrospective cohort study compared models trained and evaluated on two separate datasets between Jan 1, 2010, and Jan 1, 2020: electronic medical records (EMRs) at Boston Children's Hospital and de-identified, patient-level, administrative claims data from the IBM MarketScan research database. The study population comprised patients with an initial diagnosis of either epilepsy or convulsions before the age of 21 years, based on International Classification of Diseases, Clinical Modification (ICD-CM) codes. We compared machine learning-based predictive modelling using structured data (logistic regression and XGBoost) with emerging techniques in natural language processing by use of large language models.

FINDINGS

The primary cohort comprised 14 021 patients at Boston Children's Hospital matching inclusion criteria with an initial seizure-like event and the comparison cohort comprised 15 062 patients within the IBM MarketScan research database. Seizure recurrence based on a composite expert-derived definition occurred in 57% of patients at Boston Children's Hospital and 63% of patients within IBM MarketScan. Large language models with additional domain-specific and location-specific pre-training on patients excluded from the study (F1-score 0·826 [95% CI 0·817-0·835], AUC 0·897 [95% CI 0·875-0·913]) performed best. All large language models, including the base model without additional pre-training (F1-score 0·739 [95% CI 0·738-0·741], AUROC 0·846 [95% CI 0·826-0·861]) outperformed models trained with structured data. With structured data only, XGBoost outperformed logistic regression and XGBoost models trained with the Boston Children's Hospital EMR (logistic regression: F1-score 0·650 [95% CI 0·643-0·657], AUC 0·694 [95% CI 0·685-0·705], XGBoost: F1-score 0·679 [0·676-0·683], AUC 0·725 [0·717-0·734]) performed similarly to models trained on the IBM MarketScan database (logistic regression: F1-score 0·596 [0·590-0·601], AUC 0·670 [0·664-0·675], XGBoost: F1-score 0·678 [0·668-0·687], AUC 0·710 [0·703-0·714]).

INTERPRETATION

Physician's clinical notes about an initial seizure-like event include substantial signals for prediction of seizure recurrence, and additional domain-specific and location-specific pre-training can significantly improve the performance of clinical large language models, even for specialised cohorts.

FUNDING

UCB, National Institute of Neurological Disorders and Stroke (US National Institutes of Health).

Outcomes of resective surgery in pediatric patients with drug-resistant epilepsy: A single-center study from the Eastern Mediterranean Region

OBJECTIVE

Epilepsy surgery is widely accepted as an effective therapeutic option for carefully selected patients with drug-resistant epilepsy (DRE). There is limited data on the outcome of epilepsy surgery, especially in pediatric patients from the Eastern Mediterranean region. Hence, we performed a retrospective study examining the outcomes of resective surgery in 53 pediatric patients with focal DRE.

METHODS

Patients with focal DRE who had undergone epilepsy surgery were included in the present study. All patients underwent a comprehensive presurgical evaluation. Postoperative seizure outcomes were classified using the Engel Epilepsy Surgery Outcome Scale.

RESULTS

After surgery, 33 patients (62.2%) were Class I according to the Engel classification of surgical outcomes; eight patients (15.0%) were Class II, 11 (20.7%) were Class III, and one (1.8%) was Class IV. The relationships of presurgical, surgical, and postsurgical clinical variables to seizure outcomes were compared. Older age at seizure onset, older age at the time of surgery, the presence of focal to bilateral tonic-clonic seizures, seizure duration over 2 minutes, unsuccessful treatment with three or fewer antiseizure medications, lesions confined to one lobe (as demonstrated via magnetic resonance imaging [MRI]), surgical site in the temporal lobe, and histopathology including developmental tumors were significantly linked to an Engel Class I outcome. A univariate analysis of excellent surgical outcomes showed that lateralized semiology, localized interictal and ictal electroencephalogram (EEG) discharges, lateralized single-photon emission computed tomography and positron emission tomography findings, and temporal lobe resections were significantly related to excellent seizure outcomes.

SIGNIFICANCE

The results of our study are encouraging and similar to those found in other centers around the world. Epilepsy surgery remains an underutilized treatment for children with DRE and should be offered early.

Normative brain mapping using scalp EEG and potential clinical application

A normative electrographic activity map could be a powerful resource to understand normal brain function and identify abnormal activity. Here, we present a normative brain map using scalp EEG in terms of relative band power. In this exploratory study we investigate its temporal stability, its similarity to other imaging modalities, and explore a potential clinical application. We constructed scalp EEG normative maps of brain dynamics from 17 healthy controls using source-localised resting-state scalp recordings. We then correlated these maps with those acquired from MEG and intracranial EEG to investigate their similarity. Lastly, we use the normative maps to lateralise abnormal regions in epilepsy. Spatial patterns of band powers were broadly consistent with previous literature and stable across recordings. Scalp EEG normative maps were most similar to other modalities in the alpha band, and relatively similar across most bands. Towards a clinical application in epilepsy, we found abnormal temporal regions ipsilateral to the epileptogenic hemisphere. Scalp EEG relative band power normative maps are spatially stable across time, in keeping with MEG and intracranial EEG results. Normative mapping is feasible and may be potentially clinically useful in epilepsy. Future studies with larger sample sizes and high-density EEG are now required for validation.

Emerging glyco-risk prediction model to forecast response to immune checkpoint inhibitors in colorectal cancer

BACKGROUND

Aberrant glycosylation is one of the most common post-translational modifications leading to heterogeneity in colorectal cancer (CRC). This study aims to construct a risk prediction model based on glycosyltransferase to forecast the response to immune checkpoint inhibitors in CRC patients.

METHODS

Based on the TCGA dataset and glycosyltransferase genes, the NMF algorithm and WGCNA were used to identify molecular subtypes and co-expressed genes, respectively. Lasso and multivariate COX regression were used to identify prognostic glycosyltransferase genes and construct a glyco-risk prediction model in CRC patients. Univariate and multivariate Cox regression, Kaplan-Meier, and ROC curves were applied to further verify the prognostic performance of the model in CRC patients in the training and validation sets. We compared the responsiveness of immunotherapy and chemotherapy between the two groups. In vitro experiments and clinical specimens verified the specific function of the key glycosyltransferase genes in CRC.

RESULTS

The CRC cohort was divided into two subtypes with prominent differences in survival based on the well-robust seven-gene glyco-risk prediction model (composed of ALG1L2, HAS1, PYGL, COLGALT2, B3GNT4, POFUT2, and GALNT7). The nomograms based on the risk model could predict the prognosis of CRC patients independently of other clinicopathologic characteristics. Our prediction model showed a better overall prediction performance than other models. Compared with the low-risk group, the high-risk CRC patients showed a lower immune infiltration state, but a higher TMB and a lower response to anti-PD-1, anti-PD-L1, and anti-CTLA-4 therapy. Clinical specimen validation showed an obvious difference in the expression of seven glycosyltransferase genes between the low- and high-risk groups. Significant reduction in POFUT2 expression in high-risk groups was associated with reduced N-glycans production.

CONCLUSION

Our study constructed a robust glyco-risk prediction model that could provide direction for immunotherapy and chemotherapy in CRC patients, which could help clinicians make personalized treatment decisions.

Normative brain mapping using scalp EEG and potential clinical application

A normative electrographic activity map could be a powerful resource to understand normal brain function and identify abnormal activity. Here, we present a normative brain map using scalp EEG in terms of relative band power. In this exploratory study we investigate its temporal stability, its similarity to other imaging modalities, and explore a potential clinical application. We constructed scalp EEG normative maps of brain dynamics from 17 healthy controls using source-localised resting-state scalp recordings. We then correlated these maps with those acquired from MEG and intracranial EEG to investigate their similarity. Lastly, we use the normative maps to lateralise abnormal regions in epilepsy. Spatial patterns of band powers were broadly consistent with previous literature and stable across recordings. Scalp EEG normative maps were most similar to other modalities in the alpha band, and relatively similar across most bands. Towards a clinical application in epilepsy, we found abnormal temporal regions ipsilateral to the epileptogenic hemisphere. Scalp EEG relative band power normative maps are spatially stable across time, in keeping with MEG and intracranial EEG results. Normative mapping is feasible and may be potentially clinically useful in epilepsy. Future studies with larger sample sizes and high-density EEG are now required for validation.

Prediction of epilepsy surgery outcome using foramen ovale EEG - A machine learning approach

INTRODUCTION

Patients with drug-resistant focal epilepsy may benefit from ablative or resective surgery. In presurgical work-up, intracranial EEG markers have been shown to be useful in identification of the seizure onset zone and prediction of post-surgical seizure freedom. However, in most cases, implantation of depth or subdural electrodes is performed, exposing patients to increased risks of complications.

METHODS

We analysed EEG data recorded from a minimally invasive approach utilizing foramen ovale (FO) and epidural peg electrodes using a supervised machine learning approach to predict post-surgical seizure freedom. Power-spectral EEG features were incorporated in a logistic regression model predicting one-year post-surgical seizure freedom. The prediction model was validated using repeated 5-fold cross-validation and compared to outcome prediction based on clinical and scalp EEG variables.

RESULTS

Forty-seven patients (26 patients with post-surgical 1-year seizure freedom) were included in the study, with 31 having FO and 27 patients having peg onset seizures. The area under the receiver-operating curve for post-surgical seizure freedom (Engel 1A) prediction in patients with FO onset seizures was 0.74 ± 0.23 using electrophysiology features, compared to 0.66 ± 0.22 for predictions based on clinical and scalp EEG variables (p < 0.001). The most important features for prediction were spectral power in the gamma and high gamma ranges. EEG data from peg electrodes was not informative in predicting post-surgical outcomes.

CONCLUSION

In this hypothesis-generating study, a data-driven approach based on EEG features derived from FO electrodes recordings outperformed the predictive ability based solely on clinical and scalp EEG variables. Pending validation in future studies, this method may provide valuable post-surgical prognostic information while minimizing risks of more invasive diagnostic approaches.

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.

Seizure outcome determinants in children after surgery for single unilateral lesions on magnetic resonance imaging: Role of preoperative ictal and interictal electroencephalography

OBJECTIVE

To determine whether an ictal electroencephalographic (EEG) recording as part of presurgical evaluation of children with a demarcated single unilateral magnetic resonance imaging (MRI) lesion is indispensable for surgical decision-making, we investigated the relationship of interictal/ictal EEG and seizure semiology with seizure-free outcome.

METHODS

Data were obtained retrospectively from consecutive patients (≤18 years old) undergoing epilepsy surgery with a single unilateral MRI lesion at our institution over a 6-year period. Video-telemetry EEG (VT-EEG) was classified as concordant or nonconcordant/noninformative in relation to the MRI lesion location. The odds of seizure-free outcome associated with nonconcordant versus concordant for semiology, interictal EEG, and ictal EEG were compared separately. Multivariate logistic regression was conducted to correct for confounding variables.

RESULTS

After a median follow-up of 26 months (interquartile range = 17-37.5), 73 (69%) of 117 children enrolled were seizure-free. Histopathological diagnoses included low-grade epilepsy-associated tumors, n = 46 (39%); focal cortical dysplasia (FCD), n = 33 (28%); mesial temporal sclerosis (MTS), n = 23 (20%); polymicrogyria, n = 3 (3%); and nondiagnostic findings/gliosis, n = 12 (10%). The odds of seizure freedom were lower with a nonconcordant interictal EEG (odds ratio [OR] = .227, 95% confidence interval [CI] = .079-.646, p = .006) and nonconcordant ictal EEG (OR = .359, 95% CI = .15-.878, p = .035). In the multivariate logistic regression model, factors predicting lower odds for seizure-free outcome were developmental delay/intellectual disability and higher number of antiseizure medications tried, with a nonsignificant trend for "nonconcordant interictal EEG." In the combined subgroup of patients with FCD and tumors (n = 79), there was no significant relationship of VT-EEG factors and seizure outcomes, whereas in children with MTS and acquired lesions (n = 25), a nonconcordant EEG was associated with poorer seizure outcomes (p = .003).

SIGNIFICANCE

An ictal EEG may not be mandatory for presurgical evaluation, particularly when a well-defined single unilateral MRI lesion has been identified and the interictal EEG is concordant.

Distribution of postictal slowing has an additional yield to interictal epileptiform discharge in predicting surgical outcomes in temporal lobe epilepsy

OBJECTIVE

To investigate whether the slowing of bilateral postictal scalp electroencephalography (EEG) after focal impaired awareness seizures is associated with poor seizure outcomes after temporal lobe epilepsy (TLE) surgery.

METHODS

This retrospective cohort study was conducted in the Department of Epileptology, Tohoku University Hospital from 2010 to 2020. The study included 42 patients with TLE who underwent a detailed presurgical evaluation and sequential resective surgery for the unilateral probable epileptogenic temporal lobe with 1 year or more of follow-up. We reviewed the interictal epileptiform distribution and those of the ictal and postictal epochs of the first focal impaired awareness seizure recorded in presurgical scalp EEG. We classified patients either with postoperative seizure-free status (Engel I) as group A or those with seizure persistence (Engel II-IV) as group B.

RESULTS

Of 42 patients, 29 (69%) were classified into group A. Compared with group B, group A had a lower number of bilateral postictal polymorphic delta activity (PPDA) (10.3%: 61.5%) and bilateral interictal epileptiform discharges (IEDs) (13.8%: 69.2%) (P = 0.003, P = 0.001, respectively). A combined analysis of bilateral PPDA and IEDs per individual patient showed significantly more frequent seizure persistence after surgery (P < 0.0001) than a single analysis of bilateral IEDs or PPDA alone (P = 0.001). The regression analysis revealed that bilaterally distributed PPDA or IEDs had 13.50 or 13.72 times higher odds of persisting seizures within 1 year of surgery (95% confidence interval: 1.90-95.88; 2.12-88.87, respectively) (P = 0.009, 0.006).

SIGNIFICANCE

The results of this study revealed that the bilateral distribution of PPDA was associated with poor postoperative seizure outcomes in patients with TLE, as well as bilateral IEDs. Additionally, the concomitant bilateral distribution of interictal and postictal changes is a strong indicator of poor surgical outcomes.

VEP for VIP Care: Patient-Specific Modeling in Epilepsy Surgery

[Box: see text]

A prediction model integrating synchronization biomarkers and clinical features to identify responders to vagus nerve stimulation among pediatric patients with drug-resistant epilepsy

Aims

Vagus nerve stimulation (VNS) is a neuromodulation therapy for children with drug‐resistant epilepsy (DRE). The efficacy of VNS is heterogeneous. A prediction model is needed to predict the efficacy before implantation.

Methods

We collected data from children with DRE who underwent VNS implantation and received regular programming for at least 1 year. Preoperative clinical information and scalp video electroencephalography (EEG) were available in 88 children. Synchronization features, including phase lag index (PLI), weighted phase lag index (wPLI), and phase‐locking value (PLV), were compared between responders and non‐responders. We further adapted a support vector machine (SVM) classifier selected from 25 clinical and 18 synchronization features to build a prediction model for efficacy in a discovery cohort (n = 70) and was tested in an independent validation cohort (n = 18).

Results

In the discovery cohort, the average interictal awake PLI in the high beta band was significantly higher in responders than non‐responders (p < 0.05). The SVM classifier generated from integrating both clinical and synchronization features had the best prediction efficacy, demonstrating an accuracy of 75.7%, precision of 80.8% and area under the receiver operating characteristic (AUC) of 0.766 on 10‐fold cross‐validation. In the validation cohort, the prediction model demonstrated an accuracy of 61.1%.

Conclusion

This study established the first prediction model integrating clinical and baseline synchronization features for preoperative VNS responder screening among children with DRE. With further optimization of the model, we hope to provide an effective and convenient method for identifying responders before VNS implantation.

Quantitative Analysis of Visually Reviewed Normal Scalp EEG Predicts Seizure Freedom Following Anterior Temporal Lobectomy

Objective

Anterior temporal lobectomy (ATL) is a widely performed and successful intervention for drug‐resistant temporal lobe epilepsy (TLE). However, up to one third of patients experience seizure recurrence within 1 year after ATL. Despite the extensive literature on presurgical electroencephalography (EEG) and magnetic resonance imaging (MRI) abnormalities to prognosticate seizure freedom following ATL, the value of quantitative analysis of visually reviewed normal interictal EEG in such prognostication remains unclear. In this retrospective multicenter study, we investigate whether machine learning analysis of normal interictal scalp EEG studies can inform the prediction of postoperative seizure freedom outcomes in patients who have undergone ATL.

Methods

We analyzed normal presurgical scalp EEG recordings from 41 Mayo Clinic (MC) and 23 Cleveland Clinic (CC) patients. We used an unbiased automated algorithm to extract eyes closed awake epochs from scalp EEG studies that were free of any epileptiform activity and then extracted spectral EEG features representing (a) spectral power and (b) interhemispheric spectral coherence in frequencies between 1 and 25 Hz across several brain regions. We analyzed the differences between the seizure‐free and non–seizure‐free patients and employed a Naïve Bayes classifier using multiple spectral features to predict surgery outcomes. We trained the classifier using a leave‐one‐patient‐out cross‐validation scheme within the MC data set and then tested using the out‐of‐sample CC data set. Finally, we compared the predictive performance of normal scalp EEG‐derived features against MRI abnormalities.

Results

We found that several spectral power and coherence features showed significant differences correlated with surgical outcomes and that they were most pronounced in the 10–25 Hz range. The Naïve Bayes classification based on those features predicted 1‐year seizure freedom following ATL with area under the curve (AUC) values of 0.78 and 0.76 for the MC and CC data sets, respectively. Subsequent analyses revealed that (a) interhemispheric spectral coherence features in the 10–25 Hz range provided better predictability than other combinations and (b) normal scalp EEG‐derived features provided superior and potentially distinct predictive value when compared with MRI abnormalities (>10% higher F1 score).

Significance

These results support that quantitative analysis of even a normal presurgical scalp EEG may help prognosticate seizure freedom following ATL in patients with drug‐resistant TLE. Although the mechanism for this result is not known, the scalp EEG spectral and coherence properties predicting seizure freedom may represent activity arising from the neocortex or the networks responsible for temporal lobe seizure generation within vs outside the margins of an ATL.

Neuropathology and epilepsy surgery

PURPOSE OF REVIEW

Neurosurgical treatment of patients suffering from drug-resistant focal epilepsy is recognized as a successful, yet underutilized medical treatment option. By searching PubMed for articles published between January 2020 and September 2021 with the broad search terms 'neuropathology' AND 'epilepsy surgery', this review highlights the active field of etiology-based epilepsy research in human tissue.

RECENT FINDINGS

All papers addressing the most common epileptogenic human brain disease entities, i.e. focal cortical dysplasia (FCD), brain tumors or hippocampal sclerosis, and written in English language were eligible for our review. We can conclude from this review that etiology-based studies are of foremost interest for (1) the development of prediction models for postsurgical seizure outcome; (2) decipher genetic and molecular alterations to better define disease entities and underlying molecular pathomechanisms, and (3) the translation of human tissue-derived biomarker into clinically useful diagnostics or novel therapeutic targets in the near future.

SUMMARY

Highlighting FCD brain somatic gain-of-function variants in mammalian target of Rapamycin are a leading pathway to better classify FCD. An integrated genotype-phenotype analysis enables to classify the broad spectrum of low-grade and epilepsy-associated brain tumors. Further DNA-methylation-based disease classification will increase the mechanistic understanding and diagnostic precision of difficult to classify pathologies in the future.

Development and Validation of the 5-SENSE Score to Predict Focality of the Seizure-Onset Zone as Assessed by Stereoelectroencephalography

Importance

Stereoelectroencephalography (SEEG) has become the criterion standard in case of inconclusive noninvasive presurgical epilepsy workup. However, up to 40% of patients are subsequently not offered surgery because the seizure-onset zone is less focal than expected or cannot be identified.

Objective

To predict focality of the seizure-onset zone in SEEG, the 5-point 5-SENSE score was developed and validated.

Design, Setting, and Participants

This was a monocentric cohort study for score development followed by multicenter validation with patient selection intervals between February 2002 to October 2018 and May 2002 to December 2019. The minimum follow-up period was 1 year. Patients with drug-resistant epilepsy undergoing SEEG at the Montreal Neurological Institute were analyzed to identify a focal seizure-onset zone. Selection criteria were 2 or more seizures in electroencephalography and availability of complete neuropsychological and neuroimaging data sets. For validation, patients from 9 epilepsy centers meeting these criteria were included. Analysis took place between May and July 2021.

Main Outcomes and Measures

Based on SEEG, patients were grouped as focal and nonfocal seizure-onset zone. Demographic, clinical, electroencephalography, neuroimaging, and neuropsychology data were analyzed, and a multiple logistic regression model for developing a score to predict SEEG focality was created and validated in an independent sample.

Results

A total of 128 patients (57 women [44.5%]; median [range] age, 31 [13-58] years) were analyzed for score development and 207 patients (97 women [46.9%]; median [range] age, 32 [16-70] years) were analyzed for validation. The score comprised the following 5 predictive variables: focal lesion on structural magnetic resonance imaging, absence of bilateral independent spikes in scalp electroencephalography, localizing neuropsychological deficit, strongly localizing semiology, and regional ictal scalp electroencephalography onset. The 5-SENSE score had an optimal mean (SD) probability cutoff for identifying a focal seizure-onset zone of 37.6 (3.5). Area under the curve, specificity, and sensitivity were 0.83, 76.3% (95% CI, 66.7-85.8), and 83.3% (95% CI, 72.30-94.1), respectively. Validation showed 76.0% (95% CI, 67.5-84.0) specificity and 52.3% (95% CI, 43.0-61.5) sensitivity.

Conclusions and Relevance

High specificity in score development and validation confirms that the 5-SENSE score predicts patients where SEEG is unlikely to identify a focal seizure-onset zone. It is a simple and useful tool for assisting clinicians to reduce unnecessary invasive diagnostic burden on patients and overutilization of limited health care resources.