Search Results (25)

Epileptic seizure detection and classification from EEG recordings faces significant challenges due to extreme class imbalance. Analysis of the Temple University Hospital Seizure (TUSZ) dataset reveals imbalance ratios of 150:1 between common and rare seizure types, with high temporal heterogeneity (seizure durations of 1–1638 s). We propose a cascaded deep learning architecture with two specialized CNNs: a binary detector followed by a multi-class classifier. This approach decomposes the classification problem, reducing the maximum imbalance from 150:1 to manageable levels (9:1 binary, 5:1 type). The architecture implements a high-confidence filtering mechanism (threshold = 0.9), creating a 99.5% pure dataset for type classification, dynamic class-weighted optimization proportional to inverse class frequencies, and information flow refinement through progressive stages. Loss dynamics analysis reveals that our weighting scheme strategically redistributes optimization attention, reducing variance by 90.7% for majority classes while increasing variance for minority classes, ensuring all seizure types receive proportional learning signals regardless of representation. The binary classifier achieves 99.64% specificity and 98.23% sensitivity (ROC-AUC = 0.995). The type classifier demonstrates >99% accuracy across seven seizure categories with perfect (100%) classification for three seizure types despite minimal representation. Cross-dataset validation on the University of Bonn dataset confirms robust generalization (96.0% accuracy) for binary seizure detection. This framework effectively addresses multi-level imbalance in neurophysiological signal classification with hierarchical class structures. Full article

This article examines the complex relationship between seizures, epilepsy, and multiple sclerosis (MS) in pediatric patients, based on detailed findings from a single-center study. Background: Although multiple sclerosis is primarily recognized as an adult-onset disease, its occurrence in children presents distinctive challenges, especially related to seizure disorders. Methods: We reviewed 120 pediatric MS patients evaluated over 7 years; six of these (5%) experienced seizures (including one case of acute status epilepticus), and five were diagnosed with epilepsy according to the latest International League Against Epilepsy (ILAE) classification. This study aimed to evaluate the occurrence rates and types of seizures while investigating their management strategies in this specific group. Results: Through a detailed case analysis and patient follow-up, we identified key factors contributing to seizure onset and explored implications for treatment and care. In our cohort, children with MS and seizures showed a higher risk for disease progression and greater cumulative disability, evidenced by a significantly higher last Expanded Disability Status Scale (EDSS) score (after a minimum 2-year follow-up) in the seizure group (p < 0.006). The analysis recognized early MS onset and highly active disease types as further risk factors that led to worse health outcomes. Conclusions: Genetic causes of epilepsy in children are common and may interact with MS-related inflammation in the same patient; our observations underscore the need to investigate how these two conditions interact. This work contributes to the broader understanding of epilepsy comorbid with MS among pediatric patients, seeking to facilitate the creation of improved interdisciplinary clinical practices in pediatric neurology. Full article

In the medical field, there are several very different movement disorders, such as tremors, Parkinson’s disease, or Huntington’s disease. A wide range of motor and non-motor symptoms characterizes them. It is evident that in the modern era, the use of smart wrist devices, such as smartwatches, wristbands, and smart bracelets is spreading among all categories of people. This diffusion is justified by the limited costs, ease of use, and less invasiveness (and consequently greater acceptability) than other types of sensors used for health status monitoring. This systematic review aims to synthesize research studies using smart wrist devices for a specific class of movement disorders. Following PRISMA-S guidelines, 130 studies were selected and analyzed. For each selected study, information is provided relating to the smartwatch/wristband/bracelet model used (whether it is commercial or not), the number of end-users involved in the experimentation stage, and finally the characteristics of the benchmark dataset possibly used for testing. Moreover, some articles also reported the type of raw data extracted from the smart wrist device, the implemented designed algorithmic pipeline, and the data classification methodology. It turned out that most of the studies have been published in the last ten years, showing a growing interest in the scientific community. The selected articles mainly investigate the relationship between smart wrist devices and Parkinson’s disease. Epilepsy and seizure detection are also research topics of interest, while there are few papers analyzing gait disorders, Huntington’s Disease, ataxia, or Tourette Syndrome. However, the results of this review highlight the difficulties still present in the use of the smartwatch/wristband/bracelet for the identified categories of movement disorders, despite the advantages these technologies could bring in the dissemination of low-cost solutions usable directly within living environments and without the need for caregivers or medical personnel. Full article

Background/Objectives: Epilepsy is a major public health issue in Sub-Saharan Africa, particularly among children, due to limited healthcare resources, socioeconomic inequalities, and cultural stigma that often result in underdiagnosis and undertreatment. This review examines pediatric epilepsy’s diagnosis, classification, and management in this setting, highlighting the need for culturally appropriate interventions to improve care quality and address these challenges. Methods: A review of the literature was conducted using MEDLINE, Embase, Scopus, and Web of Science databases to identify pertinent studies published between 2013 and 2024. This review included studies examining the epidemiology, seizure classification and etiologies of epilepsy among children in Sub-Saharan Africa. Results: This review revealed higher incidence and prevalence of epilepsy in Sub-Saharan Africa compared to high-income countries, primarily attributable to factors such as infectious diseases, perinatal injuries, and limited diagnostic resources. The most frequently reported types of epilepsy were generalized and focal seizures, with significant etiological contributions from structural and infectious causes, including nodding syndrome and HIV-related epilepsy. The treatment gap remains considerable, with up to 80% of children not receiving appropriate antiseizure medications. Conclusions: The diagnosis and treatment of epilepsy in pediatric populations in Sub-Saharan Africa is complicated by several factors, including cultural stigma and the lack of adequate healthcare infrastructure. There is an urgent need for culturally tailored diagnostic tools, improved access to affordable treatments, and public health initiatives aimed at reducing stigma. Addressing these gaps through enhanced research, improved healthcare access, and targeted educational campaigns is crucial for improving the quality of life for children with epilepsy. Full article

Background: Evaluating the differential impact of vagus nerve stimulation (VNS) therapy across various seizure types, our study explores its efficacy specifically in patients with categorized minor and major seizures. Methods: We conducted a retrospective cohort study involving 76 patients with pharmacoresistant epilepsy treated at the University Emergency Hospital of Bucharest between 2021 and 2024. Seizures were classified as ‘minor’ (including focal-aware and non-motor/absence seizures) and ‘major’ (including focal to bilateral tonic-clonic and generalized motor seizures), based on modified International League Against Epilepsy (ILAE) criteria. This classification allowed us to assess the response to VNS therapy, defined by a 50% or greater reduction in seizure frequency at the 12-month follow-up. Results: Our findings reveal that major seizures respond more favorably to VNS therapy, significantly reducing both frequency and intensity. In contrast, minor seizures showed a less pronounced response in frequency reduction but noted improvements in neurocognitive functions, suggesting a nuanced benefit of VNS in these cases. Conclusion: The study underscores the importance of seizure type in determining the efficacy of VNS therapy, advocating for personalized treatment approaches based on seizure classification. This approach could potentially enhance clinical outcomes by tailoring VNS settings to specific seizure types, improving overall management strategies in pharmacoresistant epilepsy. Full article

With the widespread use of the 13-valent pneumonia vaccine (PCV13) in China, monitoring adverse events following immunization (AEFIs) is critical. We conducted a descriptive analysis of the AEFI occurrences reported within Hangzhou between the years 2020 and 2023, including the temporal trend of case reports and variables such as sex, age, type of PCV13, dose number, type of reporter, cause-specific classification, severity, and onset from vaccination. Vaccine safety signals were analyzed using reporting odds ratios (RORs). Over the 4 years analyzed in the study, 2564 AEFI cases were reported, including seven severe cases. Most AEFIs occurred within 0–1 days after vaccination (2398, 93.53%), with over half affecting infants aged 1.5–6 months of age. No statistically significant difference was observed between PCV13-TT and PCV-CRM197. Seasonal differences in AEFI reports were noted. Positive signals were detected for fever (ROR-1.96SE: 1.64) and persistent crying (ROR-1.96SE: 1.61). Four serious AEFI cases were coincidental events, while three others were considered vaccine-related cases (including one case each of allergic reaction, febrile seizure, and thrombocytopenia). The safety and tolerability of PCV13 are good, and attention should be paid to severe AEFIs, as well as long-term safety disparities between different types of PCV13. Full article

With limited sample sizes and varying study outcomes regarding complete blood count (CBC)-associated biomarkers and their febrile seizure (FS) classification, along with limited research on osmolarity, this study aims to evaluate CBC-associated biomarkers, including osmolarity, for a comprehensive view of their diagnostic value. This single-center retrospective study used data from 364 children (aged 5–60 months) diagnosed with FS. The patients were categorized into simple FS (n = 221) and complex FS (n = 143) groups. CBC and biochemical tests, including sodium, potassium, chloride, glucose, blood urea nitrogen, and C-reactive protein levels, were evaluated. The neutrophil-to-lymphocyte ratio (NLR), mean platelet volume-to-lymphocyte ratio, and osmolarity were calculated and compared between FS types and the number of seizures. Receiver operating characteristic (ROC) curve analysis was conducted to assess the predictive utility of these markers. Inflammatory markers, including NLR, were ineffective in predicting FS types. Complex FS cases exhibited a significantly lower osmolarity than simple FS cases. The area under the ROC curve for osmolarity to distinguish complex FS was 0.754, while other markers did not reach the desired threshold of 0.700. Including osmolarity in the classification of FS has clinical applicability. Physicians may consider osmolarity as an additional tool to aid in clinical decision-making. Full article

Deep learning (DL) is emerging as a successful technique for automatic detection and differentiation of spontaneous seizures that may otherwise be missed or misclassified. Herein, we propose a system architecture based on top-performing DL models for binary and multigroup classifications with the non-overlapping window technique, which we tested on the TUSZ dataset. The system accurately detects seizure episodes (87.7% Sn, 91.16% Sp) and carefully distinguishes eight seizure types (95–100% Acc). An increase in EEG sampling rate from 50 to 250 Hz boosted model performance: the precision of seizure detection rose by 5%, and seizure differentiation by 7%. A low sampling rate is a reasonable solution for training reliable models with EEG data. Decreasing the number of EEG electrodes from 21 to 8 did not affect seizure detection but worsened seizure differentiation significantly: 98.24 ± 0.17 vs. 85.14 ± 3.14% recall. In detecting epileptic episodes, all electrodes provided equally informative input, but in seizure differentiation, their informative value varied. We improved model explainability with interpretable ML. Activation maximization highlighted the presence of EEG patterns specific to eight seizure types. Cortical projection of epileptic sources depicted differences between generalized and focal seizures. Interpretable ML techniques confirmed that our system recognizes biologically meaningful features as indicators of epileptic activity in EEG. Full article

The accurate classification of seizure types using electroencephalography (EEG) signals plays a vital role in determining a precise treatment plan and therapy for epilepsy patients. Among the available deep network models, Convolutional Neural Networks (CNNs) are the most widely adopted models for learning and representing EEG signals. However, typical CNNs have high computational complexity, leading to overfitting problems. This paper proposes the design of two effective, lightweight deep network models; the 1D multiscale neural network (1D-MSCNet) model and the Long Short-term Memory (LSTM)-based compact CNN (EEG-LSTMNet) model. The 1D-MSCNet model comprises three modules: a spectral–temporal convolution module, a spatial convolution module, and a classification module. It extracts features from input EEG trials at multiple frequency/time ranges, identifying relationships between the spatial distribution of their channels. The EEG-LSTMNet model includes three convolutional layers, namely temporal, depthwise, and separable layers, a single LSTM layer, and two fully connected classification layers to extract discriminative EEG feature representations. Both models have been applied to the same EEG trials collected from the Temple University Hospital (TUH) database. Results revealed F1-score values of 96.9% and 98.4% for the 1D-MSCNet and EEG-LSTMNet, respectively. Based on the demonstrated outcomes, both models outperform related state-of-the-art methods due to their architectures’ adoption of 1D modules and layers that reduce the computational effort needed, solve the overfitting problem, and enhance classification efficiency. Hence, both models could be valuable additions for neurologists to help them decide upon precise treatments and drugs for patients depending on their type of seizure. Full article

Accurately identifying the seizure type is vital in the treatment plan and drug prescription for epileptic patients. The most commonly adopted test for identifying epileptic seizures is electroencephalography (EEG). EEG signals include important information about the brain’s electrical activities and are widely used for epilepsy analysis. Among various deep network architectures, convolutional neural networks (CNNs) have been widely used for EEG signal representation learning for epilepsy analysis. However, most of the existing CNN-based methods suffer from the overfitting problem due to a small number of EEG trials and the huge number of learnable parameters. This paper introduces the design of an efficient, lightweight, and expressive deep network model based on ResNet theory and long short-term memory (LSTM) for classifying seizure types from EEG trials. A 1D ResNet module is adopted to train a deeper network without encountering vanishing gradient problems and to avoid the overfitting problem of CNN models. The LSTM module encodes and learns long-term dependencies over time. The synthetic minority oversampling technique (SMOTE) is applied to balance the data by increasing the trials of minority classes. The proposed method was evaluated using the public domain benchmark TUH database. Experimental results revealed the superior performance of the proposed model over other state-of-the-art models with an F1-score of 97.4%. The proposed deep learning model will help neurologists precisely interpret and classify epileptic seizure types and enhance the patient’s life. Full article

Objectives: The purpose of this research was to investigate whether MRI and Simultaneous Hybrid PET/MRI images were consistent in the histological classification of patients with focal cortical dysplasia. Additionally, this research aimed to evaluate the postoperative outcomes with the MRI and Simultaneous Hybrid PET/MRI images of focal cortical dysplasia. Methods: A total of 69 cases in this research were evaluated preoperatively for drug-resistant seizures, and then surgical resection procedures of the epileptogenic foci were performed. The postoperative result was histopathologically confirmed as focal cortical dysplasia, and patients then underwent PET and MRI imaging within one month of the seizure. In this study, head MRI was performed using a 3.0 T magnetic resonance scanner (Philips) to obtain 3D T1WI images. The Siemens Biograph 16 scanner was used for a routine scanning of the head to obtain PET images. BrainLAB’s iPlan software was used to fuse 3D T1 images with PET images to obtain PET/MRI images. Results: Focal cortical dysplasia was divided into three types according to ILAE: three patients were classified as type I, twenty-five patients as type II, and forty-one patients as type III. Patients age of onset under 18 and age of operation over 18 had a longer duration (p = 0.036, p = 0.021). MRI had a high lesion detection sensitivity of type III focal cortical dysplasia (p = 0.003). Simultaneous Hybrid PET/MRI showed high sensitivity in detecting type II and III focal cortical dysplasia lesions (p = 0.037). The lesions in Simultaneous Hybrid PET/MRI-positive focal cortical dysplasia patients were mostly located in the temporal and multilobar (p = 0.005, 0.040). Conclusion: Simultaneous Hybrid PET/MRI has a high accuracy in detecting the classification of focal cortical dysplasia. The results of this study indicate that patients with focal cortical dysplasia with positive Simultaneous Hybrid PET/MRI have better postoperative prognoses. Full article

Focal cortical dysplasia (FCD) represents a heterogeneous group of morphological changes in the brain tissue that can predispose the development of pharmacoresistant epilepsy (recurring, unprovoked seizures which cannot be managed with medications). This group of neurological disorders affects not only the cerebral cortex but also the subjacent white matter. This work reviews the literature describing the morphological substrate of pharmacoresistant epilepsy. All illustrations presented in this study are obtained from brain biopsies from refractory epilepsy patients investigated by the authors. Regarding classification, there are three main FCD types, all of which involve cortical dyslamination. The 2022 revision of the International League Against Epilepsy (ILAE) FCD classification includes new histologically defined pathological entities: mild malformation of cortical development (mMCD), mild malformation of cortical development with oligodendroglial hyperplasia in frontal lobe epilepsy (MOGHE), and “no FCD on histopathology”. Although the pathomorphological characteristics of the various forms of focal cortical dysplasias are well known, their aetiologic and pathogenetic features remain elusive. The identification of genetic variants in FCD opens an avenue for novel treatment strategies, which are of particular utility in cases where total resection of the epileptogenic area is impossible. Full article

Defect of GTPBP3, the human mitochondrial tRNA-modifying enzyme, can lead to Combined Oxidative Phosphorylation Deficiency 23 (COXPD23). Up to now, about 20 different variants of the GTPBP3 gene have been reported; however, genotype–phenotype analysis has rarely been described. Here, we reported a 9-year-old boy with COXPD23 who presented with hyperlactatemia, hypertrophic cardiomyopathy, seizures, feeding difficulties, intellectual disability and motor developmental delay, and abnormal visual development. Biallelic pathogenic variants of the GTPBP3 gene were identified in this boy, one novel variant c.1102dupC (p. Arg368Profs*22) inherited from the mother and the other known variant c.689A>C (p. Gln230Pro) inherited from father. We curated 18 COXPD23 patients with GTPBP3 variants to investigate the genotype–phenotype correlation. We found that hyperlactatemia and cardiomyopathy were critical clinical features in COXPD23 and the average onset age was 1.7 years (3 months of age for the homozygote). Clinical classification of COXPD23 for the two types, severe and mild, was well described in this study. We observed arrhythmia and congestive heart failure frequently in the severe type with early childhood mortality, while developmental delay was mainly observed in the mild type. The proportion of homozygous variants (71.4%) significantly differed from that of compound heterozygous variants (18.1%) in the severe type. Compared with the variants in gnomAD, the proportion of LOFVs in GTPBP3 was higher in COXPD23 patients (48.6% versus 8.9%, p < 0.0001 ****), and 31% of them were frameshift variants, showing the LOF mechanism of GTPBP3. Additionally, the variants in patients were significantly enriched in the TrmE-type G domain, indicating that the G domain was crucial for GTPBP3 protein function. The TrmE-type G domain contained several significant motifs involved in the binding of guanine nucleotides and Mg²⁺, the hydrolysis of GTP, and the regulation of the functional status of GTPases. In conclusion, we reported a mild COXPD23 case with typical GTPBP3-related symptoms, including seizures and abnormal visual development seldom observed previously. Our study provides novel insight into understanding the clinical diagnosis and genetic counseling of patients with COXPD23 by exploring the genetic pathogenesis and genotype–phenotype correlation of COXPD23. Full article

Genetic early-onset Parkinsonism is unique due to frequent co-occurrence of hyperkinetic movement disorder(s) (MD), or additional neurological of systemic findings, including epilepsy in up to 10–15% of cases. Based on both the classification of Parkinsonism in children proposed by Leuzzi and coworkers and the 2017 ILAE epilepsies classification, we performed a literature review in PubMed. A few discrete presentations can be identified: Parkinsonism as a late manifestation of complex neurodevelopmental disorders, characterized by developmental and epileptic encephalopathies (DE-EE), with multiple, refractory seizure types and severely abnormal EEG characteristics, with or without preceding hyperkinetic MD; Parkinsonism in the context of syndromic conditions with unspecific reduced seizure threshold in infancy and childhood; neurodegenerative conditions with brain iron accumulation, in which childhood DE-EE is followed by neurodegeneration; and finally, monogenic juvenile Parkinsonism, in which a subset of patients with intellectual disability or developmental delay (ID/DD) develop hypokinetic MD between 10 and 30 years of age, following unspecific, usually well-controlled, childhood epilepsy. This emerging group of genetic conditions leading to epilepsy or DE-EE in childhood followed by juvenile Parkinsonism highlights the need for careful long-term follow-up, especially in the context of ID/DD, in order to readily identify individuals at increased risk of later Parkinsonism. Full article

In this work, an attempt has been made to develop an automated system for detecting electroclinical seizures such as tonic-clonic seizures, complex partial seizures, and electrographic seizures (EGSZ) using higher-order moments of scalp electroencephalography (EEG). The scalp EEGs of the publicly available Temple University database are utilized in this study. The higher-order moments, namely skewness and kurtosis, are extracted from the temporal, spectral, and maximal overlap wavelet distributions of EEG. The features are computed from overlapping and non-overlapping moving windowing functions. The results show that the wavelet and spectral skewness of EEG is higher in EGSZ than in other types. All the extracted features are found to have significant differences (p < 0.05), except for temporal kurtosis and skewness. A support vector machine with a radial basis kernel designed using maximal overlap wavelet skewness yields a maximum accuracy of 87%. In order to improve the performance, the Bayesian optimization technique is utilized to determine the suitable kernel parameters. The optimized model achieves the highest accuracy of 96% and an MCC of 91% in three-class classification. The study is found to be promising, and it could facilitate the rapid identification process of life-threatening seizures. Full article