Algorithm for automatic EEG classification according to the epilepsy type: Benign focal childhood epilepsy and structural focal epilepsy

Childhood epilepsy with centrotemporal spikes, had been previously considered as benign childhood epilepsy. According to the new classification proposed by Sheffer I. and colleagues the term “benign” has been changed to “self-limited”. Many studies reported that BECTS may cause transient or long lasting cognitive and behavioral disturbances. Rolandic epilepsy is the most frequent among the childhood focal epilepsy and may account for about 15–25% of all epileptic syndromes diagnosed between the ages of 5 to 15 years. The incidence range changes between 7.1–21 per 100000 in population younger than 15 years with male predominance. The age of onset in 90% of cases between 1 and 10 years with peak around 6–7 years. Seizures mainly occur during a night sleep, whereas the probability of awake seizures are less than 10%. The characteristic clinical features are: (1) focal motor seizure with unilateral orofacial tonic or clonic contractions; (2) speech arrest; (3) hypersalivation; (4) sensory symptoms represented by unilateral numbness or paresthesia of tongue, lips, gum and inner part of the check; (5) unilateral clonic jerk in leg and arm with postictal paresis; (6) generalized seizures. The EEG picture is distinctive in Rolandic epilepsy. The background activity is almost always preserved in awake state and during a sleep. The typical interictal EEG pattern is high voltage, diphasic spikes or sharp waves frequently with slow activity on central-midtemporal region. The centrotemporal spikes or rolandic spikes come from the lower rolandic region created a horizontal dipole with maximal electronegativity in the centrotemporal region and electropositivity in the frontal region usually seen unilateral or bilateral. In most cases children with RE have a good prognosis regarding both seizures and neurodevelopment. The remission of seizures usually occurs before the age of 18 years. The cognitive and behavior problem may happen in active period of disease which are reversable in most of patients.

Download Full-text

Atypical Evolutions of Idiopathic Focal Epilepsies in Childhood

Journal of Pediatric Epilepsy ◽

10.1055/s-0036-1584671 ◽

2016 ◽

Vol 05 (03) ◽

pp. 122-132

Author(s):

Roberto Caraballo ◽

Santiago Flesler ◽

Natalio Fejerman

Keyword(s):

Status Epilepticus ◽

Focal Epilepsy ◽

Childhood Epilepsy ◽

Significant Percentage ◽

Epileptic Encephalopathies ◽

Atypical Features ◽

Eeg Abnormalities ◽

Number Of Patients ◽

Panayiotopoulos Syndrome ◽

Eeg Features

Clinical and electroencephalographic (EEG) features of the three well-defined idiopathic focal epilepsies in childhood (IFEC)—benign childhood epilepsy with centrotemporal spikes (BCECTS), Panayiotopoulos syndrome, and the Gastaut type of idiopathic occipital epilepsy of childhood (IOEC-G)—have been clearly described and reported. It is also known that a significant percentage of children with IFEC present what were named atypical features in seizures and EEG abnormalities. We are studying here the small number of patients with IFEC who not only present atypical features, but also evolve into the spectrum of epileptic encephalopathies related with continuous spike-and-wave during sleep (CSWSS) or electrical status epilepticus during sleep (ESES), which includes atypical benign focal epilepsy of childhood, status of BCECTS, Landau-Kleffner syndrome, and CSWSS or ESES syndrome. We also emphasize that some patients with these encephalopathic course of IFEC present a mix or a sequence in time of the four mentioned subsyndromes.

Download Full-text

Epilepsy EEG Classification Based on Convolution Support Vector Machine

Journal of Medical Imaging and Health Informatics ◽

10.1166/jmihi.2021.3259 ◽

2021 ◽

Vol 11 (1) ◽

pp. 25-32

Author(s):

Qi Xin ◽

Shaohai Hu ◽

Shuaiqi Liu ◽

Xiaole Ma ◽

Hui Lv ◽

...

Keyword(s):

Support Vector Machine ◽

Focal Epilepsy ◽

Principal Component ◽

Support Vector ◽

Eeg Signals ◽

Eeg Classification ◽

Eeg Data ◽

Epilepsy Diagnosis ◽

Electroencephalogram Eeg

Clinical Electroencephalogram (EEG) data is of great significance to realize automatable detection, recognition and diagnosis to reduce the valuable diagnosis time. To make a classification of epilepsy, we constructed convolution support vector machine (CSVM) by integrating the advantages of convolutional neural networks (CNN) and support vector machine (SVM). To distinguish the focal and non-focal epilepsy EEG signals, we firstly reduced the dimensionality of EEG signals by using principal component analysis (PCA). After that, we classified the epilepsy EEG signals by the CSVM. The accuracy, sensitivity and specificity of our method reach up to 99.56%, 99.72% and 99.52% respectively, which are competitive than the widely acceptable algorithms. The proposed automatic end to end epilepsy EEG signals classification algorithm provides a better reference for clinical epilepsy diagnosis.

Download Full-text

Age, sex and ethnic differentials in the prevalence and control of epilepsy among Sri Lankan children: a population-based study

BMJ Paediatrics Open ◽

10.1136/bmjpo-2018-000430 ◽

2019 ◽

Vol 3 (1) ◽

pp. e000430 ◽

Cited By ~ 1

Author(s):

Jithangi Wanigasinghe ◽

Carukshi Arambepola ◽

Roshini Murugupillai ◽

Thashi Chang

Keyword(s):

Sri Lanka ◽

Focal Epilepsy ◽

Age Groups ◽

Population Based ◽

Cross Sectional Study ◽

Male Dominance ◽

Epilepsy Syndrome ◽

Childhood Epilepsy ◽

Age Group ◽

Cross Sectional

ObjectiveTo estimate the prevalence of childhood epilepsy in Sri Lanka by different age groups (0–5, 6–10 and 11–16 years), sex and ethnicity, and to describe the types and outcomes of epilepsy.Design and patientsA population-based, cross-sectional study was conducted in the district considered to be ethnically most balanced in Sri Lanka. A door-to-door survey was performed in the 0–5 year age group (60 geographically defined areas as clusters; 19 children per cluster), and a school-based survey in the 6–16 year age group (150 classes as clusters; 25 children per cluster). The screened children with epilepsy were reviewed individually for confirmation of the diagnosis of epilepsy, typing of the underlying epilepsy syndrome and assessment of control. The same group of children were re-evaluated 1 year later to reconfirm the syndromic diagnosis and to assess the stability of control of epilepsy.ResultsThe overall prevalence of childhood epilepsy was 5.7 per 10 000 children aged 0–16 years (95% CI: 38 to 87). It was higher with younger ages (73.4 per 10 000 children aged 0–5 years; 55.1 per 10 000 children aged 6–10 years and 50.4 per 10 000 children aged 11–16 years). A male dominance was noted in both age groups. In each age group, the prevalence was highest in children of Sinhalese ethnicity. Symptomatic focal epilepsy was the single most common group of epilepsy in both age groups. Majority of children remained well controlled on medications.ConclusionThe findings indicate a relatively high burden of epilepsy among children in Sri Lanka, however, these were comparable to the burden of disease reported from other countries in the region.

Download Full-text

Effect of Levetiracetam Monotherapy in Nonlesional Focal Childhood Epilepsy

Neuropediatrics ◽

10.1055/s-0037-1613680 ◽

2017 ◽

Vol 49 (02) ◽

pp. 135-141 ◽

Cited By ~ 1

Author(s):

Fumikazu Sano ◽

Tetsuo Ohyama ◽

Kanji Sugita ◽

Masao Aihara ◽

Hideaki Kanemura

Keyword(s):

Lower Risk ◽

Event Rate ◽

Focal Epilepsy ◽

Serum Levels ◽

Adverse Event Rate ◽

Efficacy Rate ◽

Childhood Epilepsy ◽

Superior Efficacy ◽

Medication Dose ◽

Valuable Treatment

AbstractThis article compares the efficacy and tolerability of carbamazepine (CBZ) and levetiracetam (LEV) when used as initial monotherapy in children with nonlesional focal epilepsy. Patients with nonlesional focal epilepsy were subdivided into two groups according to the initial monotherapy: a LEV group administered LEV at an initial dose of 5 mg/kg/day and a CBZ group. Seizure response, adverse events, medication dose, reasons for discontinuing medication, adherence, and random serum levels were recorded. The overall percentage of patients who failed initial treatment and reasons for each treatment failure were determined. Data were analyzed from 183 children who received CBZ monotherapy and 46 children who received LEV monotherapy for ≥12 months. Overall, 126 patients (68.9%) became seizure-free with CBZ, compared with 37 patients (80.4%) with LEV. Moreover, four patients in CBZ and four patients in LEV groups showed a >50% reduction in seizure frequency. The efficacy rate was significantly higher and the adverse event rate was significantly lower in the LEV group than in the CBZ group (p = 0.0129 and p = 0.0039, respectively). LEV may offer superior efficacy and a lower risk of adverse effects compared with CBZ. LEV as initial monotherapy may represent a valuable treatment option for children with nonlesional focal childhood epilepsy.

Download Full-text