EEG Seizure Research Articles

Implementation of a non-linear SVM classification for seizure EEG signal analysis on FPGA

An automatic detection method for disease diagnosis plays an important role in the medical field. A computer-aided diagnosis (CAD) system for seizure detection using electroencephalogram (EEG) data is significant in protecting patients' lives. This paper presents machine learning algorithm-based seizure detection approach implemented in both software and hardware using EEG recordings. The input EEG signal is subjected to pre-processing, decomposition using wavelet transform and classification by a support vector machine (SVM) classifier. The maximum and variance statistical values of the decomposed EEG sub-bands are computed to generate a feature vector, while classification accuracy and computational complexity are taken into account. Gaussian Radial Basis Function (RBF) kernel based Non-Linear SVM (NLSVM) is used in this proposed approach to solve both two-class and three-class classification problems. In three-class classification, the one-against-one (OAO) technique is used. The competency of the proposed technique is evaluated using two distinct datasets provided by Bonn University, Germany & Neurology and Sleep Centre, New Delhi. The adequate parameters of the NLSVM classifier model are determined by using MATLAB software in the training phase and the best performing NLSVM model is selected for implementation on the Field-Programmable Gate Array (FPGA). The proposed seizure detection method is implemented with Verilog code and examined on a Xilinx Zynq-7000 XC7Z020 FPGA board to evaluate its performance. The experimental results show that the proposed method with a minimal number of features achieved a good performance metric for seizure detection and is comparable with that of the state-of-the-art methods.

Deep Multiview Module Adaption Transfer Network for Subject-Specific EEG Recognition.

Transfer learning is one of the popular methods to solve the problem of insufficient data in subject-specific electroencephalogram (EEG) recognition tasks. However, most existing approaches ignore the difference between subjects and transfer the same feature representations from source domain to different target domains, resulting in poor transfer performance. To address this issue, we propose a novel subject-specific EEG recognition method named deep multiview module adaption transfer (DMV-MAT) network. First, we design a universal deep multiview (DMV) network to generate different types of discriminative features from multiple perspectives, which improves the generalization performance by extensive feature sets. Second, module adaption transfer (MAT) is designed to evaluate each module by the feature distributions of source and target samples, which can generate an optimal weight sharing strategy for each target subject and promote the model to learn domain-invariant and domain-specific features simultaneously. We conduct extensive experiments in two EEG recognition tasks, i.e., motor imagery (MI) and seizure prediction, on four datasets. Experimental results demonstrate that the proposed method achieves promising performance compared with the state-of-the-art methods, indicating a feasible solution for subject-specific EEG recognition tasks. Implementation codes are available at https://github.com/YangLibuaa/DMV-MAT.

Comparative subcutaneous and submuscular implantation of an electroencephalography device for long term electroencephalographic monitoring in dogs.

Implantable electroencephalography (EEG) recording devices have been used for ultra-long-term epilepsy monitoring both in clinical and home settings in people. Objective and accurate seizure detection and recording at home could be of great benefit in diagnosis, management and research in canine idiopathic epilepsy (IE). Continuous EEG monitoring would allow accurate detection of seizure patterns, seizure cycles, and seizure frequency. An EEG acquisition system usable in an "out of clinic" setting could improve owner and veterinary compliance for EEG diagnostics and seizure management. Whether a subcutaneous ultra-long term EEG monitoring device designed for humans could be implanted in dogs. Cadaver study with 8 medium to large breed dogs. Comparatively using a subcutaneous and submuscular approach to implant the UNEEG SubQ-Implant in each dog. Positioning was controlled via CT post implantation and cranial measurements were taken. In four of the eight dogs a submuscular implantation without any complications was possible. Complications were close contact to the optic nerve in the first approaches, before the implantation angle was changed and in the smallest dog contact of the implant with the orbital fat body. Cranial measurements of less than 95 mm length proved to be too small for reliable implantation via this approach. The subcutaneous approach showed severe limitations and the implant was prone to dislocation. The UNEEQ SubQ-Implant can be implanted in dogs, via submuscular approach. CT imaging and cranial measurements should be taken prior to implantation.

Methods for Extended Sleep EEG Recording in Autism: A Sleep Deprivation–Based, Natural Sleep–Recording Protocol and Feasibility Study

Background: Electroencephalography (EEG) has been used to research individuals with autism spectrum disorder (ASD). Such individuals are known to have a higher prevalence of EEG abnormalities and seizures. Extended EEG recordings, particularly those including sleep, are known to significantly improve the yield of these abnormalities. However, obtaining extended sleep recordings without sedation in autistic individuals is challenging. This study explores practical and feasible protocols for recording EEG in awake and sleep states in children and adolescents with ASD. Methods: Extended EEG recordings using detailed sleep deprivation and recording protocol were done. Results: 132 children and adolescents with ASD and sleep deprivation completed extended EEG recording in natural sleep. Only nine autistic children required melatonin along with sleep deprivation. Conclusions: Extended EEG recording can be achieved by following a rigorous sleep deprivation protocol that is customized, along with providing an autism-friendly recording room environment. This is important in identifying and increasing the yield of epileptiform discharges. We also propose technical guidelines and considerations for EEG recording in children and adolescents with ASD.

A sustainable artificial-intelligence-augmented digital care pathway for epilepsy: Automating seizure tracking based on electroencephalogram data using artificial intelligence.

Scalp electroencephalograms (EEGs) are critical for neurological evaluations, particularly in epilepsy, yet they demand specialized expertise that is often lacking in many regions. Artificial intelligence (AI) offers potential solutions to this gap. While existing AI models address certain aspects of EEG analysis, a fully automated system for routine EEG interpretation is required for effective epilepsy management and healthcare professionals' decision-making. This study aims to develop an AI-augmented model for automating EEG seizure tracking, thereby supporting a sustainable digital care pathway for epilepsy (DCPE). The goal is to improve patient monitoring, facilitate collaborative decision-making, ensure timely medication adherence, and promote patient compliance. The study proposes an AI-augmented framework using machine learning, focusing on quantitative analysis of EEG data to automate DCPE. A focus group discussion was conducted with healthcare professionals to find the problem of the current digital care pathway and assess the feasibility, usability, and sustainability of the AI-augmented system in the digital care pathway. The study found that a combination of random forest with principal component analysis and support vector machines with KBest feature selection achieved high accuracy rates of 96.52% and 95.28%, respectively. Additionally, the convolutional neural networks model outperformed other deep learning algorithms with an accuracy of 97.65%. The focus group discussion revealed that automating the diagnostic process in digital care pathway could reduce the time needed to diagnose epilepsy. However, the sustainability of the AI-integrated framework depends on factors such as technological infrastructure, skilled personnel, training programs, patient digital literacy, financial resources, and regulatory compliance. The proposed AI-augmented system could enhance epilepsy management by optimizing seizure tracking accuracy, improving monitoring and timely interventions, facilitating collaborative decision-making, and promoting patient-centered care, thereby making the digital care pathway more sustainable.

Effect of Input Channel Reduction on EEG Seizure Detection

Effect of Input Channel Reduction on EEG Seizure Detection

Usage and significance of EEG in febrile seizures: Can EEG be a diagnostic method for detecting epilepsy that may develop in the future?

Usage and significance of EEG in febrile seizures: Can EEG be a diagnostic method for detecting epilepsy that may develop in the future?

Generalized onset seizures with focal evolution (GOFE) – a largely unknown ictal variation in genetic generalized epilepsies

PurposeAt onset of generalized seizures, focal electroclinical features are commonly seen, while generalized onset seizures with focal evolution (GOFE) are largely unknown bearing the risk of misclassification. MethodsIn two German epilepsy-centers, patients with GOFE documented by video-EEG monitoring (VEM) between 2017 and 2022 were identified retrospectively. In addition to analysis of ictal electroclinical features, detailed epilepsy and family history, response to antiseizure medication (ASM), and findings from neuroimaging were considered. ResultsWe identified five patients with GOFE, three females, age 14 to 22 years. All patients developed genetic generalized epilepsy in childhood or adolescence, each presenting with two or three generalized seizure types. In each of the five patients, one GOFE was recorded by VEM. At onset, EEG seizure patterns were characterized by generalized spike-wave discharges at 2.5 to 3.5/sec for 9 to 16 sec followed by focal evolution of the discharges. Interictally, all patients presented with generalized spike-wave discharges without focal abnormalities. Semiology at onset was behavioral arrest in two patients and generalized increase in tone in one, while two onsets were clinically inapparent. Semiological signs during focal evolution were variable, comprising head and body version, figure 4 sign, unilateral arm clonic activity, and staring with oral automatisms. In one case, focality involved both hemispheres successively. ConclusionProminent focal semiological features in GOFE carry a high risk of misclassification as focal seizures and epilepsy and thus wrong choice of ASM. This calls for low-threshold VEM if any doubts of focal genesis of seizures exist.

Efficacy of Lacosamide and Rufinamide as Adjuncts to Midazolam-Ketamine Treatment Against Cholinergic-Induced Status Epilepticus in Rats.

Benzodiazepine pharmacoresistance develops when treatment of status epilepticus (SE) is delayed. This response may result from gamma-aminobutyric acid A receptors (GABAAR) internalization that follows prolonged SE; this receptor trafficking results in fewer GABAAR in the synapse to restore inhibition. Increase in synaptic N-methyl-D-aspartate receptors (NMDAR) also occurs in rodent models of SE. Lacosamide, a third-generation antiseizure medication (ASM), acts on the slow inactivation of voltage-gated sodium channels. Another ASM, rufinamide, similarly acts on sodium channels by extending the duration of time spent in the inactivation stage. Combination therapy of the benzodiazepine midazolam, NMDAR antagonist ketamine, and ASMs lacosamide (or rufinamide) was investigated for efficacy against soman (GD)-induced SE and neuropathology. Adult male rats implanted with telemetry transmitters for monitoring electroencephalographic (EEG) activity were exposed to a seizure-inducing dose of GD and treated with an admix of atropine sulfate and HI-6 1 minute later and with midazolam monotherapy or combination therapy 40 minutes after EEG seizure onset. Rats were monitored continuously for seizure activity for two weeks, after which brains were processed for assessment of neurodegeneration, neuronal loss, and neuroinflammatory responses. Simultaneous administration of midazolam, ketamine, and lacosamide (or rufinamide) was more protective against GD-induced SE compared with midazolam monotherapy. In general, lacosamide triple therapy had more positive outcomes on measures of epileptogenesis, EEG power integral, and the number of brain regions protected from neuropathology compared with rats treated with rufinamide triple therapy. Overall, both drugs were well tolerated in these combination models. SIGNIFICANCE STATEMENT: We currently report on improved efficacy of antiseizure medications lacosamide and rufinamide, each administered in combination with ketamine (NMDAR antagonist) and midazolam (benzodiazepine), in combatting soman (GD)-induced seizure, epileptogenesis, and brain pathology over that provided by midazolam monotherapy, or dual therapy of midazolam and lacosamide (or rufinamide) in rats. Administration of lacosamide as adjunct to midazolam and ketamine was particularly effective against GD-induced toxicity. However, protection was incomplete, suggesting the need for further study.

Targeted suppression of mTORC2 reduces seizures across models of epilepsy

Epilepsy is a neurological disorder that poses a major threat to public health. Hyperactivation of mTOR complex 1 (mTORC1) is believed to lead to abnormal network rhythmicity associated with epilepsy, and its inhibition is proposed to provide some therapeutic benefit. However, mTOR complex 2 (mTORC2) is also activated in the epileptic brain, and little is known about its role in seizures. Here we discover that genetic deletion of mTORC2 from forebrain neurons is protective against kainic acid-induced behavioral and EEG seizures. Furthermore, inhibition of mTORC2 with a specific antisense oligonucleotide robustly suppresses seizures in several pharmacological and genetic mouse models of epilepsy. Finally, we identify a target of mTORC2, Nav1.2, which has been implicated in epilepsy and neuronal excitability. Our findings, which are generalizable to several models of human seizures, raise the possibility that inhibition of mTORC2 may serve as a broader therapeutic strategy against epilepsy.

Hemispheric epilepsy surgery for hemimegalencephaly: The UCLA experience.

Hemimegalencephaly (HME) is a rare congenital brain malformation presenting predominantly with drug-resistant epilepsy. Hemispheric disconnective surgery is the mainstay of treatment; however, little is known about how postoperative outcomes compare across techniques. Thus we present the largest single-center cohort of patients with HME who underwent epilepsy surgery and characterize outcomes. This observational study included patients with HME at University of California Los Angeles (UCLA) from 1984 to 2021. Patients were stratified by surgical intervention: anatomic hemispherectomy (AH), functional hemispherectomy (FH), or less-than-hemispheric resection (LTH). Seizure freedom, functional outcomes, and operative complications were compared across surgical approaches. Regression analysis identified clinical and intraoperative variables that predict seizure outcomes. Of 56 patients, 43 (77%) underwent FH, 8 (14%) underwent AH, 2 (4%) underwent LTH, 1 (2%) underwent unknown hemispherectomy type, and 2 (4%) were managed non-operatively. At median last follow-up of 55 months (interquartile range [IQR] 20-92 months), 24 patients (49%) were seizure-free, 17 (30%) required cerebrospinal fluid (CSF) shunting for hydrocephalus, 9 of 43 (21%) had severe developmental delay, 8 of 38 (21%) were non-verbal, and 15 of 38 (39%) were non-ambulatory. There was one (2%) intraoperative mortality due to exsanguination earlier in this cohort. Of 12 patients (29%) requiring revision surgery, 6 (50%) were seizure-free postoperatively. AH, compared to FH, was not associated with statistically significant improved seizure freedom (hazard ratio [HR] = .48, p = .328), although initial AH trended toward greater odds of seizure freedom (75% vs 46%, p = .272). Younger age at seizure onset (HR = .29, p = .029), lack of epilepsia partialis continua (EPC) (HR = .30, p = .022), and no contralateral seizures on electroencephalography (EEG) (HR = .33, p = .039) independently predicted longer duration of seizure freedom. This study helps inform physicians and parents of children who are undergoing surgery for HME by demonstrating that earlier age at seizure onset, absence of EPC, and no contralateral EEG seizures were associated with longer postoperative seizure freedom. At our center, initial AH for HME may provide greater odds of seizure freedom with complications and functional outcomes comparable to those of FH.

EEG based automated seizure detection – A survey of medical professionals

Diagnosing and managing seizures presents substantial challenges for clinicians caring for patients with epilepsy. Although machine learning (ML) has been proposed for automated seizure detection using EEG data, there is little evidence of these technologies being broadly adopted in clinical practice. Moreover, there is a noticeable lack of surveys investigating this topic from the perspective of medical practitioners, which limits the understanding of the obstacles for the development of effective automated seizure detection. Besides the issue of generalisability and replicability seen in a small amount of studies, obstacles to the adoption of automated seizure detection remain largely unknown. To understand the obstacles preventing the application of seizure detection tools in clinical practice, we conducted a survey targeting medical professionals involved in the management of epilepsy. Our study aimed to gather insights on various factors such as the clinical utility, professional sentiment, benchmark requirements, and perceived barriers associated with the use of automated seizure detection tools.Our key findings are: I) The minimum acceptable sensitivity reported by most of our respondents (80%) seems achievable based on studies reported from most currently available ML-based EEG seizure detection algorithms, but replication studies often fail to meet this minimum. II) Respondents are receptive to the adoption of ML seizure detection tools and willing to spend time in training. III) The top three barriers for usage of such tools in clinical practice are related to availability, lack of training, and the blackbox nature of ML algorithms. Based on our findings, we developed a guide that can serve as a basis for developing ML-based seizure detection tools that meet the requirements of medical professionals, and foster the integration of these tools into clinical practice.

Interictal magnetoencephalography abnormalities to guide intracranial electrode implantation and predict surgical outcome.

Intracranial EEG is the gold standard technique for epileptogenic zone localization but requires a preconceived hypothesis of the location of the epileptogenic tissue. This placement is guided by qualitative interpretations of seizure semiology, MRI, EEG and other imaging modalities, such as magnetoencephalography. Quantitative abnormality mapping using magnetoencephalography has recently been shown to have potential clinical value. We hypothesized that if quantifiable magnetoencephalography abnormalities were sampled by intracranial EEG, then patients' post-resection seizure outcome may be better. Thirty-two individuals with refractory neocortical epilepsy underwent magnetoencephalography and subsequent intracranial EEG recordings as part of presurgical evaluation. Eyes-closed resting-state interictal magnetoencephalography band power abnormality maps were derived from 70 healthy controls as a normative baseline. Magnetoencephalography abnormality maps were compared to intracranial EEG electrode implantation, with the spatial overlap of intracranial EEG electrode placement and cerebral magnetoencephalography abnormalities recorded. Finally, we assessed if the implantation of electrodes in abnormal tissue and subsequent resection of the strongest abnormalities determined by magnetoencephalography and intracranial EEG corresponded to surgical success. We used the area under the receiver operating characteristic curve as a measure of effect size. Intracranial electrodes were implanted in brain tissue with the most abnormal magnetoencephalography findings-in individuals that were seizure-free postoperatively (T = 3.9, P = 0.001) but not in those who did not become seizure-free. The overlap between magnetoencephalography abnormalities and electrode placement distinguished surgical outcome groups moderately well (area under the receiver operating characteristic curve = 0.68). In isolation, the resection of the strongest abnormalities as defined by magnetoencephalography and intracranial EEG separated surgical outcome groups well, area under the receiver operating characteristic curve = 0.71 and area under the receiver operating characteristic curve = 0.74, respectively. A model incorporating all three features separated surgical outcome groups best (area under the receiver operating characteristic curve = 0.80). Intracranial EEG is a key tool to delineate the epileptogenic zone and help render individuals seizure-free postoperatively. We showed that data-driven abnormality maps derived from resting-state magnetoencephalography recordings demonstrate clinical value and may help guide electrode placement in individuals with neocortical epilepsy. Additionally, our predictive model of postoperative seizure freedom, which leverages both magnetoencephalography and intracranial EEG recordings, could aid patient counselling of expected outcome.

Electroencephalogram pattern predicting neurological outcomes of children with seizures secondary to abusive head trauma

BackgroundThe clinical presentations of abusive head trauma can abruptly worsen, so the occurrence of seizures and changes of EEG can be variable according to patients’ conditions. Since the changes of EEG background waves reflect the cortical function of children, we aimed to find out whether the timing of EEG background, epileptiform discharges and seizure patterns were associated with the outcomes of patients with AHT. Material and methodsUsing seizure type and acute stage electroencephalographic (EEG) characteristics to assess adverse neurological outcomes in children with seizures secondary to abusive head trauma (AHT). Children who were hospitalized with AHT at a tertiary referral hospital from October 2000 to April 2010 were evaluated retrospectively. A total of 50 children below 6 years of age admitted due to AHT were included. KOSCHI outcome scale was used to evaluate the primary outcome and neurological impairment was used as secondary outcome after 6 months discharge. ResultsChildren with apnea, cardiac arrest, reverse blood flow and skull fracture in clinic had a higher mortality rate even in the no-seizure group (3/5 [60%] vs. 3/45 [6.7%], odds ratio [OR] = 11; 95% CI = 2.3–52; p = 0.025). Seizure occurrence reduced mostly at the second day after admission in seizure groups; but children with persistent seizures for 1 week showed poor neurological outcomes. The occurrence of initial seizure was frequency associated with younger age; focal seizure, diffuse cortical dysfunction in acute-stage EEG, and low Glasgow Coma Scale (GCS) score were significantly related to poor outcomes after 6 months. Diffuse cortical dysfunction was also associated with motor, speech, and cognitive dysfunction. ConclusionsDiffuse cortical dysfunction in acute-stage EEG combined with low GCS score and focal seizure may related to poor outcomes and neurological dysfunctions in children with AHT.

Epilepsiechirurgie beim therapierefraktären Status epilepticus des Kindesalters

AbstractThe mainstay of status epilepticus (SE) treatment is pharmacotherapy with anti-seizure medications (ASM). In refractory status epilepticus (RSE), when additional ASM are not effective, high-dose suppressive therapy with either benzodiazepines, thiopental, phenobarbitone, or propofol is used to suppress clinical and EEG seizure activity. However, in selected eligible cases of RSE or in super-refractory cases, epilepsy surgery may be the treatment of choice to terminate SE. Here, we review epilepsy surgery including deep brain stimulation (DBS) for treatment for RSE with emphasis on special aspects of presurgical evaluation, patient selection, and outcome. We focus on surgical treatment options for patients in the acute phase of RSE, who have received high-dose suppressive therapy prior to surgery in the majority of the cases.

Prognostic value of early electrographic biomarkers of epileptogenesis in high-risk ischaemic stroke patients.

Post-stroke epilepsy (PSE) is frequent. Better prediction of PSE would enable individualized management and improve trial design for epilepsy prevention. The aim was to assess the complementary value of continuous electroencephalography (EEG) data during the acute phase compared with clinical risk factors currently used to predict PSE. A prospective cohort of 81 patients with ischaemic stroke who received early continuous EEG monitoring was studied to assess the association of early EEG seizures, other highly epileptogenic rhythmic and periodic patterns, and regional attenuation without delta (RAWOD, an EEG pattern of stroke severity) with PSE. Clinical risk factors were investigated using the SeLECT (stroke severity; large-artery atherosclerosis; early clinical seizures; cortical involvement; territory of middle cerebral artery) scores. Twelve (15%) patients developed PSE. The presence of any of the investigated patterns was associated with a risk of epilepsy of 46%, with a sensitivity and specificity of 83% and 78%. The association remained significant after adjusting for the SeLECT score (odds ratio 18.8, interquartile range 3.8-72.7). It was found that highly epileptogenic rhythmic and periodic patterns and RAWOD were associated with the development of PSE and complemented clinical risk factors. These findings indicate that continuous EEG provides useful information to determine patients at higher risk of developing PSE and could help individualize care.

Intracranial EEG Structure-Function Coupling and Seizure Outcomes After Epilepsy Surgery.

Surgery is an effective treatment for drug-resistant epilepsy, which modifies the brain's structure and networks to regulate seizure activity. Our objective was to examine the relationship between brain structure and function to determine the extent to which this relationship affects the success of the surgery in controlling seizures. We hypothesized that a stronger association between brain structure and function would lead to improved seizure control after surgery. We constructed functional and structural brain networks in patients with drug-resistant focal epilepsy by using presurgery functional data from intracranial EEG (iEEG) recordings, presurgery and postsurgery structural data from T1-weighted MRI, and presurgery diffusion-weighted MRI. We quantified the relationship (coupling) between structural and functional connectivity by using the Spearman rank correlation and analyzed this structure-function coupling at 2 spatial scales: (1) global iEEG network level and (2) individual iEEG electrode contacts using virtual surgeries. We retrospectively predicted postoperative seizure freedom by incorporating the structure-function connectivity coupling metrics and routine clinical variables into a cross-validated predictive model. We conducted a retrospective analysis on data from 39 patients who met our inclusion criteria. Brain areas implanted with iEEG electrodes had stronger structure-function coupling in seizure-free patients compared with those with seizure recurrence (p = 0.002, d = 0.76, area under the receiver operating characteristic curve [AUC] = 0.78 [95% CI 0.62-0.93]). Virtual surgeries on brain areas that resulted in stronger structure-function coupling of the remaining network were associated with seizure-free outcomes (p = 0.007, d = 0.96, AUC = 0.73 [95% CI 0.58-0.89]). The combination of global and local structure-function coupling measures accurately predicted seizure outcomes with a cross-validated AUC of 0.81 (95% CI 0.67-0.94). These measures were complementary to other clinical variables and, when included for prediction, resulted in a cross-validated AUC of 0.91 (95% CI 0.82-1.0), accuracy of 92%, sensitivity of 93%, and specificity of 91%. Our study showed that the strength of structure-function connectivity coupling may play a crucial role in determining the success of epilepsy surgery. By quantitatively incorporating structure-function coupling measures and standard-of-care clinical variables into presurgical evaluations, we may be able to better localize epileptogenic tissue and select patients for epilepsy surgery. This is a Class IV retrospective case series showing that structure-function mapping may help determine the outcome from surgical resection for treatment-resistant focal epilepsy.

Parameterized aperiodic and periodic components of single-channel EEG enables reliable seizure detection.

Although it is clinically important, a reliable and economical solution to automatic seizure detection for patients at home is yet to be developed. Traditional algorithms rely on multi-channel EEG signals and features of canonical EEG power description. This study is aimed to propose an effective single-channel EEG seizure detection method centered on novel EEG power parameterization and channel selection algorithms.We employed the publicly available multi-channel CHB-MIT Scalp EEG database to gauge the effectiveness of our approach. We first adapted a power spectra parameterization algorithm to characterize the aperiodic and periodic components of the ictal and inter-ictal EEGs. We selected four features based on their statistical significance and interpretability, and developed a ranking approach to channel selection for each patient. We then tested the effectiveness of our approaches to channel and feature selection for automatic seizure detection using support vector machine (SVM) as the classifier. The performance of our algorithm was evaluated using five-fold cross-validation and compared to those methods of comparable complexity (using one or two channels of EEG), in terms of accuracy, specificity, sensitivity, precision and F1 score.Some channels of EEG signals show strikingly different distributions of PSD features between the ictal and inter-ictal states. Four features including the offset and exponent parameters for the aperiodic component and the first and second highest total power (TPW1 and TPW2) form the basis of channel selection and the input of SVM classifier. The selected channel is found to be patient-specific. Our approach has achieved a mean sensitivity of 95.6%, specificity of 99.2%, accuracy of 98.6%, precision of 95.5%, and F1 score of 95.5%. Compared with algorithms in previous studies that used one or two channels of EEG signals, ours outperforms in specificity and accuracy with comparable sensitivity.EEG power spectra parameterization to feature extraction and feature ranking-based channel selection are found to enable efficient and effective automatic seizure detection based on single-channel EEG signal.

A machine learning approach to seizure detection in a rat model of post-traumatic epilepsy

Epilepsy is a common neurologic condition frequently investigated using rodent models, with seizures identified by electroencephalography (EEG). Given technological advances, large datasets of EEG are widespread and amenable to machine learning approaches for identification of seizures. While such approaches have been explored for human EEGs, machine learning approaches to identifying seizures in rodent EEG are limited. We utilized a predesigned deep convolutional neural network (DCNN), GoogLeNet, to classify images for seizure identification. Training images were generated through multiplexing spectral content (scalograms), kurtosis, and entropy for two-second EEG segments. Over 2200 h of EEG data were scored for the presence of seizures, with 95.6% of seizures identified by the DCNN and a false positive rate of 34.2% (1.52/h), as compared to visual scoring. Multiplexed images were superior to scalograms alone (scalogram-kurtosis-entropy 0.956 ± 0.010, scalogram 0.890 ± 0.028, t(7) = 3.54, p < 0.01) and a DCNN trained specifically for the individual animal was superior to using DCNNs across animals (intra-animal 0.960 ± 0.0094, inter-animal 0.811 ± 0.015, t(30) = 5.54, p < 0.01). For this dataset the DCNN approach is superior to a previously described algorithm utilizing longer local line lengths, calculated from wavelet-decomposition of EEG, to identify seizures. We demonstrate the novel use of a predesigned DCNN constructed to classify images, utilizing multiplexed images of EEG spectral content, kurtosis, and entropy, to rapidly and objectively identifies seizures in a large dataset of rat EEG with high sensitivity.

Effect of melatonin on sleep quality and EEG features in childhood epilepsy: a possible non-conventional treatment.

Sleep and epilepsy are characterized by a bidirectional relationship. Indeed, epilepsy predisposes to the development of sleep disorders, while sleep deprivation may exacerbate epilepsy. In addition, antiseizure medication can disrupt normal sleep architecture. Therefore, adequate sleep hygiene could lead to improvement in seizure control. The present study aimed to evaluate the effect of melatonin on seizure frequency, EEG tracing, and sleep in children with focal idiopathic epilepsy. This observation study evaluated the effect of 4 mg oral melatonin in ameliorating sleep-wake cycle, seizure frequency, and EEG features in children with focal idiopathic epilepsy of infancy. Twenty children were enrolled from September 2020 to August 2021. The study consisted of serial controls at enrollment (t0), at 3 months (t1), and at 6 months (t2) including neurological examination, questionnaire about sleep disturbances (CSHQ), and EEG. A significant improvement in sleep quality and daytime sleepiness was observed after melatonin supplementation. Furthermore, we observed a noteworthy improvement in EEG tracing at t2 that exhibited a significant correlation with improvements in CSHQ scores. The studies conducted so far to evaluate the effect of melatonin in persons with epilepsy do not lead to definitive conclusions. Despite the small population sample and the study design, we report sleep and EEG improvement after melatonin administration in our cohort. Larger studies are needed to further study the neuroprotective and anticonvulsant properties of melatonin.