[Clinicopathological characteristics of gangliogliomas with anaplastic morphology].

Objective: To investigate the clinical, radiological, and pathological features of anaplastic gangliogliomas (AGGs) and to determine whether these tumors represent a distinct entity. Methods: Consecutive 667 cases of ganglioglioma (GG) diagnosed at the Xuanwu Hospital, Capital Medical University, Beijing, China between January 2015 and July 2023 were screened. Among these cases, 9 pathologically confirmed AGG cases were identified. Their clinical, radiological, treatment, and outcome data were analyzed retrospectively. Most of the tumor samples were subject to next-generation sequencing, while a subset of them were subject to DNA methylation profiling. Results: Among the 9 patients, there were five males and four females, with a median age of 8 years. Epileptic seizures (5/9) were the most frequently presented symptom. Radiological examinations showed three types of radiological manifestations: four cases showed abnormal MRI signals with no significant mass effects and mild enhancement; two cases demonstrated a mixed solid-cystic density lesion with peritumoral edema, which showed significant heterogeneous enhancement and obvious mass effects, and one case displayed cystic cavity formation with nodules on MRI, which showed evident enhancements. All cases exhibited mutations that were predicted to activate the MAP kinase signaling pathway, including seven with BRAF p.V600E mutation and two with NF1 mutation. Five AGGs with mutations involving the MAP kinase signaling pathway also had concurrent mutations, including three with CDKN2A homozygous deletion, one with a TERT promoter mutation, one with a H3F3A mutation, and one with a PTEN mutation. Conclusions: AGG exhibits a distinct spectrum of pathology, genetic mutations and clinical behaviors, differing from GG. Given these characteristics suggest that AGG may be a distinct tumor type, further expansion of the case series is needed. Therefore, a comprehensive integration of clinical, histological, and molecular analyses is required to correctly diagnose AGG. It will also help guide treatments and prognostication.

Detecting local perturbations of networks in a latent hyperbolic embedding space.

This paper introduces two novel scores for detecting local perturbations in networks. For this, we consider a non-Euclidean representation of networks, namely, their embedding onto the Poincaré disk model of hyperbolic geometry. We numerically evaluate the performances of these scores for the detection and localization of perturbations on homogeneous and heterogeneous network models. To illustrate our approach, we study latent geometric representations of real brain networks to identify and quantify the impact of epilepsy surgery on brain regions. Results suggest that our approach can provide a powerful tool for representing and analyzing changes in brain networks following surgical intervention, marking the first application of geometric network embedding in epilepsy research.

Focal cortical dysplasia II caused by brain somatic mutation of IRS-1 is associated with ERK signaling pathway activation.

Somatic mutations have been identified in 10% to 63% of focal cortical dysplasia type II samples, primarily linked to the mTOR pathway. When the causative genetic mutations are not identified, this opens the possibility of discovering new pathogenic genes or pathways that could be contributing to the condition. In our previous study, we identified a novel candidate pathogenic somatic variant of IRS-1 c.1791dupG in the brain tissue of a child with focal cortical dysplasia type II. This study further explored the variant's role in causing type II focal cortical dysplasia through in vitro overexpression in 293T and SH-SY5Y cells and in vivo evaluation via in utero electroporation in fetal brains, assessing effects on neuronal migration, morphology, and network integrity. It was found that the mutant IRS-1 variant led to hyperactivity of p-ERK, increased cell volume, and was predominantly associated with the MAPK signaling pathway. In vivo, the IRS-1 c.1791dupG variant induced abnormal neuron migration, cytomegaly, and network hyperexcitability. Notably, the ERK inhibitor GDC-0994, rather than the mTOR inhibitor rapamycin, effectively rescued the neuronal defects. This study directly highlighted the ERK signaling pathway's role in the pathogenesis of focal cortical dysplasia II and provided a new therapeutic target for cases of focal cortical dysplasia II that are not treatable by rapamycin analogs.

New data-driven method to predict the therapeutic indication of redeemed prescriptions in secondary data sources: a case study on antiseizure medications users aged ≥65 identified in Danish registries.

OBJECTIVES: We aimed to develop a new data-driven method to predict the therapeutic indication of redeemed prescriptions in secondary data sources using antiepileptic drugs among individuals aged ≥65 identified in Danish registries. DESIGN: This was an incident new-user register-based cohort study using Danish registers. SETTING: The study setting was Denmark and the study period was 2005-2017. PARTICIPANTS: Participants included antiepileptic drug users in Denmark aged ≥65 with a confirmed diagnosis of epilepsy. PRIMARY AND SECONDARY OUTCOME MEASURES: Sensitivity served as the performance measure of the algorithm. RESULTS: The study population comprised 8609 incident new users of antiepileptic drugs. The sensitivity of the algorithm in correctly predicting the therapeutic indication of antiepileptic drugs in the study population was 65.3% (95% CI 64.4 to 66.2). CONCLUSIONS: The algorithm demonstrated promising properties in terms of overall sensitivity for predicting the therapeutic indication of redeemed antiepileptic drugs by older individuals with epilepsy, correctly identifying the therapeutic indication for 6 out of 10 individuals using antiepileptic drugs for epilepsy.

Quality of life and associated factors among people with epilepsy in Ethiopia: a systematic review and meta-analysis.

BACKGROUND: Epilepsy is a global health and economic burden with major problems that have an impact on physical, psychological, and social activities. Quality of life (QoL) is often disturbed and can be influenced by many factors, like anti-seizure medication side effects, the sociocultural environment, and various disease-related factors. The aim of this systematic review and meta-analysis is to provide an overview of the most recent information available regarding the pooled prevalence of poor quality of life and associated factors among adult people with epilepsy in Ethiopia. METHODS: The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) is an appropriate set of guidelines for reporting systematic reviews and meta-analyses. This systematic review and meta-analysis protocol was registered on the International Prospective Register of Systematic Reviews (PROSPERO) with CRD42024527914. To find publications for the systematic review and meta-analysis, we used both manual and electronic searches. The publications were searched by PubMed, MEDLINE, EMBASE, Cochrane Library, Scopus, and other grey publications were searched by Google Scholar. The Joanna Briggs Institute (JBI) for cross-sectional study quality assessment was employed to evaluate the methodological quality of the studies included in this review. The data was extracted in Microsoft Excel, and then it was exported into STATA 11.0 for analysis. A funnel plot and an objective examination of Egger's regression test were used to check for publication bias. RESULTS: We have included 7 studies conducted in Ethiopia with 2123 study participants, of whom 1163 (54.78%) were male individuals, and 1196 (56.34%) of the participants were living without marriage (either single, divorced, or widowed). The pooled prevalence of poor quality of life among people with epilepsy in Ethiopia is 45.07 (95% CI: 39.73-50.42%). Further, in subgroup analysis regarding the assessment tool of poor quality of life of people with epilepsy, QOLIE-31 accounted for 50.05% (95%CI: 46.65-53.45) and WHO QOL BREF accounted for 39.72% (95%CI: 27.67-51.78). Among the associated factors, being unable to read and write, anxiey and depression were significantly linked to the quality of life of people with epilepsy. CONCLUSION: This review found that there was a high pooled prevalence of poor quality of life related to people with epilepsy in Ethiopia. This study may provide further information to concerned bodies that do early screening and manage the quality of life of individuals with epilepsy. Also, screening and intervention for anxiety and depression problems should be considered in regular epilepsy care management.

Epilepsy, Antiepileptic Drugs, and Adverse Pregnancy Outcomes, 1: Examination and Interpretation of Recent Research.

The age-standardized global prevalence of epilepsy is about 0.3% in women. Seizures are associated with morbidity and mortality risks; so, women with epilepsy (WWE) are usually advised antiepileptic drug (AED) treatment even during pregnancy. Women may also knowingly or unknowingly be exposed during pregnancy to AEDs advised for other on- or off-label indications. In this context, a meta-analysis of 35 adverse gestational outcomes examined in 76 observational studies found that WWE were at increased risk of most of the adverse outcomes, regardless of gestational exposure to AEDs. AEDs, especially in polytherapy, further increased at least a few of the gestational risks, including risks of congenital conditions, neonatal intensive care unit admission, small for gestational age, low birth weight, and neonatal/infant death (it is unclear whether the lack of statistical significance for the remaining risks was because AED exposure was truly limited to these risks or whether the nonsignificant analyses were underpowered). Reassuringly, the increases in risk were mostly in the small to modest range. This meta-analysis pooled unadjusted risks (which would probably be larger than adjusted risks), so readers are informed about expected findings in the population but not about cause-effect relationships that may be cautiously hypothesized from adjusted analyses. A take-home message is that, because of the wide range of outcomes for which risk is increased, WWE should be closely monitored and followed all through pregnancy, regardless of treatment with AEDs. This article also provides readers with suggestions on how to critically interpret literature with regard to 8 matters: confounding by indication and confounding by severity of indication, as specific to the indication for AED prescription; unadjusted and adjusted analyses; the base rate of an outcome in the population; the examination of multiple outcomes; the uniform direction of findings; the sample numbers; the timing of AED exposure; and self-fulfilling prophecies.

Effect of SCN1Aand SCN2A gene polymorphisms on the efficacy of valproic acid treatment in Chinese children with epilepsy.

OBJECTIVE: Epilepsy patients exhibit considerable differences in their response to sodium valproate (VPA) therapy, a phenomenon that might be attributed to individual genetic variances. The role of genetic variations, specifically in sodium channels encoded by SCN1A and SCN2A genes, in influencing the effectiveness of VPA in treating epilepsy is still debated. This research focuses on examining the impact of these genetic polymorphisms on the efficacy of VPA therapy among pediatric epilepsy patients in China. METHODS: Five single nucleotide polymorphisms (SNPs), including SCN1A (rs10188577, rs2298771, rs3812718) and SCN2A (rs2304016, rs17183814), were genotyped in 233 epilepsy patients undergoing VPA therapy. The associations between genotypes and the antiepileptic effects of VPA were assessed, with 128 patients categorized as VPA responders and 105 as VPA non-responders. RESULTS: In the context of VPA monotherapy, SCN1A rs2298771 and SCN2A rs17183814 were found to be significantly associated with VPA response (P< 0.05). CONCLUSION: Our study suggests the findings of this investigation indicate that the polymorphisms SCN1A rs2298771 and SCN2A rs17183814 could potentially act as predictive biomarkers for the responsiveness to VPA among Chinese epilepsy patients.

Automatic detection of epilepsy from EEGs using a temporal convolutional network with a self-attention layer.

BACKGROUND: Over 60% of epilepsy patients globally are children, whose early diagnosis and treatment are critical for their development and can substantially reduce the disease's burden on both families and society. Numerous algorithms for automated epilepsy detection from EEGs have been proposed. Yet, the occurrence of epileptic seizures during an EEG exam cannot always be guaranteed in clinical practice. Models that exclusively use seizure EEGs for detection risk artificially enhanced performance metrics. Therefore, there is a pressing need for a universally applicable model that can perform automatic epilepsy detection in a variety of complex real-world scenarios. METHOD: To address this problem, we have devised a novel technique employing a temporal convolutional neural network with self-attention (TCN-SA). Our model comprises two primary components: a TCN for extracting time-variant features from EEG signals, followed by a self-attention (SA) layer that assigns importance to these features. By focusing on key features, our model achieves heightened classification accuracy for epilepsy detection. RESULTS: The efficacy of our model was validated on a pediatric epilepsy dataset we collected and on the Bonn dataset, attaining accuracies of 95.50% on our dataset, and 97.37% (A v. E), and 93.50% (B vs E), respectively. When compared with other deep learning architectures (temporal convolutional neural network, self-attention network, and standardized convolutional neural network) using the same datasets, our TCN-SA model demonstrated superior performance in the automated detection of epilepsy. CONCLUSION: The proven effectiveness of the TCN-SA approach substantiates its potential as a valuable tool for the automated detection of epilepsy, offering significant benefits in diverse and complex real-world clinical settings.

Multi-target Phenylpropanoids Against Epilepsy.

Epilepsy is a neurological disease with no defined cause, characterized by recurrent epileptic seizures. These occur due to the dysregulation of excitatory and inhibitory neurotransmitters in the central nervous system (CNS). Psychopharmaceuticals have undesirable side effects; many patients require more than one pharmacotherapy to control crises. With this in mind, this work emphasizes the discovery of new substances from natural products that can combat epileptic seizures. Using in silico techniques, this review aims to evaluate the antiepileptic and multi-target activity of phenylpropanoid derivatives. Initially, ligand-based virtual screening models (LBVS) were performed with 468 phenylpropanoid compounds to predict biological activities. The LBVS were developed for the targets alpha- amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), voltage-gated calcium channel Ttype (CaV), gamma-aminobutyric acid A (GABAA), gamma-aminobutyric acid transporter type 1 (GAT-1), voltage-gated potassium channel of the Q family (KCNQ), voltage-gated sodium channel (NaV), and N-methyl D-aspartate (NMDA). The compounds that had good results in the LBVS were analyzed for the absorption, distribution, metabolism, excretion, and toxicity (ADMET) parameters, and later, the best molecules were evaluated in the molecular docking consensus. The TR430 compound showed the best results in pharmacokinetic parameters; its oral absorption was 99.03%, it did not violate any Lipinski rule, it showed good bioavailability, and no cytotoxicity was observed either from the molecule or from the metabolites in the evaluated parameters. TR430 was able to bind with GABAA (activation) and AMPA (inhibition) targets and demonstrated good binding energy and significant interactions with both targets. The studied compound showed to be a promising molecule with a possible multi-target activity in both fundamental pharmacological targets for the treatment of epilepsy.

Spectrum of NMDA Receptor Variants in Neurodevelopmental Disorders and Epilepsy.

N-methyl-D-aspartate receptors (NMDAR) are ligand-gated ion channels mediating excitatory neurotransmission and are important for normal brain development, cognitive abilities, and motor functions. Pathogenic variants in the Glutamate receptor Ionotropic N-methyl-D-aspartate (GRIN) genes (GRIN1, GRIN2A-D) encoding NMDAR subunits have been associated with a wide spectrum of neurodevelopmental disorders and epilepsies ranging from treatable focal epilepsies to devastating early-onset developmental and epileptic encephalopathies. Genetic variants in NMDA receptor genes can cause a range of complex alterations to receptor properties resulting in various degrees of loss-of-function, gain-of-function, or mixtures thereof. Understanding how genetic variants affect the function of the receptors, therefore, represents an important first step in the ongoing development towards targeted therapies. Currently, targeted treatment options for GRIN-related diseases are limited. However, treatment with memantine has been reported to significantly reduce seizure frequency in a few individuals with developmental and epileptic encephalopathies harboring de novo gain-of-function GRIN2A missense variants, and supplementary treatment with L-serine has been associated with improved motor and cognitive performance as well as reduced seizure frequency in patients with GRIN2B loss-of-function missense variants as well as GRIN2A and GRIN2B null variants.

Efficacy and safety of herbal medicine combined with acupuncture in pediatric epilepsy treatment: A meta-analysis of randomized controlled trials.

OBJECTIVE: To evaluate the efficacy and safety of herbal medicine and acupuncture combination for pediatric epilepsy treatment. METHODS: Databases were searched from their interception until October 2023 to identify randomized controlled trials focusing on the therapeutic efficacy of herbal medicine-acupuncture combination (intervention group) for pediatric epilepsy. The primary outcome was the risk of treatment failure, whereas the secondary outcomes included the risk of post-treatment electroencephalogram (EEG) abnormalities and adverse events. Subgroup analyses were conducted based on the type of herbal compound formulas. Meta-regression analysis was conducted to examine the influence of patient demographics and clinical history on the therapeutic efficacy of herbal medicine-acupuncture combination for pediatric epilepsy. To assess the cumulative evidence, trial sequential analysis (TSA) was performed. RESULTS: The analysis included 10 trials involving a total of 882 pediatric patients. Meta-analysis revealed that the intervention group had a lower risk of treatment failure than the control group (risk ratio [RR] = 0.3, 95% confidence interval [CI]: 0.19-0.47, P<0.00001, I2 = 0%, 10 trials). Subgroup analyses showed that therapeutic efficacy was consistent among the different herbal compound formulas. Meta-regression analysis revealed that the efficacy of the treatments did not significantly vary with patient age, male sex, and duration of seizure history. TSA suggested that herbal medicine-acupuncture combination exerted a robust and conclusive effect on seizure treatment. Although the combined used of herbal medicine and acupuncture was not associated with a lower risk of post-treatment EEG abnormalities (RR = 0.82, 95%CI:0.6-1.11, P = 0.2, 3 trials), the risk of adverse events was reduced (RR = 0.27, 95%CI:0.18-0.41, P<0.00001, 4 trials). CONCLUSION: The meta-analysis suggested that combined use of herbal medicine and acupuncture is a promising and safe clinical approach for pediatric epilepsy treatment. Further large-scale studies are necessary to conclusively determine the efficacy and safety of herbal medicine and acupuncture in pediatric epilepsy treatment.

Diagnostic biomarker discovery from brain EEG data using LSTM, reservoir-SNN, and NeuCube methods in a pilot study comparing epilepsy and migraine.

The study introduces a new online spike encoding algorithm for spiking neural networks (SNN) and suggests new methods for learning and identifying diagnostic biomarkers using three prominent deep learning neural network models: deep BiLSTM, reservoir SNN, and NeuCube. EEG data from datasets related to epilepsy, migraine, and healthy subjects are employed. Results reveal that BiLSTM hidden neurons capture biological significance, while reservoir SNN activities and NeuCube spiking dynamics identify EEG channels as diagnostic biomarkers. BiLSTM and reservoir SNN achieve 90 and 85% classification accuracy, while NeuCube achieves 97%, all methods pinpointing potential biomarkers like T6, F7, C4, and F8. The research bears implications for refining online EEG classification, analysis, and early brain state diagnosis, enhancing AI models with interpretability and discovery. The proposed techniques hold promise for streamlined brain-computer interfaces and clinical applications, representing a significant advancement in pattern discovery across the three most popular neural network methods for addressing a crucial problem. Further research is planned to study how early can these diagnostic biomarkers predict an onset of brain states.

Therapeutic Drug Monitoring of Levetiracetam in Different Trimesters of Pregnant Women with Epilepsy in a Tertiary Care Center: A Prospective Study.

BACKGROUND: Levetiracetam is the most commonly used antiepileptic drug in pregnant women due to its low teratogenic risk profile, favorable pharmacokinetic characteristics, and safety profile. Serum levels of levetiracetam vary in epilepsy during pregnancy. Therefore, the aim of the study was to evaluate the serum levels of levetiracetam during different trimesters of pregnancy by using therapeutic drug monitoring (TDM). MATERIALS AND METHODS: This was a single-center, prospective study. Pregnant women with epilepsy on levetiracetam were enrolled after getting written informed consent from them. Serum trough levels of levetiracetam were estimated at all trimesters by high-performance liquid chromatography (HPLC). RESULTS: The study included 16 participants with mean ± standard deviation (SD) age of 27.75 ± 4 years. There were nine (56.2%) participants with generalized seizure disorder and seven (43.8%) participants of focal seizure disorder. Among 16 patients, 10 (62.5%) participants were on levetiracetam alone and six (37.5%) participants were on levetiracetam combined with other antiepileptic drugs. In a total of 48 trough samples, 45 sample concentrations were below the therapeutic range of 12-46 mg/l and three sample concentrations were within the therapeutic range. There was a statistically significant difference in the concentration-dose ratio (CDR) of levetiracetam between the third and first trimesters (P-value 0.018). CONCLUSION: There was a statistically significant difference in serum levetiracetam concentration between the third and first trimesters. A well-conducted, intensive pharmacokinetic sampling study in PWWE with a control group is needed in future to evaluate the whole pharmacokinetic profile of levetiracetam and to correlate the clinical outcome.

A multiscale symbolic approach to decoding delta and ripple oscillation bands as biomarkers for epileptiform discharges.

We use a multiscale symbolic approach to study the complex dynamics of temporal lobe refractory epilepsy employing high-resolution intracranial electroencephalogram (iEEG). We consider the basal and preictal phases and meticulously analyze the dynamics across frequency bands, focusing on high-frequency oscillations up to 240 Hz. Our results reveal significant periodicities and critical time scales within neural dynamics across frequency bands. By bandpass filtering neural signals into delta, theta, alpha, beta, gamma, and ripple high-frequency bands (HFO), each associated with specific neural processes, we examine the distinct nonlinear dynamics. Our method introduces a reliable approach to pinpoint intrinsic time lag scales τ within frequency bands of the basal and preictal signals, which are crucial for the study of refractory epilepsy. Using metrics such as permutation entropy (H), Fisher information (F), and complexity (C), we explore nonlinear patterns within iEEG signals. We reveal the intrinsic τmax that maximize complexity within each frequency band, unveiling the nonlinear subtle patterns of the temporal structures within the basal and preictal signal. Examining the H×F and C×F values allows us to identify differences in the delta band and a band between 200 and 220 Hz (HFO 6) when comparing basal and preictal signals. Differences in Fisher information in the delta and HFO 6 bands before seizures highlight their role in capturing important system dynamics. This offers new perspectives on the intricate relationship between delta oscillations and HFO waves in patients with focal epilepsy, highlighting the importance of these patterns and their potential as biomarkers.

[Risk of epilepsy after a first epileptic seizure with unknown etiology in elderly patients]. / Riesgo de epilepsia tras una primera crisis epiléptica de etiología desconocida en pacientes de edad avanzada.

AIM: Patients whose epilepsy begins with seizures with unknown etiology in old age have been studied to a limited extent. The aim is to clinically characterise these patients, and predict their risk of developing epilepsy in the long term. MATERIALS AND METHODS: This is a retrospective observational study of patients over 55 years old experiencing a first epileptic seizure with unknown etiology. The data were collected from their clinical history, including electroencephalogram (EEG) and brain magnetic resonance imaging (MRI) results. RESULTS: Eighty-seven patients (58.6% male; 71.5 ± 8.1 years) were included. The mean follow-up was 7.3 ± 4.9 years. The most common vascular risk factor was arterial hypertension (77%; n = 67). Focal seizures with altered consciousness were the most frequent type of seizure (44.8%; n = 39), followed by focal seizures evolving to bilateral tonic-clonic seizures (39.1%; n = 34). Brain MRI showed cortical atrophy (50%; n = 42) and signs of small-vessel vascular disease (SVVD) (67.8%; n = 57). Interictal epileptiform EEG abnormalities were observed in 43.7% (n = 38) of the patients, mostly with temporal localisations (94.7%; n = 36). 44.8% (n = 39) had mild cognitive impairment at baseline. Recurrence of seizures, which was observed in 49 patients (56.1%), occurred after a median of 12 months (interquartile range: 4.4-25.9). Finally, 71 patients (81.6%) developed epilepsy. CONCLUSION: The risk of epilepsy in the long term following a single seizure of unknown etiology in elderly patients is greater than 80%. Arterial hypertension and mild cognitive impairment at baseline are the most common clinical features. Cortical atrophy and the presence of SVVD are frequent in MRI, and routine EEGs do not usually show epileptiform alterations.

TITLE: Riesgo de epilepsia tras una primera crisis epiléptica de etiología desconocida en pacientes de edad avanzada.Objetivo. Los pacientes que comienzan con crisis de origen desconocido en la edad avanzada no están bien estudiados. El objetivo es caracterizar clínicamente a estos pacientes y predecir el riesgo de desarrollar epilepsia a largo plazo. Materiales y métodos. Es un estudio observacional retrospectivo en pacientes mayores de 55 años con una primera crisis epiléptica de causa desconocida. Se recogieron los datos desde la historia clínica, incluyendo electroencefalograma (EEG) y resonancia magnética (RM) cerebral. Resultados. Se incluyó a 87 pacientes (58,6% varones; 71,5 ± 8,1 años). El seguimiento medio fue de 7,3 ± 4,9 años. El factor de riesgo vascular más frecuente fue la hipertensión arterial (77%; n = 67). Las crisis focales con alteración de la conciencia fueron el tipo de crisis más frecuente (44,8%; n = 39), seguidas de las crisis focales con evolución a bilaterales tonicoclónicas (39,1%; n = 34). La RM cerebral mostró atrofia cortical (50%; n = 42) y signos de enfermedad vascular de pequeño vaso (EVPV) (67,8%; n = 57). Se observaron anomalías epileptiformes intercríticas en el EEG en un 43,7% (n = 38) de los pacientes, mayoritariamente con localización temporal (94,7%; n = 36). Hasta un 44,8% (n = 39) presentaba deterioro cognitivo leve basalmente. La recurrencia de crisis, observada en 49 pacientes (56,1%), sucedió con una mediana de 12 meses (rango intercuartílico: 4,4-25,9). Finalmente, 71 pacientes (81,6%) desarrollaron epilepsia. Conclusión. El riesgo de epilepsia a largo plazo tras una crisis única de etiología desconocida en pacientes de edad avanzada es superior al 80%. La hipertensión arterial y el deterioro cognitivo leve en el inicio son las características clínicas más frecuentes. En la RM, la atrofia cortical y la presencia de EVPV son frecuentes, y los EEG de rutina no suelen mostrar alteraciones epileptiformes.

Performance investigation of epilepsy detection from noisy EEG signals using base-2-meta stacking classifier.

Epilepsy is a chronic neurological disease, characterized by spontaneous, unprovoked, recurrent seizures that may lead to long-term disability and premature death. Despite significant efforts made to improve epilepsy detection clinically and pre-clinically, the pervasive presence of noise in EEG signals continues to pose substantial challenges to their effective application. In addition, discriminant features for epilepsy detection have not been investigated yet. The objective of this study is to develop a hybrid model for epilepsy detection from noisy and fragmented EEG signals. We hypothesized that a hybrid model could surpass existing single models in epilepsy detection. Our approach involves manual noise rejection and a novel statistical channel selection technique to detect epilepsy even from noisy EEG signals. Our proposed Base-2-Meta stacking classifier achieved notable accuracy (0.98 ± 0.05), precision (0.98 ± 0.07), recall (0.98 ± 0.05), and F1 score (0.98 ± 0.04) even with noisy 5-s segmented EEG signals. Application of our approach to the specific problem like detection of epilepsy from noisy and fragmented EEG data reveals a performance that is not only superior to others, but also is translationally relevant, highlighting its potential application in a clinic setting, where EEG signals are often noisy or scanty. Our proposed metric DF-A (Discriminant feature-accuracy), for the first time, identified the most discriminant feature with models that give A accuracy or above (A = 95 used in this study). This groundbreaking approach allows for detecting discriminant features and can be used as potential electrographic biomarkers in epilepsy detection research. Moreover, our study introduces innovative insights into the understanding of these features, epilepsy detection, and cross-validation, markedly improving epilepsy detection in ways previously unavailable.