Early Seizure Detection Research Articles

Subdural Electroencephalogram Monitoring for Early Seizure Detection in Patients With Subdural Hematoma: Considering It in the Real World.

Subdural Electroencephalogram Monitoring for Early Seizure Detection in Patients With Subdural Hematoma: Considering It in the Real World.

HyEpiSeiD: a hybrid convolutional neural network and gated recurrent unit model for epileptic seizure detection from electroencephalogram signals

Epileptic seizure (ES) detection is an active research area, that aims at patient-specific ES detection with high accuracy from electroencephalogram (EEG) signals. The early detection of seizure is crucial for timely medical intervention and prevention of further injuries of the patients. This work proposes a robust deep learning framework called HyEpiSeiD that extracts self-trained features from the pre-processed EEG signals using a hybrid combination of convolutional neural network followed by two gated recurrent unit layers and performs prediction based on those extracted features. The proposed HyEpiSeiD framework is evaluated on two public datasets, the UCI Epilepsy and Mendeley datasets. The proposed HyEpiSeiD model achieved 99.01% and 97.50% classification accuracy, respectively, outperforming most of the state-of-the-art methods in epilepsy detection domain.

Seizure Detection of EEG Signals Based on Multi-Channel Long- and Short-Term Memory-Like Spiking Neural Model.

Seizure is a common neurological disorder that usually manifests itself in recurring seizure, and these seizures can have a serious impact on a person's life and health. Therefore, early detection and diagnosis of seizure is crucial. In order to improve the efficiency of early detection and diagnosis of seizure, this paper proposes a new seizure detection method, which is based on discrete wavelet transform (DWT) and multi-channel long- and short-term memory-like spiking neural P (LSTM-SNP) model. First, the signal is decomposed into 5 levels by using DWT transform to obtain the features of the components at different frequencies, and a series of time-frequency features in wavelet coefficients are extracted. Then, these different features are used to train a multi-channel LSTM-SNP model and perform seizure detection. The proposed method achieves a high seizure detection accuracy on the CHB-MIT dataset: 98.25% accuracy, 98.22% specificity and 97.59% sensitivity. This indicates that the proposed epilepsy detection method can show competitive detection performance.

Evaluating the Impact of Point-of-Care Electroencephalography on Length of Stay in the Intensive Care Unit: Subanalysis of the SAFER-EEG Trial.

Electroencephalography (EEG) is needed to diagnose nonconvulsive seizures. Prolonged nonconvulsive seizures are associated with neuronal injuries and deleterious clinical outcomes. However, it is uncertain whether the rapid identification of these seizures using point-of-care EEG (POC-EEG) can have a positive impact on clinical outcomes. In a retrospective subanalysis of the recently completed multicenter Seizure Assessment and Forecasting with Efficient Rapid-EEG (SAFER-EEG) trial, we compared intensive care unit (ICU) length of stay (LOS), unfavorable functional outcome (modified Rankin Scale score ≥ 4), and time to EEG between adult patients receiving a US Food and Drug Administration-cleared POC-EEG (Ceribell, Inc.) and those receiving conventional EEG (conv-EEG). Patient records from January 2018 to June 2022 at three different academic centers were reviewed, focusing on EEG timing and clinical outcomes. Propensity score matching was applied using key clinical covariates to control for confounders. Medians and interquartile ranges (IQRs) were calculated for descriptive statistics. Nonparametric tests (Mann-Whitney U-test) were used for the continuous variables, and the χ2 test was used for the proportions. A total of 283 ICU patients (62 conv-EEG, 221 POC-EEG) were included. The two populations were matched using demographic and clinical characteristics. We found that the ICU LOS was significantly shorter in the POC-EEG cohort compared to the conv-EEG cohort (3.9 [IQR 1.9-8.8] vs. 8.0 [IQR 3.0-16.0] days, p = 0.003). Moreover, modified Rankin Scale functional outcomes were also different between the two EEG cohorts (p = 0.047). This study reveals a significant association between early POC-EEG detection of nonconvulsive seizures and decreased ICU LOS. The POC-EEG differed from conv-EEG, demonstrating better functional outcomes compared with the latter in a matched analysis. These findings corroborate previous research advocating the benefit of early diagnosis of nonconvulsive seizure. The causal relationship between the type of EEG and metrics of interest, such as ICU LOS and functional/clinical outcomes, needs to be confirmed in future prospective randomized studies.

LMA-EEGNet: A Lightweight Multi-Attention Network for Neonatal Seizure Detection Using EEG signals

Neonatal epilepsy is an early postnatal brain disorder, and automatic seizure detection is crucial for timely diagnosis and treatment to reduce potential brain damage. This work proposes a novel Lightweight Multi-Attention Network, LMA-EEGNet, for diagnosing neonatal epileptic seizures from multi-channel EEG signals employing dilated depthwise separable convolution (DDS Conv) for feature extraction and using pointwise convolution followed by global average pooling for classification. The proposed approach substantially reduces the model size, number of parameters, and computational complexity, which are crucial for real-time detection and clinical diagnosis of neonatal epileptic seizures. LMA-EEGNet integrates temporal and spectral features through distinct temporal and spectral branches. The temporal branch uses DDS Conv to extract temporal features, enhanced by a channel attention mechanism. The spectral branch utilizes similar convolutions alongside a spatial attention mechanism to highlight key frequency components. Outputs from both branches are merged and processed through a pointwise convolution layer and a global average pooling layer for efficient neonatal seizure detection. Experimental results show that our model, with only 2471 parameters and a size of 23 KB, achieves an accuracy of 95.71% and an AUC of 0.9862, demonstrating its potential for practical deployment. This study provides an effective deep learning solution for the early detection of neonatal epileptic seizures, improving diagnostic accuracy and timeliness.

A comprehensive study of machine learning-based methods to predict epileptic seizures

People with epilepsy have many difficulties as a result of this complicated brain condition, which is typified by frequent convulsions. Symptoms of these seizures include bizarre behaviors, odd sensations, and in extreme cases, loss of awareness. These seizures appear as episodes of aberrant electrical impulses in the central nervous system. Successful epilepsy management depends on early seizure detection and identification, which allows for appropriate intervention to minimize risks and improve patient outcomes. Two major reasons have contributed to the extraordinary advancements in the area of epilepsy investigations in the last few years: the explosive development of machine learning techniques and the decreasing cost of non-invasive electroencephalography (EEG) apparatus. The availability of low-cost EEG equipment has made it easier to gather information on brain activity, which has opened up new avenues for monitoring and analyzing episodes of epileptic seizures away from conventional medical settings. The abundance of data and the advancement of machine learning methods have created new opportunities for the early identification and forecasting of seizures. Machine learning algorithms can predict seizures based on EEG data, providing patients with epilepsy with more control and informed decision-making. This paper offers a current review of current methods for treating epileptic seizures. The feature extraction techniques and classification algorithms receive particular focus. The most popular EEG datasets and their accessibility are listed. The approaches that are examined range from those that use more established machine learning techniques, such as naive Bayes models, Support Vector Machines (SVM), and Linear Discriminant Analysis (LDA), to those that take advantage of more recent deep learning techniques, like (Long-Short Term Memory, or LSTM), and deep Convolutional-Neural-Networks (CNN).

Medical Devices for the Management of Patients with Epilepsy

The increasing number of medical devices developed and marketed towards management of patients with epilepsy reflects the growing interest in translating technological advances and knowledge about epilepsy into better healthcare for this population. The objective of this narrative literature review is to analyze the available options of medical devices for detecting, treating, and recording epileptic seizures, and their potential clinical application. The included articles were selected from the PubMed database using the query “(Epilepsy[MeSH Terms]) AND (SUDEP)) AND (Medical Device)) AND (English[Language])”. The detection of epileptic seizures is essential for early intervention and to optimize the therapy for each patient. In outpatient settings, this detection is further challenging due to their unpredictability. Traditionally electroencephalography is the direct detection method used in a hospital environment. Indirect methods, such as electrocardiogram, photoplethysmography, oximeter, electrodermal activity, accelerometer, and electromyography, have shown potential for detecting seizures in the outpatient setting. Several medical devices have been developed based on the mentioned methods, with the aim of providing patients with solutions they can use in their daily lives. Behind-the-ear EEG, wristbands, armbands and bed sensors are some of the designs available. Equipped with different features, these devices can answer the need for early seizure detection and improve patients’ and caregivers’ quality of life. There are also devices available for the treatment of epileptic seizures. Through neuromodulation techniques such as vagus nerve stimulation, deep brain stimulation, and responsive neurostimulation, these devices are presented as solutions for patients with refractory epilepsy not eligible for ressective surgery. Patients with epilepsy have several apps available online for proper recording of seizures. These apps help doctors optimize therapy based on clinical evolution. The wide range of devices available creates the opportunity to personalize the approach to patient’s specific needs. Understanding each device’s characteristics can help clinicians improve management of patients with epilepsy.

Diagnosing epileptic seizures using combined features from independent components and prediction probability from EEG data.

Epileptic seizures are neurological events that pose significant risks of physical injuries characterized by sudden, abnormal bursts of electrical activity in the brain, often leading to loss of consciousness and uncontrolled movements. Early seizure detection is essential for timely treatments and better patient outcomes. To address this critical issue, there is a need for an advanced artificial intelligence approach for the early detection of epileptic seizure disorder. This study primarily focuses on designing a novel ensemble approach to perform early detection of epileptic seizure disease with high performance. A novel ensemble approach consisting of a fast, independent component analysis random forest (FIR) and prediction probability is proposed, which uses electroencephalography (EEG) data to investigate the efficacy of the proposed approach for early detection of epileptic seizures. The FIR model extracts independent components and class prediction probability features, creating a new feature set. The proposed model combined integrated component analysis (ICA) with predicting probability to enhance seizure recognition accuracy scores. Extensive experimental evaluations demonstrate that FIR assists machine learning models to obtain superior results compared to original features. The research gap is addressed using combined features to improve the performance of epileptic seizure detection compared to a single feature set. In particular, the ensemble model FIR with support vector machine (FIR + SVM) outperforms other methods, achieving an accuracy of 98.4% for epileptic seizure detection. The proposed FIR has the potential for early diagnosis of epileptic seizures and can significantly help the medical industry with enhanced detection and timely interventions.

Automatic epileptic seizure detection based on EEG using a moth-flame optimization of one-dimensional convolutional neural networks.

Frequent epileptic seizures can cause irreversible damage to the brains of patients. A potential therapeutic approach is to detect epileptic seizures early and provide artificial intervention to the patient. Currently, extracting electroencephalogram (EEG) features to detect epileptic seizures often requires tedious methods or the repeated adjustment of neural network hyperparameters, which can be time- consuming and demanding for researchers. This study proposes an automatic detection model for an EEG based on moth-flame optimization (MFO) optimized one-dimensional convolutional neural networks (1D-CNN). First, according to the characteristics and need for early epileptic seizure detection, a data augmentation method for dividing an EEG into small samples is proposed. Second, the hyperparameters are tuned based on MFO and trained for an EEG. Finally, the softmax classifier is used to output EEG classification from a small-sample and single channel. The proposed model is evaluated with the Bonn EEG dataset, which verifies the feasibility of EEG classification problems that involve up to five classes, including healthy, preictal, and ictal EEG from various brain regions and individuals. Compared with existing advanced optimization algorithms, such as particle swarm optimization, genetic algorithm, and grey wolf optimizer, the superiority of the proposed model is further verified. The proposed model can be implemented into an automatic epileptic seizure detection system to detect seizures in clinical applications.

Seizure Detection Via Time Series Classification Using Modified Metaheuristic Optimized Recurrent Networks

Epilepsy, colloquially termed seizure disorder, constitutes a neurological condi-tion characterized by unpredictable and sudden episodes of heightened electrical activity in the brain. The inherent risks associated with seizures, particularly in their nascent stages, stem from their challenging predictability, often leading to individuals endangering themselves through falls or lapses into unconsciousness in precarious settings. Additionally, the potential for neurological damage further compounds the complexity of managing this condition. This research endeavors to explore the viability of Recurrent Neural Networks (RNNs) for the early detection of seizures using electroencephalogram (EEG) recordings derived from real-world scenarios. Acknowledging the pivotal role of network topology and training parameters in achieving optimal performance, this study employs meta-heuristic algorithms to fine-tune performance through judicious hyperparameter selection. Notably, a modified iteration of a recently introduced algorithm is introduced herein. The application of RNNs to seizure detection yields promising outcomes, with constructed models achieving an accuracy surpassing 99%. The results obtained underscore the potential of employing metaheuristic-optimized RNNs in seizure detection as a means of substantially enhancing the quality of life for individuals afflicted by epilepsy. The demonstrated accuracy of these models suggests a robust and reliable methodology, paving the way for future advancements in the realm of neurotechnological interventions. This investigation contributes valuable insights into the intersection of machine learning, neuroscience, and healthcare, offering a foundation for further research and technological developments aimed at ameliorating the challenges posed by epilepsy.

BIOM-14. INCREASED RATIO OF GLUTAMATE TO TOTAL CREATINE IN GLIOBLASTOMA PATIENTS WITH THE POST-OPERATIVE EPILEPSY

Abstract Epilepsy in glioblastoma patients significantly reduces their quality of life; however, little is known about the association between predicting epilepsy and metabolites in tumors. In this study, we used 3.0-T magnetic resonance imaging (MRI) and 1H-magnetic resonance spectroscopy (MRS) to quantify metabolite concentrations in patients with varying epilepsy histories. MRS data were quantified with linear combination model (LC-Model) software. From January 2013 to April 2020, fifty-one patients with primary glioblastoma underwent pretreatment 3.0-T MRI/1H-MRS scanning. Twenty-one patients (41%) suffered from epilepsy during the postoperative course, although forty-seven patients (97%) took anti-epileptic drug after surgery. MRS data were compared between groups with and without the post-operative epilepsy. The higher ratios of glutamate (Glu) and glutamate + glutamine (Glx) to total creatine (tCr) in the tumor were significantly associated with epilepsy history (Glu/tCr P < 0.001, Glx/tCr P = 0.004, respectively). The receiver operating characteristic curve analysis showed that a Glu/tCr value of 1.81 was 70% sensitive and 90% specific for the prediction of epilepsy during the postoperative course (area under curve: 0.82). These results suggest that the Glu/tCr ratio in tumors can be a predictive maker for early detection of epileptic seizures after surgery in glioblastoma patients.

Prediction Of Seizures In Patients With Temporal Lobe Epilepsy, More Than 30 Minutes Before Onset.

Dear Madam, Around 52.5 million people worldwide suffer from the neurological illness known as epilepsy, which is frequently characterized by recurring seizures. (1, 2) As epilepsy is a chronic disorder that can lead to cognitive deficits, sadness, and anxiety in sufferers, it has a considerable negative impact on individuals and society. The most common type of seizures, affecting around 61% of persons with epilepsy, are focal seizures, which raise the risk of injury and early death (2). One of the most common forms of focal epilepsy, temporal lobe epilepsy (TLE), is characterized by recurring spontaneous seizures that begin in the temporal lobe. (s). (3) In the last two decades, many new antiseizure medications (ASMs)have been developed; however, about 40% of people with epilepsy may be drug-resistant, meaning ASMs are unable to control their seizures; therefore, despite the advent of new treatments, a high unmet need remains unaddressed. (2) Surgery is an alternate treatment, however, it cannot be used if larger areas of the brain are affected, in such cases, neuromodulation is a novel and successful therapeutic tool developed. (4) Conventionally, neural activity in epilepsy is categorized into interictal, and post-ictal states. With the help of continuous EEG, researchers have subdivided interictal state into a pre-ictal and pro-ictal state in temporal lobe epilepsy patients (which is the typical drug-resistant focal epilepsy targeted for interventions such as neuromodulation). The pre-ictal states indicate imminent seizure onset, while pro-ictal states reflect changes in neural activity that create a propensity for seizures over longer periods. The EEG-based pro-ictal state detection is critical to adaptive neuromodulation, with early detection of seizure allowing electrodes to be applied therapeutically to the brain's seizure onset zone and thalamus enabling treatment to be modified as per seizure risk. (5) A prospective cohort study was conducted on adults with drug-resistant temporal lobe epilepsy at UTHealth Houston, in which pro-ictal state – i.e., pathologic brain activities during periods of heightened seizure risk could be detected up to one-half hour prior to seizure onset. A classifier was developed that could reasonably accurately distinguish between physiological and pathologic brain states in temporal lobe epilepsy. Thalamocortical dynamics were used to identify pro-ictal states in each patient. As the limbic thalamus exerts a diverse influence on cortex activity, thalamocortical EEG was used for pro-ictal state detection as opposed to previous paradigms derived solely from cortical structures. ---Continue

Electroencephalography Quality and Application Times in a Pediatric Emergency Department Setting: A Feasibility Study

BackgroundStatus epilepticus is the most common neurological emergency presenting to pediatric emergency departments. Nonconvulsive status epilepticus can be extremely challenging to diagnose, however, requiring electroencephalographic (EEG) confirmation for definitive diagnosis. We aimed to determine the feasibility of achieving a good-quality pediatric EEG recording within 20 minutes of presentation to the emergency department. MethodsSingle-center prospective feasibility study in Cork University Hospital, Ireland, between July 2021 and June 2022. Two-channel continuous EEG was recorded from children (1) aged <16 years and (2) with Glasgow Coma Scale <11 or a reduction in baseline Glasgow Coma Scale in the case of a child with a neurodisability. ResultsTwenty patients were included. The median age at presentation was 65.8 months (interquartile range, 23.2 to 119.0); 50% had a background diagnosis of epilepsy. The most common reason for EEG monitoring was status epilepticus (85%) followed by suspected nonconvulsive status (10%) and reduced consciousness of unknown etiology (5%). The mean length of recording was 93.1 minutes (S.D. 47.4). The mean time to application was 41.3 minutes (S.D. 11.7). The mean percent of artifact in all recordings was 19.3% (S.D. 15.9). Thirteen (65%) EEGs had <25% artifact. Artifact was higher in cases in which active airway management was ongoing. ConclusionsEEG monitoring can be achieved in a pediatric emergency department setting within one hour of presentation. Overall, artifact percentage was low outside of periods of airway manipulation. Future studies are required to determine its use in early seizure detection and its support role in clinical decision-making in these patients.

Automated Seizure Detection in Neonatal EEG using Signal Processing Algorithms

The crucial challenge of identifying seizures in newborn electroencephalogram (EEG) data is the subject of this research. (ElectroEncephaloGram) EEG analysis has traditionally been labor-intensive and subjective due to the heavy reliance on human inspection by skilled electrophysiologists and neurologists. The authors suggest a novel online automatic detection approach to get over these restrictions. This approach intends to open the door for effective and reliable early seizure detection, delivering a potential development in neonatal care and neurological research. It does this by utilizing the capabilities of signal processing and brain-inspired modeling’s suggested machine-learning method makes effective use of real-time EEG signal processing methods. Its processing of incoming EEG data streams makes it appropriate for neonates to be closely tracked in critical care situations. Automation of the detection method minimizes the need for manual inspection by highly specialized experts, allowing for more rapid responses and possibly increasing seizure management's overall results. Using metal electrodes, the electroencephalogram (EEG) analyses the electrical activity of the brain. European Data Format (EDF) BROWSER and EEG STUDIO are used to analyses EEG signals. Root Mean Square (RMS) and signal samples are performed by EDF BROWSER, as well as mean, frequency, and frequency from filtered output using a band-pass filter. The standard deviation and average frequency are examined by EEG STUDIO. Spike detection, frequency domain analysis, and nonlinear techniques are used in seizure prediction. EOG, EKG, and EMG artifacts produced by the eyes, the heart, and the muscles, respectively may be present in EEG readings.

Incidence of Neonatal Seizures in China Based on Electroencephalogram Monitoring in Neonatal Neurocritical Care Units

Neonatal seizures pose a significant challenge in critical care, and continuous video electroencephalography (cEEG) monitoring holds promise for early detection of seizures. However, large-scale data on the incidence of neonatal seizures and monitoring systems in China are lacking. To determine the incidence of neonatal seizures in infants with high risk in China. A large, cross-sectional multicenter study was conducted from January 2017 to December 2018 in the neonatal intensive care units (NICUs) of 7 tertiary medical centers in China. Neonates with high risk were included, and cEEG monitoring was conducted. Data were collected between January 1, 2017, and January 31, 2020. The data were analyzed between January 2021 and January 2022. The incidence of neonatal seizures, categorized by etiology, and seizure burden. A total of 20 310 neonates with high risk were included (10 495 [51.7%] male; mean [SD] postmenstrual age, 37.7 [3.7] weeks), and seizures were observed in 3423 infants (16.9%). The highest proportion of seizures was attributed to acute neonatal encephalopathy (1448 [42.3%]). The incidence of seizures decreased with postmenstrual age and birth weight, with the highest occurrence observed in neonates with postmenstrual age of less than 28 weeks (237 of 879 [27.0%]) or birth weight of less than 1.0 kg (269 of 914 [29.4%]). Preterm infants had a higher proportion of moderate and severe seizure burdens compared with full-term infants (moderate severity: 248 of 1199 [20.7%] vs 454 of 2224 [20.4%]), but no significant differences were observed in etiology. Seizure burden was highest with genetic syndromes (49 of 188 [26.1%]), central nervous system malformations (33 of 127 [26.0%]), and inborn errors of metabolism (27 of 113 [23.9%]). During hospitalization, 7.8% of neonates with seizures died (267 neonates), with 81.3% of these cases having a moderate or severe seizure burden (217 neonates). Mortality was generally higher in preterm vs full-term infants (98 of 1199 [8.2%] vs 169 of 2224 [7.6%]) and increased with the severity of seizure burden (217 of 267 neonates with moderate or severe burden [81.3%]). This cross-sectional study of neonatal seizures underscores the substantial burden seizures pose to high-risk infants with brain injury in China, particularly those who are born prematurely or who have congenital conditions.

Effective Early Detection of Epileptic Seizures through EEG Signals Using Classification Algorithms Based on t-Distributed Stochastic Neighbor Embedding and K-Means.

Epilepsy is a neurological disorder in the activity of brain cells that leads to seizures. An electroencephalogram (EEG) can detect seizures as it contains physiological information of the neural activity of the brain. However, visual examination of EEG by experts is time consuming, and their diagnoses may even contradict each other. Thus, an automated computer-aided diagnosis for EEG diagnostics is necessary. Therefore, this paper proposes an effective approach for the early detection of epilepsy. The proposed approach involves the extraction of important features and classification. First, signal components are decomposed to extract the features via the discrete wavelet transform (DWT) method. Principal component analysis (PCA) and the t-distributed stochastic neighbor embedding (t-SNE) algorithm were applied to reduce the dimensions and focus on the most important features. Subsequently, K-means clustering + PCA and K-means clustering + t-SNE were used to divide the dataset into subgroups to reduce the dimensions and focus on the most important representative features of epilepsy. The features extracted from these steps were fed to extreme gradient boosting, K-nearest neighbors (K-NN), decision tree (DT), random forest (RF) and multilayer perceptron (MLP) classifiers. The experimental results demonstrated that the proposed approach provides superior results to those of existing studies. During the testing phase, the RF classifier with DWT and PCA achieved an accuracy of 97.96%, precision of 99.1%, recall of 94.41% and F1 score of 97.41%. Moreover, the RF classifier with DWT and t-SNE attained an accuracy of 98.09%, precision of 99.1%, recall of 93.9% and F1 score of 96.21%. In comparison, the MLP classifier with PCA + K-means reached an accuracy of 98.98%, precision of 99.16%, recall of 95.69% and F1 score of 97.4%.

Protein S100B and Amplitude-Integrated EEG as Early Predictive Methods for Brain Injury and Seizures in Preterm Neonates

Background: Neonatal brain injury (NBI) is a serious adverse outcome in premature neonates. The most common form of neonatal brain injury (NBI) are intraventricular hemorrhage (IVH) and periventricular leukomalacia (PVL). Improved survival of preterm infants leads to short- and long-term hazards of neurological, cognitive, respiratory, digestive, renal, cardiovascular, metabolic, immune, and psychosocial disturbances. We sought to determine the levels and prognostic value of serum S100B and the role of aEEG during the first three days of life in premature neonates (&lt; 37 weeks) that later developed NBI in the form of intraventricular hemorrhage (IVH) or neonatal seizures to rule out the sensitivity and specificity of them in early detection of brain lesions in preterm neonates.&#x0D; Aim of the Study: To evaluate the role of Protein S100B as an early predictor for neonatal brain injury in preterm neonates and the predictive and prognostic value of amplitude-integrated electroencephalography for neonatal brain injury susceptibility and severity.&#x0D; Subject and Methods: This study was carried out on 50 preterm neonate (less than 37 weeks GA) who were divided according to presence or absence of brain injury into case and control groups. They were admitted in Neonatal Intensive Care Unit (NICU) of Tanta University Hospital during the period from March 2021 to March 2022.serum S100B at (Day 1 and Day 3), serial trans-cranial sonar and aEEG were done for all patients.&#x0D; Results: Neonates with NBI, had significantly higher S100B concentration during the first three days of life, its level was higher in the third day than the first day, the cut-off value &gt;810.4 ng/ml serum S100B performed a sensitivity of 72.7%and a specificity of 71.4% to predict adverse neonatal outcome. 60% neonates had normal aEEG and 40% had abnormal aEEG.72.7% of neonates with NBI had abnormal aEEG interpretation. There was a significant relationship between the electrografic seizures on aEEG and occurrence of clinical seizures as 96.7% of neonates who had abnormal aEEG suffered of clinical convulsion.&#x0D; Conclusion: Protein S100B has a good predictive value regarding NBI in premature neonates and aEEG has a great role in monitoring neonatal brain function and early detection of neonatal seizures.

ECG-Based Semi-Supervised Anomaly Detection for Early Detection and Monitoring of Epileptic Seizures

Epilepsy is one of the most common brain diseases, characterized by frequent recurrent seizures or "ictal" states. A patient experiences uncontrollable muscular contractions, inducing loss of mobility and balance, which may result in injury or even death during these ictal states. Extensive investigation is vital to establish a systematic approach for predicting and informing patients about oncoming seizures ahead of time. Most methodologies developed are focused on the detection of abnormalities using mostly electroencephalogram (EEG) recordings. In this regard, research has indicated that certain pre-ictal alterations in the Autonomic Nervous System (ANS) can be detected in patient electrocardiogram (ECG) signals. The latter could potentially provide the basis for a robust seizure prediction approach. The recently proposed ECG-based seizure warning systems utilize machine learning models to classify a patient's condition. Such approaches require the incorporation of large, diverse, and thoroughly annotated ECG datasets, limiting their application potential. In this work, we investigate anomaly detection models in a patient-specific context with low supervision requirements. Specifically, we consider One-Class SVM (OCSVM), Minimum Covariance Determinant (MCD) Estimator, and Local Outlier Factor (LOF) models to quantify the novelty or abnormality of pre-ictal short-term (2-3 min) Heart Rate Variability (HRV) features of patients, trained on a reference interval considered to contain stable heart rate as the only form of supervision. Our models are evaluated against labels that were either hand-picked or automatically generated (weak labels) by a two-phase clustering procedure for samples of the "Post-Ictal Heart Rate Oscillations in Partial Epilepsy" (PIHROPE) dataset recorded by the Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, achieving detection in 9 out of 10 cases, with average AUCs of over 93% across all models and warning times ranging from 6 to 30 min prior to seizure. The proposed anomaly detection and monitoring approach can potentially pave the way for early detection and warning of seizure incidents based on body sensor inputs.

SJLA_2024_A5: Seizure Risk, Contributing Factors and Management Post Bariatric Surgery: Systematic

Introduction: The prevalence of bariatric surgery as a treatment for obesity has brought to light its implications for various health outcomes, including neurological disorders such as seizures and epilepsy. Obesity itself has been linked to neurological health, raising questions about the impact of its treatments, such as bariatric surgery, on neurological outcomes. Background: A study in this domain was conducted; their research investigated the risk of epilepsy following bariatric surgery in comparison to a nonsurgical cohort with an obesity diagnosis. Their results indicated a higher incidence of epilepsy in the postbariatric surgery group. These findings underscore the necessity of this systematic review. The relationship between bariatric surgery and neurological outcomes such as epilepsy remains inadequately explored, with potential long-term risks that warrant further investigation. Therefore, this review aims to synthesize available evidence, evaluate the effects of bariatric surgery on epilepsy risk, and identify potential mechanisms and risk factors. Methods: This review includes eligible case reports and cohort studies on people who developed postbariatric seizure between September 1962 and January 2024, selected from three primary databases: PubMed (NCBI), Embase (OvidSP), and Web of Science. The selection process was based on three phases: title, abstract, and full-text screening, and we included 19 articles. Results: Among the case reports, 14 out of 15 are associated with Roux-en-Y gastric bypass (RYGB) surgery. The causes of postbariatric seizures were diverse, with 56% attributed to hypoglycemia, 19% to hyperammonemia, and the remaining cases linked to hypocalcemia, vitamin deficiency, and postbariatric stroke. Cohort studies assessing postbariatric seizure risk showed significant findings. One study reported HR of 1.45, and another found HRs of 3 and 7.3 for epilepsy and seizures, respectively. These findings underline increased risk after bariatric surgeries. Management varies widely, including surgical and medical treatment. Surgical treatment by RYGB reversal was reported in several cases, with two approaches, including laparoscopic restoration using a Henley-Longmire interposition and distal pancreatectomy. Medical management involved antiepileptic drugs, antihypoglycemic agents, and specific therapy for hyperammonemia and hypocalcemia. Discussion: The results of this review suggest a link between bariatric surgeries and developing seizures. However, the majority of the studies concluded hypoglycemia was primarily associated with these outcomes. Few studies suggest hyperammonemia as a contributing factor; two studies implicate urea cycle disorders. Furthermore, surgical management showed the most effective way to control postbariatric hypoglycemia alongside other specific treatments for certain conditions, such as hyperammonemia. The main limitation of our review is the need for primary studies. Conclusion: Findings from this review highlight the need for continuous monitoring postoperatively to address any risks of seizures following bariatric surgery, as early detection of seizures can prevent adverse outcomes. Moreover, healthcare providers should focus on patients’ education regarding the risk of seizures, enabling them to recognize symptoms and seek medical care. Furthermore, these findings highlight the importance of continuous follow-up to optimize patient safety and improve long-term outcomes for individuals undergoing bariatric surgery.

Artificial intelligence system, based on mjn-SERAS algorithm, for the early detection of seizures in patients with refractory focal epilepsy: A cross-sectional pilot study.

Around one-third of epilepsy patients develop drug-resistant seizures; early detection of seizures could help improve safety, reduce patient anxiety, increase independence, and enable acute treatment. In recent years, the use of artificial intelligence techniques and machine learning algorithms in different diseases, including epilepsy, has increased significantly. The main objective of this study is to determine whether the mjn-SERAS artificial intelligence algorithm developed by MJN Neuroserveis, can detect seizures early using patient-specific data to create a personalized mathematical model based on EEG training, defined as the programmed recognition of oncoming seizures before they are primarily initiated, usually within a period of a few minutes, in patients diagnosed of epilepsy. Retrospective, cross-sectional, observational, multicenter study to determine the sensitivity and specificity of the artificial intelligence algorithm. We searched the database of the Epilepsy Units of three Spanish medical centers and selected 50 patients evaluated between January 2017 and February 2021, diagnosed with refractory focal epilepsy who underwent video-EEG monitoring recordings between 3 and 5days, a minimum of 3 seizures per patient, lasting more than 5s and the interval between each seizure was greater than 1h. Exclusion criteria included age <18years, intracranial EEG monitoring, and severe psychiatric, neurological, or systemic disorders. The algorithm identified pre-ictal and interictal patterns from EEG data using our learning algorithm and was compared to a senior epileptologist's evaluation as a gold standard. Individual mathematical models of each patient were trained using this feature dataset. A total of 1963h of 49 video-EEG recordings were reviewed, with an average of 39.26h per patient. The video-EEG monitoring recorded 309 seizures as subsequently analyzed by the epileptologists. The mjn-SERAS algorithm was trained on 119 seizures and split testing was performed on 188 seizures. The statistical analysis includes the data from each model and reports 10 false negatives (no detection of episodes recorded by video-EEG) and 22 false positives (alert detected without clinical correlation or abnormal EEG signal within 30min). Specifically, the automated mjn-SERAS AI algorithm achieved a sensitivity of 94.7% (95%; CI 94.67-94.73), and an F-Score representing specificity of 92.2% (95%; CI 92.17-92.23) compared to the reference performance represented by a mean (harmonic mean or average) and a positive predictive value of 91%, with a false positive rate of 0.55 per 24h in the patient-independent model. This patient-specific AI algorithm for early seizure detection shows promising results in terms of sensitivity and false positive rate. Although the algorithm requires high computational requirements on specialized servers cloud for training and computing, its computational load in real-time is low, allowing its implementation on embedded devices for online seizure detection.