Classification Of Seizure Types Research Articles

Neural Memory Networks for Seizure Type Classification.

Classification of seizure type is a key step in the clinical process for evaluating an individual who presents with seizures. It determines the course of clinical diagnosis and treatment, and its impact stretches beyond the clinical domain to epilepsy research and the development of novel therapies. Automated identification of seizure type may facilitate understanding of the disease, and seizure detection and prediction have been the focus of recent research that has sought to exploit the benefits of machine learning and deep learning architectures. Nevertheless, there is not yet a definitive solution for automating the classification of seizure type, a task that must currently be performed by an expert epileptologist. Inspired by recent advances in neural memory networks (NMNs), we introduce a novel approach for the classification of seizure type using electrophysiological data. We first explore the performance of traditional deep learning techniques which use convolutional and recurrent neural networks, and enhance these architectures by using external memory modules with trainable neural plasticity. We show that our model achieves a state-of-the-art weighted F1 score of 0.945 for seizure type classification on the TUH EEG Seizure Corpus with the IBM TUSZ preprocessed data. This work highlights the potential of neural memory networks to support the field of epilepsy research, along with biomedical research and signal analysis more broadly.

Epileptic Seizure Classification With Symmetric and Hybrid Bilinear Models.

Epilepsy affects nearly [Formula: see text] of the global population, of which two thirds can be treated by anti-epileptic drugs and a much lower percentage by surgery. Diagnostic procedures for epilepsy and monitoring are highly specialized and labour-intensive. The accuracy of the diagnosis is also complicated by overlapping medical symptoms, varying levels of experience and inter-observer variability among clinical professions. This paper proposes a novel hybrid bilinear deep learning network with an application in the clinical procedures of epilepsy classification diagnosis, where the use of surface electroencephalogram (sEEG) and audiovisual monitoring is standard practice. Hybrid bilinear models based on two types of feature extractors, namely Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), are trained using Short-Time Fourier Transform (STFT) of one-second sEEG. In the proposed hybrid models, CNNs extract spatio-temporal patterns, while RNNs focus on the characteristics of temporal dynamics in relatively longer intervals given the same input data. Second-order features, based on interactions between these spatio-temporal features are further explored by bilinear pooling and used for epilepsy classification. Our proposed methods obtain an F1-score of [Formula: see text] on the Temple University Hospital Seizure Corpus and [Formula: see text] on the EPILEPSIAE dataset, comparing favourably to existing benchmarks for sEEG-based seizure type classification. The open-source implementation of this study is available at https://github.com/NeuroSyd/Epileptic-Seizure-Classification.

EEG based multi-class seizure type classification using convolutional neural network and transfer learning

Recognition of epileptic seizure type is essential for the neurosurgeon to understand the cortical connectivity of the brain. Though automated early recognition of seizures from normal electroencephalogram (EEG) was existing, no attempts have been made towards the classification of variants of seizures. Therefore, this study attempts to classify seven variants of seizures with non-seizure EEG through the application of convolutional neural networks (CNN) and transfer learning by making use of the Temple University Hospital EEG corpus. The objective of our study is to perform a multi-class classification of epileptic seizure type, which includes simple partial, complex partial, focal non-specific, generalized non-specific, absence, tonic, and tonic–clonic, and non-seizures. The 19 channels EEG time series was converted into a spectrogram stack before feeding as input to CNN. The following two different modalities were proposed using CNN: (1) Transfer learning using pretrained network, (2) Extract image features using pretrained network and classify using the support vector machine classifier. The following ten pretrained networks were used to identify the optimal network for the proposed study: Alexnet, Vgg16, Vgg19, Squeezenet, Googlenet, Inceptionv3, Densenet201, Resnet18, Resnet50, and Resnet101. The highest classification accuracy of 82.85% (using Googlenet) and 88.30% (using Inceptionv3) was achieved using transfer learning and extract image features approach respectively. Comparison results showed that CNN based approach outperformed conventional feature and clustering based approaches. It can be concluded that the EEG based classification of seizure type using CNN model could be used in pre-surgical evaluation for treating patients with epilepsy.

Natural resonance frequency identification for sea waves and EEG epileptic patients using Pade approximation and neural network

Pade method is used in the spectral domain estimation to classify both the remote sensing and the biomedical signals. The first example presented from remote sensing is the sea wave classification while the second example depicted from biomedical engineering field is the Epilepsy seizure type classification. Feature extractions of both the Global Navigation Satellite Systems (GNSS) signal and the epilepsy seizure from a human Electroencephalograph (EEG) signal are based on the poles location of the signal.

Natural resonance frequency identification for remote sensing and biomedical engineering using Prony method and fuzzy logic

Prony method is applied to classify both the remote sensing and the biomedical signals. The first example presented from remote sensing is the sea wave classification while the second example depicted from biomedical engineering field is the Epilepsy seizure type classification. Feature extractions of both the Global navigation satellite systems (GNSS) signal and the epilepsy seizure from a human Electroencephalograph (EEG) signal are based on the poles location of the signal.

A convolutional neural network based framework for classification of seizure types.

Epileptic seizures are caused by a disturbance in the electrical activity of the brain and classified as many different types of epileptic seizures based on the characteristics of EEG and other parameters. Till now research has been conducted to classify EEG as seizure and non-seizures, but the classification of seizure types has not been explored. Thus, in this paper, we have proposed the 8-class classification problem in order to classify different seizure types using convolutional neural networks (CNN). This research study suggests a CNN based framework for classification of epileptic seizure types that include simple partial, complex partial, focal non-specific, generalized non-specific, absence, tonic, and tonic-clonic, and non-seizures. EEG time series was converted into spectrogram stacks and used as input for CNN. To the best of authors knowledge, ours is the very first study that classified the seizures types using the computational algorithm. The four CNN models, namely AlexNet, VGG16, VGG19, and basic CNN model was applied to study the performance of 8-class classification problem. The proposed study showed a classification accuracy of 84.06%, 79.71%, 76.81%, and 82.14% using AlexNet, VGG16, VGG19 and basic CNN models respectively. The experimental results suggest that the proposed framework could be helpful to the neurology community for recognition of seizures types.

Instruction manual for the ILAE 2017 operational classification of seizure types

Instruction manual for the ILAE 2017 operational classification of seizure types

国際抗てんかん連盟によるてんかん発作型の操作的分類：ILAE分類・用語委員会の公式声明

国際抗てんかん連盟によるてんかん発作型の操作的分類：ILAE分類・用語委員会の公式声明

Validation of a Hausa Childhood Epilepsy Screening and Seizure Classification Tool

Background: A validated epilepsy screening and seizure classification instrument, used by community health workers (CHWs), can enhance epilepsy diagnosis and classification to bring care to the 60-90% of children with epilepsy who remain untreated in low- and middle-income countries (LMICs). Methods: An epilepsy screening and seizure type classification tool was translated to the local language (Hausa) and administered by CHWs in clinics in Kano, Nigeria. Children were then evaluated by Kano epileptologists (gold standard). An epileptologist in the U.S., blinded to the clinical evaluations in Kano also interpreted the questionnaire results. Sensitivity, specificity, predictive values, and receiver operating characteristics (ROCs) of the questionnaire were determined. Findings: Of 765 consecutive children screened, the 82 who screened positive and 63 randomly selected from those who screened negative were evaluated by epileptologists. The screening tool demonstrated a sensitivity of 0.95 (95% CI= 0.90, 0.99), specificity of 0.88 (0.80, 0.96), and positive and negative predictive values of 0.90 (0.84, 0.97) and 0.94 (0.88, 0.99), respectively, for epilepsy diagnosis. The ROC for predicting a physician diagnosis of epilepsy was 0.92 (0.87, 0.96) and for predicting classification of focal onset and generalized seizures were 0.89 (0.84, 0.95) and 0.88 (0.81,0.94), respectively. Among children who screened positive for epilepsy, positive and negative predictive values were 1.00 (0.90, 1.00) and 0.73 (0.59, 0.86), respectively, for generalized seizures, and 0.67 (0.51, 0.83) and 1.00 (0.90, 1.00), respectively, for focal onset seizures. Interpretation: This childhood epilepsy screening and classification tool performed well when used by CHWs in northern Nigeria. Funding Statement: Fogarty International Center/NIH (1R21TW010899). Declaration of Interests: This study (ATA, ET, US, HA, ASS, HX, UW, ZI, DJI, HMY, MHA) was supported by Fogarty International Center/National Institutes of Health (FIC/NIH, 1R21TW010899). US was also supported by the VECD Fogarty Global Health Fellows Program (FIC/NIH, D43 TW009337). MHA also reports funding from the US Centers for Disease Control (CDC). DJI also reports funding from the US Agency for International Development (USAID). AAP reports funding from the American Academy of Neurology (Practice Research training scholarship). Ethics Approval Statement: This validation study protocol was approved by the Ethics Committees at AKTH and at the Kano State Ministry of Health, and by the Vanderbilt University Medical Center Internal Review Board.

Electroencephalographic profile of children attending a neurophysiology centre in Lagos, Nigeria

Background: Electroencephalography (EEG) is an important diagnostic tool for the evaluation of neurological disorders. There is a paucity of information about EEG recordings in Nigerian children. This study, therefore, aimed to describe the pattern and prevalence of EEG abnormalities among children referred for EEG assessment at a neurophysiology centre in Lagos, Nigeria. Methods: This was a prospective study carried out at a neurophysiology centre, Lagos, Nigeria. Consecutive 200 patients referred for EEG at a neurophysiology centre between 1st October, 2015-30th April 2016 were recruited into the study. Information was obtained on demographic, clinical presentation and EEG recordings. Results: The age of children ranges from 1-192 months with a median age of 42 (IQR 25-88) months. There were 128 males (64%) and 62 females (36%). Commonest clinical indications for making EEG request were seizure disorder (55%), cerebral palsy (24.5%) and febrile seizure (6.5%). EEG recording was abnormal in 161 (80.5%) children. There was a significantly high prevalence of abnormal epileptiform discharges in all cases with the exception of children with febrile seizures (p=0.002). The commonest abnormal epileptiform discharges were focal (70.2%). Others were generalized epileptiform activity (24.8%) and slow and diffuse waves suggestive of encephalopathies (5%). Ten (2%) children from the study population were diagnosed with absence seizure, Lennox-Gaustaut, and infantile spasm Conclusion: Secondary causes of seizure are an important contributor to seizure disorders in Nigerian children. The EEG is useful in the detection and classification of seizure types.

Chapter 11 - Video-EEG

Indications for video-EEG monitoring (VEM) include differential diagnosis of paroxysmal events including epileptic seizures, organic nonepileptic seizures, and psychogenic nonepileptic seizures; classification of seizure types and electroclinical syndromes; quantification of seizures and of interictal and ictal epileptiform discharges; and presurgical evaluation in medically refractory epilepsy patients. Standardized questionnaires and examinations should be used on admission to the epilepsy monitoring unit (EMU). Patients should be provided with comprehensive information concerning purpose and procedures during VEM and need to sign informed consents. Staff working in the EMU needs to be properly trained in the management of seizures and periictal testing according to written protocols as well as in cardiopulmonary resuscitation. Minimum staffing ratios of dedicated healthcare professionals to patients have been recommended. Antiepileptic drug tapering/withdrawal needs to be individualized for each patient. EEG recordings have to be performed according to established guidelines. Cardiorespiratory monitoring including continuous ECG monitoring and continuous measurement of oxygen saturation is strongly recommended. Patient safety is of utmost importance during VEM. Indicators for reporting quality and safety have been developed. Standardized ictal testing protocols and standardized computer-based organized reporting should further improve standards of VEM and training of EMU staff, and facilitate data exchange and collaborations between EMUs.

The effect of atmospheric pressure on frequency and patient's subjective perception of epileptic seizure severity

Epilepsy is a paroxysmal cerebral dysrhythmia defined by three important characteristics: the suddenness of the process, brain origin and the disorder of the rhythm of the electrical brain activity (1). Seizures (crises, ictus events, attacks) are common to all epilepsies. They include a variety of clinical manifestations or sudden episodes of disturbance of motor, sensibility, behaviour, perception, awareness or other psychic functions, autonomic and other regulations. The classification of seizures and epilepsy syndrome takes the central place in modern epileptology, although the classification process itself is as old as the study of epilepsy. The international classification, based mainly on the clinical form of seizure and the electroencephalographic findings, was adopted in 1981 and has been further modified. In 1989, ILAE (International League Against Epilepsy) adopted the International classification of epilepsies, epilepsy syndromes and disorders, which highlights the symptoms and signs that show the localization of initial cerebral dysfunction and its spread (2). In 2017, the ILAE released a new classification of seizure types, based upon the existing classification formulated in 1981. and its extension from 2010. The differences include the following: "partial seizures" become "focal"; consciousness is used as a classifier of focal seizures; terms such as „dyscognitive”, „simple partial”, „complex partial”, „psychic” and „secondary generalized” are eliminated; new focal seizures include automatisms, behavioural changes, hyperkinetic, autonomic, cognitive and emotional; atonic, clonic, epileptic spasms, myoclonic, and tonic seizures can be either focal or generalized phenomena; „focal seizure evolving to a bilateral tonic-clonic seizure” replace the term „secondary generalized seizure”; new generalized types of attack are absence with eyelid myoclonia, myoclonic absence, myoclonic-atonic and myoclonic-tonic-clonic and seizures of unknown onset (3). Biometeorology is an interdisciplinary science that studies the interaction system between living organisms and the environment. The biometerological warning, bioforecast, contain a description of the meteorological situation and announces the arrival of those weather conditions which may adversely affect human health in the next days. The goal of bioprognosis is the organization of preventive measures for the protection of the vulnerable groups health (4). Variations in biometeorological factors have an influence on the overall balance of the human body, as well as to changes in the condition of patients, primarily those with chronic diseases. Le Blanc and Mills (1932) stated that one of the oldest and best known disease whose association with climatic factors is mentioned, is epilepsy (5). Fluctuation in some climatic factors, e.g. air pressure, might influence one or more of the above mentioned factors, and thus lead to seizure. The purpose of this study was to determine the effect of atmospheric pressure on the specific seizure types frequency as well as on patient's subjective perception of the seizure "severity".

Nonlesional late-onset epilepsy: Semiology, EEG, cerebrospinal fluid, and seizure outcome characteristics

IntroductionIncidence and prevalence of epilepsy increase with advancing age. Although the majority of late-onset epilepsies are of lesional origin, a considerable proportion of patients present with unknown etiology. The aim of this study was to evaluate the semiological, electroencephalographic (EEG), and cerebrospinal fluid (CSF) characteristics as well as the 12-month seizure outcome in a cohort of patients with nonlesional late-onset epilepsy (≥55 years). MethodA total of 54 patients with newly diagnosed nonlesional late-onset epilepsy (NLLOE) were retrospectively evaluated for seizure type using the most recent International League Against Epilepsy (ILAE) classification of seizure types, EEG characteristics, and CSF profile and followed-up for at least 12 months after epilepsy onset. Results were compared with a gender-matched control group of 58 patients with nonlesional early-onset epilepsy (NLEOE). ResultsThe predominant seizure types in NLLOE were focal to bilateral tonic–clonic seizures (30%) as well as focal onset impaired awareness motor seizures (IAMS) (22%) and focal onset impaired awareness nonmotor seizures (IANMS) (22%). The predominant seizure types in NLEOE were focal to bilateral tonic–clonic seizures (43%) as well as focal onset aware nonmotor seizures (ANMS) (31%) and IAMS (31%). Focal onset impaired awareness nonmotor seizures were found to be more characteristic in patients with NLLOE (p = 0.019; α < 0.05; NLLOE: 22.2% vs. NLEOE: 8.6%). Electroencephalography revealed no significant differences between groups. Of interest, three patients with NLLOE (8%) presented with oligoclonal bands (OCB) in CSF albeit absence of antineuronal antibodies. Seizure-free rate was 70%. Adverse effects from medication leading to antiepileptic drug (AED) change were reported in 12 patients (22%), valproate was the best tolerated AED in patients with NLLOE [adverse effects in 9%, compared with 12% (gabapentin) and 26% (levetiracetam)]. ConclusionsUsing the most recent classification system, different patterns of semiological characteristics were identified: NLLOE more frequently present with IANMS, whereas patients with NLEOE rather have ANMS. Oligoclonal bands were only detected in patients with NLLOE, indicating that careful exclusion of autoimmune encephalitis in this patient group is warranted. Our findings may help to more accurately identify and characterize patients with NLLOE to improve targeted diagnostics and adequate treatment in this challenging group of patients.

The 2017 ILAE classification of seizure types and the epilepsies: what do people with epilepsy and their caregivers need to know?

The International League against Epilepsy (ILAE) published in the April 2017 edition of Epilepsia three companion articles on the classification of seizures and the epilepsies. These represent a long-awaited update on the original 1981 and 1989 publications and provide a modern descriptive template. The new classification presents three levels of terminology, involving seizure types, epilepsy types, and syndromes. In this fourth paper, we present an interpretation of these new concepts for people with epilepsy and those who care for them, as well as for young medical doctors not specialized in epilepsy and nurses. Our goal in writing this paper is to ensure that everyone is speaking and understanding the same language, which is fundamental to the optimal management of people with epilepsy.

The Revised Definition and Classification of Epilepsy for Neurodiagnostic Technologists

ABSTRACTThe definition of who has epilepsy, classification of seizure types, and types of epilepsy have all recently been revised. The classical definition of epilepsy as a person having two or more unprovoked seizures more than 24 hours apart has been expanded also to include those with one seizure and a high likelihood (more than 60%) of having another. In the new definition, epilepsy is considered to be resolved when a person is seizure-free for 10 years, the terminal 5 being off seizure medicines, or when an age-dependent syndrome has been outgrown. The new seizure type classification revises the 1981 system but maintains the primary distinction of focal- versus generalized-onset seizures. Seizures also can be of unknown onset. Focal seizures may demonstrate retention or impairment of awareness, resulting in focal-aware or focal-impaired awareness seizures. Several new focal and generalized seizure types are introduced. Classification of the epilepsies is now by grouping of seizure types, etiologies, comorbidities, and epilepsy syndromes. The goal of the new terminology is greater clarity of communication and more accurate grouping of seizure types for research. Neurodiagnostic technologists can be of great help in observing clinical and electrographic features that will define the type of seizure.

The New Classification of Seizures by the International League Against Epilepsy 2017

This review presents the newly developed International League Against Epilepsy (ILAE) 2017 classification of seizure types. The fundamental distinction is between seizures that begin focally in one hemisphere of the brain, generalized onset seizures that apparently originate in both hemispheres, and seizures of unknown onset. Focal seizures optionally can be subclassified according to whether awareness (a surrogate marker for consciousness) is intact or impaired. The next level of classification for focal seizures is motor (with subgroups automatisms, atonic, clonic, epileptic spasms, hyperkinetic, myoclonic, tonic), non-motor (with subgroups autonomic, behavior arrest, cognitive, emotional, sensory), and focal to bilateral tonic-clonic. Generalized seizures are categorized as motor (tonic-clonic, clonic, tonic, myoclonic, myoclonic-tonic-clonic, myoclonic-atonic, atonic, epileptic spasms) and non-motor/absence (typical, atypical, myoclonic, eyelid myoclonia). The classification allows new types of focal seizures and a few new generalized seizures, and clarifies terms used to name seizures.

Instruction manual for the ILAE 2017 operational classification of seizure types.

This companion paper to the introduction of the International League Against Epilepsy (ILAE) 2017 classification of seizure types provides guidance on how to employ the classification. Illustration of the classification is enacted by tables, a glossary of relevant terms, mapping of old to new terms, suggested abbreviations, and examples. Basic and extended versions of the classification are available, depending on the desired degree of detail. Key signs and symptoms of seizures (semiology) are used as a basis for categories of seizures that are focal or generalized from onset or with unknown onset. Any focal seizure can further be optionally characterized by whether awareness is retained or impaired. Impaired awareness during any segment of the seizure renders it a focal impaired awareness seizure. Focal seizures are further optionally characterized by motor onset signs and symptoms: atonic, automatisms, clonic, epileptic spasms, or hyperkinetic, myoclonic, or tonic activity. Nonmotor-onset seizures can manifest as autonomic, behavior arrest, cognitive, emotional, or sensory dysfunction. The earliest prominent manifestation defines the seizure type, which might then progress to other signs and symptoms. Focal seizures can become bilateral tonic-clonic. Generalized seizures engage bilateral networks from onset. Generalized motor seizure characteristics comprise atonic, clonic, epileptic spasms, myoclonic, myoclonic-atonic, myoclonic-tonic-clonic, tonic, or tonic-clonic. Nonmotor (absence) seizures are typical or atypical, or seizures that present prominent myoclonic activity or eyelid myoclonia. Seizures of unknown onset may have features that can still be classified as motor, nonmotor, tonic-clonic, epileptic spasms, or behavior arrest. This "users' manual" for the ILAE 2017 seizure classification will assist the adoption of the new system.

Operational classification of seizure types by the International League Against Epilepsy: Position Paper of the ILAE Commission for Classification and Terminology.

The International League Against Epilepsy (ILAE) presents a revised operational classification of seizure types. The purpose of such a revision is to recognize that some seizure types can have either a focal or generalized onset, to allow classification when the onset is unobserved, to include some missing seizure types, and to adopt more transparent names. Because current knowledge is insufficient to form a scientifically based classification, the 2017 Classification is operational (practical) and based on the 1981 Classification, extended in 2010. Changes include the following: (1) "partial" becomes "focal"; (2) awareness is used as a classifier of focal seizures; (3) the terms dyscognitive, simple partial, complex partial, psychic, and secondarily generalized are eliminated; (4) new focal seizure types include automatisms, behavior arrest, hyperkinetic, autonomic, cognitive, and emotional; (5) atonic, clonic, epileptic spasms, myoclonic, and tonic seizures can be of either focal or generalized onset; (6) focal to bilateral tonic-clonic seizure replaces secondarily generalized seizure; (7) new generalized seizure types are absence with eyelid myoclonia, myoclonic absence, myoclonic-atonic, myoclonic-tonic-clonic; and (8) seizures of unknown onset may have features that can still be classified. The new classification does not represent a fundamental change, but allows greater flexibility and transparency in naming seizure types.

Revised Operational Classification of Seizure Types

The International League Against Epilepsy (ILAE) issued a revised operational classification of seizure types. The ILAE retained much of the old structure, but new features include new types of focal motor seizures: automatism, tonic, atonic, myoclonic, hyperkinetic, and spasm motor seizures. Focal nonmotor seizures can be autonomic, behavior arrest, cognitive, emotional, and sensory. A few new generalized types are included, as well as seizures of unknown onset. Partial is renamed focal. Simple partial is renamed focal aware and complex partial, focal impaired awareness. The new classification should enhance clarity and communication.

ILAE CLASSIFICATION OF THE EPILEPSIES: THE 2017 REVISION AND UPDATE

The International League Against Epilepsy (ILAE) has developed and approved the Operational classification of seizure types 2017 and the Classification of the Epilepsies 2017 . According to the ILAE 2017 Operational classification of seizure types, seizures can be initially divided into focal or generalized; and the seizure onset is of crucial importance in this case. Focal seizures are optionally subdivided into focal aware seizures and focal impaired awareness seizures. Several new types of generalized-onset seizures were implemented. Specific motor and non-motor classifiers may be added. Unknown onset seizures are placed separately. Three levels of diagnostics are specified in the 2017 ILAE Classification of the Epilepsies: seizure type (defined in accordance with the 2017 ILAE Operational classification of seizure types), epilepsy type (focal, generalized, combined generalized and focal, unknown), and epilepsy syndrome. An etiologic diagnosis should be considered at each step along the diagnostic pathway. A patient ’s epilepsy may be classified into more than one etiological category. The term “benign” is replaced by the terms “self-limited” and “pharmacoresponsive” to be used where appropriate. The term “developmental and epileptic encephalopathy” can be applied in whole or in part where appropriate. The 2017 Operational classification of seizure types and the 2017 ILAE Classification of the Epilepsies can be helpful both in routine clinical practice and scientific research that will assist in improving epilepsy care and quality of life of patients with epilepsies.