Crize epileptice și managementul lor.
127 de articoleThis study proposes a deep learning approach for predicting epileptic seizures using EEG signal analysis, employing convolutional neural networks for feature extraction and Bi-LSTM for classification. The method achieved 99.61% accuracy and aims to improve patient quality of life through early seizure detection and timely intervention.
This systematic review of 37 studies examines how machine learning models can predict individual patient responses to anti-seizure medications by analyzing clinical data, brain imaging, EEG, and genetic information. While some models show promising accuracy, results vary widely, and the field faces significant limitations related to data quality and quantity that require further research and standardization.
This study examines temporal patterns and prognostic factors in seizures associated with glioblastoma, a primary brain malignancy. The research aims to characterize how seizure presentation timing relates to patient outcomes and disease progression.
This pharmacovigilance study analyzed FDA adverse event reports to investigate whether ALK inhibitors (crizotinib, ceritinib, alectinib, brigatinib, lorlatinib) are associated with seizures using Bayesian statistical methods. Results showed varying signal detection scores across different ALK inhibitors, with ceritinib, brigatinib, alectinib, and lorlatinib demonstrating stronger associations with seizure disorders compared to crizotinib.
This retrospective study analyzed EEG recordings to determine optimal recording duration for detecting epileptiform abnormalities in patients with first seizure versus established epilepsy. Results showed detection rates of 37.6% and 57.4% respectively, with factors like anti-seizure medication use and seizure frequency significantly influencing detection, suggesting individualized EEG protocols based on clinical presentation.
This systematic review and meta-analysis examined alterations in gut microbiota composition in epilepsy patients compared to healthy controls across multiple databases. While alpha diversity did not differ significantly, specific bacterial taxa showed significant changes including increased Verrucomicrobia and Akkermansia but decreased Lactobacillus in epilepsy patients.
This study presents a machine learning model designed to anticipate epileptic seizures by analyzing EEG scalogram images through a multi-serial cascaded network with optimized feature selection. The approach uses the Archimedes Optimization algorithm for feature selection followed by Bi-directional LSTM processing to provide timely seizure detection.
This scientific commentary examines how anti-seizure medications affect brain network organization following an initial seizure event. The work provides insights into neurobiological mechanisms underlying individual differences in medication response in epilepsy patients.
This article addresses alcohol withdrawal seizures as a serious medical complication, explores the relationship between chronic alcohol use and epilepsy development, and examines how traumatic brain injury can increase seizure risk. The content covers clinically recognized mechanisms linking substance withdrawal, neurological injury, and seizure disorders.
This study examines 13 events in 8 patients initially misdiagnosed with convulsive status epilepticus who actually had psychogenic non-epileptic seizures (PNES). The research highlights that poor response to benzodiazepines and anti-seizure medications should prompt consideration of PNES as a differential diagnosis to avoid unnecessary intubation and inappropriate treatment escalation.
SCN3A-related neurodevelopmental disorder is a rare genetic condition presenting with developmental delays and seizures of varying severity, ranging from isolated speech delay to severe developmental and epileptic encephalopathy (DEE). Seizures typically begin in infancy and are often drug-resistant, particularly in cases with cortical malformations, with additional complications including hypotonia, movement disorders, and potential autonomic dysfunction.
This review addresses the clinical challenge of drug-resistant frontal lobe epilepsy, the second most common epilepsy type, which is complicated by the frontal lobe's large size making seizure focus localization difficult. The article examines epidemiology, etiologic factors (particularly cortical developmental abnormalities), symptomatology, and advances in neurophysiological and neuroimaging techniques including magnetoencephalography, quantitative PET, 7T MRI, and artificial intelligence-assisted morphometric analysis to improve lesion detection and surgical outcomes.
This review examines how genetic variations in Arab populations influence the efficacy, safety, and individual responses to anti-seizure medications, identifying specific genetic markers that can predict drug resistance and optimize dosing. Genetic testing for variants in genes such as MTHFR, MDR1, ABCB1, and others enables personalized treatment approaches to improve seizure control and minimize adverse drug reactions.
The International League Against Epilepsy released an updated seizure classification in 2025 with six main revisions to improve clinical clarity, including replacing 'onset' in seizure class names, substituting 'consciousness' for 'awareness' to encompass both awareness and responsiveness, and replacing 'motor' with 'observable' features. The update also emphasizes chronological seizure semiology patterns, recognizes epileptic negative myoclonus as a new seizure type, and expands the use of epileptic spasms as a descriptor across focal and unknown seizures.
This survey examines current clinical practices among Italian epileptologists regarding prophylactic anti-seizure medication use in brain tumor patients. The study provides insights into specialist approaches to seizure prevention in this high-risk population.
Psychiatric disorders commonly coexist with epilepsy but remain underdiagnosed and undertreated due to insufficient mental health resources and neurologist training gaps. This clinical resource provides neurologists with evidence-based strategies to identify and pharmacologically manage reactive symptoms, interictal psychiatric disorders, periictal episodes, and iatrogenic psychiatric effects in their epilepsy patients.