OBJECTIVE: To identify the optimal machine learning algorithm for predicting post-stroke epilepsy (PSE) within one year following acute stroke, establish a nomogram model based on this algorithm, and perform external validation to achieve accurate prediction of secondary epilepsy. METHODS: A total of 870 acute stroke patients admitted to the emergency department of Xiang'an Hospital of Xiamen University from June 2019 to June 2023 were enrolled for model development (model group). An external validation cohort of 435 acute stroke patients admitted to the Fifth Hospital of Xiamen during the same period was used to validate the machine learning algorithms and nomogram model. Patients were classified into control and epilepsy groups based on the development of PSE within one year. Clinical and laboratory data, including baseline characteristics, stroke location, vascular status, complications, hematologic parameters, and National Institutes of Health Stroke Scale (NIHSS) score, were collected for analysis. Nine machine learning algorithms such as logistic regression, CN2 rule induction, K-nearest neighbors, adaptive boosting, random forest, gradient boosting, support vector machine, naive Bayes, and neural network were applied to evaluate predictive performance. The area under the curve (AUC) of receiver operator characteristic curve (ROC curve) was used to identify the optimal algorithm. Logistic regression was used to screen risk factors for PSE, and the top 10 predictors were selected to construct the nomogram model. The predictive performance of the model was evaluated using the ROC curve in both the model and validation groups. RESULTS: Among the 870 patients in the model group, 29 developed PSE within one year. Among the nine algorithms tested, logistic regression demonstrated the best performance and generalizability, with an AUC of 0.923. Univariate logistic regression identified several risk factors for PSE, including platelet count, white blood cell count, red blood cell count, glycated hemoglobin (HbA1c), C-reactive protein (CRP), triglycerides, high-density lipoprotein (HDL), aspartate aminotransferase (AST), alanine aminotransferase (ALT), activated partial thromboplastin time (APTT), thrombin time, D-dimer, fibrinogen, creatine kinase (CK), creatine kinase-MB (CK-MB), lactate dehydrogenase (LDH), serum sodium, lactic acid, anion gap, NIHSS score, brain herniation, periventricular stroke, and carotid artery plaque. Further multivariate logistic regression analysis showed that white blood cell count, HDL, fibrinogen, lactic acid and brain herniation were independent risk factors [odds ratio (OR) were 1.837, 198.039, 47.025, 11.559, 70.722, respectively, all P < 0.05]. In the external validation group, univariate logistic regression analysis showed that platelet count, white blood cell count, CRP, triacylglycerol, APTT, D-dimer, fibrinogen, CK, CK-MB, LDH, NIHSS score, and cerebral herniation were risk factors for PSE one year after acute stroke. Further multiple logistic regression analysis showed that APTT and cerebral herniation were independent predictors (OR were 0.587 and 116.193, respectively, both P < 0.05). The nomogram model, constructed using 10 key variables-brain herniation, periventricular stroke, carotid artery plaque, white blood cell count, triglycerides, thrombin time, D-dimer, serum sodium, lactic acid, and NIHSS score-achieved an AUC of 0.908 in the model group and 0.864 in the external validation group. CONCLUSIONS The logistic regression-based prediction model for epilepsy one year after acute stroke, developed using machine learning algorithms, showed optimal predictive performance. The nomogram model based on the logistic regression-derived predictors showed strong discriminative power and was successfully validated externally, suggesting favorable clinical applicability and generalizability.
Humans ; Machine Learning ; Stroke/complications* ; Nomograms ; Epilepsy/etiology* ; Algorithms ; Male ; Female ; Logistic Models ; Middle Aged ; Aged ; Risk Factors ; Bayes Theorem

Psychogenic non-epileptic seizures are accompanied by motor,behavioral,sensory,and/or cognitive changes,with the clinical manifestations similar to epileptic seizures.This disease is easy to be misdiagnosed and neglected in clinical work.At present,most intervention measures still depend on the experience of clinicians.This article reviews the comorbidities of psychogenic non-epileptic seizures,including mental and cognitive disorders,somatic syndrome,sleep disorders,and epilepsy.This review aims to strengthen the precision of clinical treatment and management of patients with psychogenic non-epileptic seizures and provide more efficient individualized diagnosis and treatment programs for patients.
Humans ; Seizures/diagnosis* ; Comorbidity ; Epilepsy ; Sleep Wake Disorders ; Mental Disorders ; Psychophysiologic Disorders ; Cognition Disorders

The zebrafish model has attracted much attention due to its strong reproductive ability, short research cycle, and ease of maintenance. It has always been an important vertebrate model system, often used to carry out human disease research. Its motor behavior features have the advantages of being simpler, more intuitive, and quantifiable. In recent years, it has received widespread attention in the study of traditional Chinese medicine(TCM)for the treatment of sleep disorders, neurodegenerative diseases, fatigue, epilepsy, and other diseases. This paper reviews the characteristics of zebrafish motor behavior and its applications in the pharmacodynamic verification and mechanism research of TCM extracts, active ingredients, and TCM compounds, as well as in active ingredient screening and safety evaluation. The paper also analyzes its advantages and disadvantages, with the aim of improving the breadth and depth of zebrafish and its motor behavior applications in the field of TCM research.
Zebrafish/physiology* ; Medicine, Chinese Traditional ; Drugs, Chinese Herbal/therapeutic use* ; Disease Models, Animal ; Drug Evaluation, Preclinical/methods* ; Animals ; Sleep Wake Disorders/physiopathology* ; Epilepsy/physiopathology* ; Neurodegenerative Diseases/physiopathology* ; Fatigue/physiopathology* ; Behavior, Animal/physiology* ; Motor Activity/physiology*

The existing epilepsy seizure detection algorithms have problems such as overfitting and poor generalization ability due to high reliance on manual labeling of electroencephalogram's data and data imbalance between seizure and interictal periods. An unsupervised learning detection method for epileptic seizure that jointed graph attention network (GAT) and Transformer framework (GAT-T) was proposed. In this method, channel correlations were adaptively learned by GAT encoder. Temporal information was captured by one-dimensional convolution decoder. Combining outputs of the two mentioned above, predicted values for electroencephalogram were generated. The collective anomaly score was calculated and the detection threshold was determined. The results demonstrated that GAT-T achieved the average performance exceeding 90% (or 99%) with a 0.25 s (or 2 s) time segment length, which could effectively detect epileptic seizures. Moreover, the channel association probability matrix was expected to assist clinicians in the initial screening of the epileptogenic zone, and ablation experiments also reflected the significance of each module in GAT-T. This study may assist clinicians in making more accurate diagnostic and therapeutic decisions for epilepsy patients.
Humans ; Electroencephalography/methods* ; Epilepsy/physiopathology* ; Algorithms ; Seizures/physiopathology* ; Neural Networks, Computer ; Signal Processing, Computer-Assisted

Current epilepsy prediction methods are not effective in characterizing the multi-domain features of complex long-term electroencephalogram (EEG) data, leading to suboptimal prediction performance. Therefore, this paper proposes a novel multi-scale sparse adaptive convolutional network based on multi-head attention mechanism (MS-SACN-MM) model to effectively characterize the multi-domain features. The model first preprocesses the EEG data, constructs multiple convolutional layers to effectively avoid information overload, and uses a multi-layer perceptron and multi-head attention mechanism to focus the network on critical pre-seizure features. Then, it adopts a focal loss training strategy to alleviate class imbalance and enhance the model's robustness. Experimental results show that on the publicly created dataset (CHB-MIT) by MIT and Boston Children's Hospital, the MS-SACN-MM model achieves a maximum accuracy of 0.999 for seizure prediction 10 ~ 15 minutes in advance. This demonstrates good predictive performance and holds significant importance for early intervention and intelligent clinical management of epilepsy patients.
Humans ; Electroencephalography/methods* ; Epilepsy/physiopathology* ; Neural Networks, Computer ; Seizures/physiopathology* ; Signal Processing, Computer-Assisted ; Algorithms

OBJECTIVES: To investigate the efficacy and safety of perampanel (PER) add-on therapy in children with epilepsy of genetic etiology. METHODS: A retrospective analysis was conducted on the clinical data of 53 children who attended the Department of Neurology, Wuhan Children's Hospital, from November 2020 to April 2023. All children received PER add-on therapy and were diagnosed with epilepsy of genetic etiology based on whole-exome sequencing. The primary outcome measure was the proportion of children with a reduction in seizure frequency of ≥50% at month 12 of PER treatment (i.e., response rate), and the secondary outcome measures were response rates at months 3 and 6 of treatment. The influencing factors for the efficacy of PER add-on therapy in the treatment of epilepsy of genetic etiology were analyzed, and adverse events were recorded. RESULTS: The median follow-up duration was 13.10 months. After 12 months of follow-up, 42 children were included in the analysis, comprising 25 boys (60%) and 17 girls (40%). The median initial dose of PER was 1.5 (1.0, 2.0) mg/d, and the median maintenance dose was 4.0 (3.0, 8.0) mg/d. The response rates to PER at months 3, 6, and 12 of treatment were 61% (30/49), 54% (25/46), and 48% (20/42), respectively. No significant difference in the efficacy of PER was observed between children with mutations in genes encoding different protein functions (P>0.05). The most common adverse event reported was fatigue, observed in 3 children (6%). CONCLUSIONS PER add-on therapy demonstrates good efficacy and safety in children with epilepsy of genetic etiology. No influencing factors for the efficacy of PER have been identified to date.
Humans ; Male ; Female ; Nitriles ; Child ; Pyridones/administration & dosage* ; Child, Preschool ; Retrospective Studies ; Anticonvulsants/administration & dosage* ; Epilepsy/etiology* ; Adolescent ; Infant ; Drug Therapy, Combination

OBJECTIVES: To summarize the clinical and genetic characteristics of epilepsy with intellectual disability caused by SETD1B gene variants in children. METHODS: A retrospective analysis was conducted on the clinical data of three children with SETD1B gene variants diagnosed and treated at the Department of Pediatric Neurology of Xiangya Hospital of Central South University. Relevant literature was reviewed to summarize the clinical characteristics of this condition. RESULTS: All three children presented with symptoms during infancy or early childhood, including mild intellectual disability and myoclonic seizures, with two cases exhibiting eyelid myoclonia. After treatment with three or more antiepileptic drugs, two cases achieved seizure control or partial control, while one case remained refractory. Each of the three children was found to have a heterozygous variant in the SETD1B gene (one deletion, one frameshift, and one missense variant). To date, 54 cases with SETD1B gene variants have been reported, involving a total of 56 variants, predominantly missense variants (64%, 36/56). The main clinical manifestations included varying degrees of developmental delay (96%, 52/54) and seizures (81%, 44/54). Among the 44 patients with seizures, myoclonic (20%, 9/44) and absence seizures (34%, 15/44) were common, with eyelid myoclonia reported in six cases. Approximately one-fifth of these patients had poorly controlled seizures. CONCLUSIONS The primary phenotypes associated with SETD1B gene variants are intellectual disability and seizures, and seizures exhibit distinct characteristics. Eyelid myoclonia is not uncommon.
Humans ; Intellectual Disability/complications* ; Epilepsy/complications* ; Male ; Female ; Histone-Lysine N-Methyltransferase/genetics* ; Child, Preschool ; Child ; Retrospective Studies

The patient was a girl, aged 10 years, who was admitted due to fever for 5 days and pancytopenia in peripheral blood for 2 days. Bone marrow examination showed the presence of phagocytic activity, and peripheral blood tests showed pancytopenia, an increase in ferritin, a reduction in fibrinogen, increases in triglyceride and sCD25, and a reduction in natural killer cell activity, which led to the diagnosis of hemophagocytic lymphohistiocytosis (HLH). On the day of admission, the child developed convulsions and rapidly progressed to refractory status epilepticus, which was consistent with the manifestations of febrile infection-related epilepsy syndrome. HLH was controlled after active immunotherapy, with the sequela of refractory epilepsy, and her cognitive function was essentially within normal limits. This article reports the condition of febrile infection-related epilepsy syndrome caused by HLH for the first time in China, in order to improve the awareness of this disease among clinicians.
Humans ; Lymphohistiocytosis, Hemophagocytic/complications* ; Female ; Child ; Epilepsy/etiology* ; Fever/etiology* ; Epileptic Syndromes/etiology*

BACKGROUND AND OBJECTIVES

Epilepsy is a very common pediatric neurologic disorder, and the mainstay of treatment is the use of anti-seizure medication. Several factors may cause inadequate adherence leading to uncontrolled seizures, lower quality of life, and neurodevelopmental consequences. This study aimed to determine medication adherence of adolescents with epilepsy and identify factors that may be associated in medication adherence.

This is a prospective cross-sectional study involving adolescents with epilepsy. A self-reported survey was used to measure adherence. Data on demographics and epilepsy were then assessed for presence of association with adherence.

Fifty-one participants were included. Of these, 19.6% were non-adherent, 35.3% had medium adherence, and 45.1% had high adherence. Simple logistic regression analysis showed that unemployed primary caregiver is associated with 7.0 times higher odds of having moderate-high adherence and consuming at least three drugs is associated with 0.3 lower odds of having moderate-high adherence.

As high as 80.4% of adolescents were adherent to their medications. The presence of a caregiver who can closely monitor the patient is associated with adherence while intake of several drugs is associated with nonadherence. Future studies may need larger sample size and explore knowledge, attitude, and other social factors that may influence medication adherence.

OBJECTIVE: To analyze the clinical characteristics and genetic etiology of a child with Christianson syndrome (CS). METHODS: A 1-year-and-5-month-old boy with CS diagnosed at the First Affiliated Hospital of Zhengzhou University in April 2021 was selected as the study subject. Clinical data were retrospectively analyzed. Peripheral blood samples were obtained from the child and his parents, followed by genomic DNA extraction and whole exome sequencing (WES). Candidate variant was validated by Sanger sequencing. This study has been approved by the Medical Ethics Committee of the Hospital of Zhengzhou University (Ethics No. 2024-KY-1103-001). RESULTS: The child has manifested with seizures, microcephaly, and global developmental delay. WES revealed that he has harbored a novel de novo hemizygous nonsense variant of the SLC9A6 gene, namely c.1014G>A (p.W338*). Based on the guidelines from the American College of Medical Genetics and Genomics (ACMG), the variant was rated as pathogenic. CONCLUSION The hemizygous c.1014G>A nonsense variant of the SLC9A6 gene probably underlay the pathogenesis in this child. Above discovery has expanded mutational spectrum of the SLC9A6 gene and enabled definite diagnosis of the child.
Humans ; Male ; Infant ; Microcephaly/genetics* ; Spasms, Infantile/genetics* ; Sodium-Hydrogen Exchangers/genetics* ; Exome Sequencing ; Intellectual Disability/genetics* ; Genetic Diseases, X-Linked/genetics* ; Mutation ; Seizures/genetics* ; Ataxia ; Epilepsy ; Ocular Motility Disorders