Localizational concepts in epilepsy: past, present and future

The clinical seizure pattern, particularly the initial phenomena, plus the EEG, when satisfactory recording of the seizure onset can be achieved, determine the primary localization of epileptic phenomena. The EEG has also demonstrated, by the presence of interictal epileptiform spike discharges, the presence of a second-order localization of epileptic phenomena, namely, the location and extent of cortex adjacent to the site of origin of the neuronal seizure discharge that is recruited into action in a clinical epileptic seizure. Experience with cortical resection in the treatment of focal epilepsy has demonstrated the importance of a third-order localization of epileptic phenomena, namely, how much of the potentially epileptogenic cortex must be excised in order to produce a satisfactory reduction of the seizure tendency.     

Dynamic Imaging of Coherent Sources Reveals Different Network Connectivity Underlying the Generation and Perpetuation of Epileptic Seizures

The concept of focal epilepsies includes a seizure origin in brain regions with hyper synchronous activity (epileptogenic zone and seizure onset zone) and a complex epileptic network of different brain areas involved in the generation, propagation, and modulation of seizures. The purpose of this work was to study functional and effective connectivity between regions involved in networks of epileptic seizures. The beginning and middle part of focal seizures from ictal surface EEG data were analyzed using dynamic imaging of coherent sources (DICS), an inverse solution in the frequency domain which describes neuronal networks and coherences of oscillatory brain activities. The information flow (effective connectivity) between coherent sources was investigated using the renormalized partial directed coherence (RPDC) method. In 8/11 patients, the first and second source of epileptic activity as found by DICS were concordant with the operative resection site; these patients became seizure free after epilepsy surgery. In the remaining 3 patients, the results of DICS / RPDC calculations and the resection site were discordant; these patients had a poorer post-operative outcome. The first sources as found by DICS were located predominantly in cortical structures; subsequent sources included some subcortical structures: thalamus, Nucl. Subthalamicus and cerebellum. DICS seems to be a powerful tool to define the seizure onset zone and the epileptic networks involved. Seizure generation seems to be related to the propagation of epileptic activity from the primary source in the seizure onset zone, and maintenance of seizures is attributed to the perpetuation of epileptic activity between nodes in the epileptic network. Despite of these promising results, this proof of principle study needs further confirmation prior to the use of the described methods in the clinical praxis.     

Incidence of seizures and EEG abnormalities among offspring of epileptic patients

Summary The marriage rate of epileptic patients was 62% in males und 78% in females. Compared with the rates in the general population, the male patients had a 15% lower rate, but there was no difference in females. There were 263 patients with at least one offspring selected for the study. There were 243 sons and 272 daughters (506 total, 1.9 per patient). Distribution by types of seizure was awakening grand mal, absence or myoclonic petit mal in 24%, grand mal with no aura in 21%, grand mal during sleep in 23%, diffuse grand mal in 7%, grand mal with aura in 13%, psychomotor seizure in 9%, and focal seizure in 3%. The probands were composed of 79% idiopathic and 21% symptomatic in pathogenetic classification. An epileptic EEG abnormality was demonstrated in 22% of male and 44% of female probands.The incidence of seizures among offspring was 2.4% (4.2% age-corrected) in a narrow sense (epilepsy) and 9.1% in a broad sense including febrile convulsions. The latter morbidity was 11.0% for the idiopathic and 3.2% for the symptomatic group; 11.0% for female and 6.9% for male probands; 10.2% for sons and 8.1% for daughters. The figure was higher for the probands with the age range at onset of seizure of 0–4 years (20.6%) and 20–29 years (12.6%) than for those with other age ranges; higher for those with awakening grand mal, absence, myoclonic petit mal, or grand mal with no aura than for those with other types of seizure; and higher for those with family history of epilepsy than those without it.Possible correlation of types of seizure between probands and offspring was demonstrated. Thirty-seven percent of offspring exhibited epileptic EEG abnormalities, and the ratio of epileptic EEG abnormalities to clinical manifestation is about 4:1.Possible existence of familial aggregation of EEG abnormalities and of two kinds of families with large or small epileptic predisposition was indicated.The importance of the role of hereditary and environmental factors in epileptic pathogenesis is proved, and the results of an investigation of congenital malformation among offspring of epileptic mothers are presented. These results were considered to be useful for genetic counseling of epileptic patients.     

Application of a dissimilarity index of EEG and its sub-bands on prediction of induced epileptic seizures from rat's EEG signals

ObjectiveEpileptic seizures are defined as manifest of excessive and hyper-synchronous activity of neurons in the cerebral cortex that cause frequent malfunction of the human central nervous system. Therefore, finding precursors and predictors of epileptic seizure is of utmost clinical relevance to reduce the epileptic seizure induced nervous system malfunction consequences. Researchers for this purpose may even guide us to a deep understanding of the seizure generating mechanisms. The goal of this paper is to predict epileptic seizures in epileptic rats.MethodsSeizures were induced in rats using pentylenetetrazole (PTZ) model. EEG signals in interictal, preictal, ictal and postictal periods were then recorded and analyzed to predict epileptic seizures. Epileptic seizures were predicted by calculating an index in consecutive windows of EEG signal and comparing the index with a threshold. In this work, a newly proposed dissimilarity index called Bhattacharyya Based Dissimilarity Index (BBDI), dynamical similarity index and fuzzy similarity index were investigated.ResultsBBDI, dynamical similarity index and fuzzy similarity index were examined on case and control groups and compared to each other. The results show that BBDI outperforms dynamical and fuzzy similarity indices. In order to improve the results, EEG sub-bands were also analyzed. The best result achieved when the proposed dissimilarity index was applied on Delta sub-band that predicts epileptic seizures in all rats with a mean of 299.5 s.ConclusionThe dissimilarity of neural network activity between reference window and present window of EEG signal has a significant increase prior to an epileptic seizure and the proposed dissimilarity index (BBDI) can reveal this variation to predict epileptic seizures. In addition, analyzing EEG sub-bands results in more accurate information about constituent neuronal activities underlying the EEG since certain changes in EEG signal may be amplified when each sub-band is analyzed separately.SignificanceThis paper presents application of a dissimilarity index (BBDI) on EEG signals and its sub-bands to predict PTZ-induced epileptic seizures in rats. Based on the results of this work, BBDI will predict epileptic seizures more accurately and more reliably compared to current indices that increases epileptic patient comfort and improves patient outcomes.     

Brain Source Imaging in Preclinical Rat Models of Focal Epilepsy using High-Resolution EEG Recordings

Electroencephalogram (EEG) has been traditionally used to determine which brain regions are the most likely candidates for resection in patients with focal epilepsy. This methodology relies on the assumption that seizures originate from the same regions of the brain from which interictal epileptiform discharges (IEDs) emerge. Preclinical models are very useful to find correlates between IED locations and the actual regions underlying seizure initiation in focal epilepsy. Rats have been commonly used in preclinical studies of epilepsy¹; hence, there exist a large variety of models for focal epilepsy in this particular species. However, it is challenging to record multichannel EEG and to perform brain source imaging in such a small animal. To overcome this issue, we combine a patented-technology to obtain 32-channel EEG recordings from rodents² and an MRI probabilistic atlas for brain anatomical structures in Wistar rats to perform brain source imaging. In this video, we introduce the procedures to acquire multichannel EEG from Wistar rats with focal cortical dysplasia, and describe the steps both to define the volume conductor model from the MRI atlas and to uniquely determine the IEDs. Finally, we validate the whole methodology by obtaining brain source images of IEDs and compare them with those obtained at different time frames during the seizure onset.     

Epileptic Seizure Detection Based on New Hybrid Models with Electroencephalogram Signals

Objectives: Epileptic seizures are one of the most common diseases in society and difficult to detect. In this study, a new method was proposed to automatically detect and classify epileptic seizures from EEG (Electroencephalography) signals.Methods: In the proposed method, EEG signals classification five-classes including the cases of eyes open, eyes closed, healthy, from the tumor region, an epileptic seizure, has been carried out by using the support vector machine (SVM) and the normalization methods comprising the z-score, minimum-maximum, and MAD normalizations. To classify the EEG signals, the support vector machine classifiers having different kernel functions, including Linear, Cubic, and Medium Gaussian, have been used. In order to evaluate the performance of the proposed hybrid models, the confusion matrix, ROC curves, and classification accuracy have been used. The used SVM models are Linear SVM, Cubic SVM, and Medium Gaussian SVM.Results: Without the normalizations, the obtained classification accuracies are 76.90%, 82.40%, and 81.70% using Linear SVM, Cubic SVM, and Medium Gaussian SVM, respectively. After applying the z-score normalization to the multi-class EEG signals dataset, the obtained classification accuracies are 77.10%, 82.30%, and 81.70% using Linear SVM, Cubic SVM, and Medium Gaussian SVM, respectively. With the minimum-maximum normalization, the obtained classification accuracies are 77.20%, 82.40%, and 81.50% using Linear SVM, Cubic SVM, and Medium Gaussian SVM, respectively. Moreover, finally, after applying the MAD normalization to the multi-class EEG signals dataset, the obtained classification accuracies are 76.70%, 82.50%, and 81.40% using Linear SVM, Cubic SVM, and Medium Gaussian SVM, respectively.Conclusion: The obtained results have shown that the best hybrid model is the combination of cubic SVM and MAD normalization in the classification of EEG signals classification five-classes.     

基于脑电信号的癫痫发作预测特征及识别

解码癫痫发作前脑电信号的神经元集群异常痫样放电活动,对癫痫发作进行有效预测并实施病前干预,可显著减少疾病病损,是癫痫防治的研究热点之一。基于脑电信号的癫痫发作预测研究关键在于发作间期和前期的异常状态识别,研究上述两状态间的神经动力学特征差异对明确癫痫发病机制、选取高分辨特征,进而有效识别该渐进性疾病所处的发作阶段具有重要价值。目前,研究者已对当前主流特征提取及模式识别方法进行了充分的调研梳理,但忽视了神经动态特征变化对于癫痫发作预测的重要意义。基于此,本文归纳总结了5类典型的发作预测特征分析方法及其优缺点,重点剖析了发作间期至前期神经生理特征的动态变化及其动力学特性,类比分析了当前该领域主流的机器学习和深度学习特征识别方法,以期为进一步建立精准、高效的癫痫发作预测技术提供新思路。     

Electroencephalographic study of the effect of neurotoxin DSP-4 in iron model of chronic focal epilepsy.

The effect of the noradrenergic neurotoxin N-(2-chloroethyl)-N-ethyl-2-bromobenzylamine (DSP-4) on electroencephalographic activity (EEG) was studied in the model of chronic focal epilepsy induced by intracortical injection of FeCl3 in the rat. EEG activity was recorded from the epileptogenic focus (ipsilateral and contralateral) in chronic experiments before and after DSP-4 treatment. In some experiments EEG activity was also simultaneously recorded from the cortical epileptogenic focus and locus coeruleus before and after DSP-4 treatment to study the effect of iron-induced seizure activity and of DSP-4 on the locus coeruleus electrical activity. The results showed that DSP-4 aggravated the iron-induced epileptiform activity as well as the locus-coeruleus electrical activity. The data also showed that, induction of epilepsy by FeCl3 is accompanied by enhancement of the locus coeruleus electrical activity. Our study demonstrates that DSP-4 intensifies and modifies the epileptic activity in the iron-induced chronic epilepsy model and that the effects of toxin persist for a longer duration.     

Epileptic seizures may begin hours in advance of clinical onset: a report of five patients

Mechanisms underlying seizure generation are traditionally thought to act over seconds to minutes before clinical seizure onset. We analyzed continuous 3- to 14-day intracranial EEG recordings from five patients with mesial temporal lobe epilepsy obtained during evaluation for epilepsy surgery. We found localized quantitative EEG changes identifying prolonged bursts of complex epileptiform discharges that became more prevalent 7 hr before seizures and highly localized subclinical seizure-like activity that became more frequent 2 hr prior to seizure onset. Accumulated energy increased in the 50 min before seizure onset, compared to baseline. These observations, from a small number of patients, suggest that epileptic seizures may begin as a cascade of electrophysiological events that evolve over hours and that quantitative measures of preseizure electrical activity could possibly be used to predict seizures far in advance of clinical onset.     

Multiway analysis of epilepsy tensors

MOTIVATION: The success or failure of an epilepsy surgery depends greatly on the localization of epileptic focus (origin of a seizure). We address the problem of identification of a seizure origin through an analysis of ictal electroencephalogram (EEG), which is proven to be an effective standard in epileptic focus localization. SUMMARY: With a goal of developing an automated and robust way of visual analysis of large amounts of EEG data, we propose a novel approach based on multiway models to study epilepsy seizure structure. Our contributions are 3-fold. First, we construct an Epilepsy Tensor with three modes, i.e. time samples, scales and electrodes, through wavelet analysis of multi-channel ictal EEG. Second, we demonstrate that multiway analysis techniques, in particular parallel factor analysis (PARAFAC), provide promising results in modeling the complex structure of an epilepsy seizure, localizing a seizure origin and extracting artifacts. Third, we introduce an approach for removing artifacts using multilinear subspace analysis and discuss its merits and drawbacks. RESULTS: Ictal EEG analysis of 10 seizures from 7 patients are included in this study. Our results for 8 seizures match with clinical observations in terms of seizure origin and extracted artifacts. On the other hand, for 2 of the seizures, seizure localization is not achieved using an initial trial of PARAFAC modeling. In these cases, first, we apply an artifact removal method and subsequently apply the PARAFAC model on the epilepsy tensor from which potential artifacts have been removed. This method successfully identifies the seizure origin in both cases.     

Nonlinear analysis of EEG signals: Surrogate data analysis

ObjectivesThe electroencephalogram (EEG) signal contains information about the state and condition of the brain. The aim of the study is to conduct a nonlinear analysis of the EEG signals and to compare the differences in the nonlinear characteristics of the EEG during normal state and the epileptic state.DataThe EEG data used for this study – which consisted of epileptic EEG and normal EEG – were obtained from the EEG database available with the Bonn University, Germany.ResultsThe attractors seen in normal and epileptic human brain dynamics were studied and compared. Surrogate data analyses were conducted on two nonlinear measures, namely the largest Lyapunov exponent and the correlation dimension, to test the hypothesis whether EEG signals were in accordance with linear stochastic models.DiscussionsThe existence of deterministic chaos in brain activity is confirmed by the existence of a chaotic attractor; also, saturation of the correlation dimension towards a definite value is the manifestation of a deterministic dynamics. Also a reduction is observed between the dimensionalities of the brain attractors from normal state to the epileptic state. The evaluation of the largest Lyapunov exponent also confirms the lowering of complexity during an episode of seizure.ConclusionIn case of Lyapunov exponent of EEG data, the change due to surrogating is small suggesting that it is not representing the system complexity properly but there is a marked change in the case of correlation dimension value due to surrogating.     

Widespread EEG Changes Precede Focal Seizures

The process by which the brain transitions into an epileptic seizure is unknown. In this study, we investigated whether the transition to seizure is associated with changes in brain dynamics detectable in the wideband EEG, and whether differences exist across underlying pathologies. Depth electrode ictal EEG recordings from 40 consecutive patients with pharmacoresistant lesional focal epilepsy were low-pass filtered at 500 Hz and sampled at 2,000 Hz. Predefined EEG sections were selected immediately before (immediate preictal), and 30 seconds before the earliest EEG sign suggestive of seizure activity (baseline). Spectral analysis, visual inspection and discrete wavelet transform were used to detect standard (delta, theta, alpha, beta and gamma) and high-frequency bands (ripples and fast ripples). At the group level, each EEG frequency band activity increased significantly from baseline to the immediate preictal section, mostly in a progressive manner and independently of any modification in the state of vigilance. Preictal increases in each frequency band activity were widespread, being observed in the seizure-onset zone and lesional tissue, as well as in remote regions. These changes occurred in all the investigated pathologies (mesial temporal atrophy/sclerosis, local/regional cortical atrophy, and malformations of cortical development), but were more pronounced in mesial temporal atrophy/sclerosis. Our findings indicate that a brain state change with distinctive features, in the form of unidirectional changes across the entire EEG bandwidth, occurs immediately prior to seizure onset. We postulate that these changes might reflect a facilitating state of the brain which enables a susceptible region to generate seizures.     

Patient-Specific Detection of Cerebral Blood Flow Alterations as Assessed by Arterial Spin Labeling in Drug-Resistant Epileptic Patients

Electrophysiological and hemodynamic data can be integrated to accurately and precisely identify the generators of abnormal electrical activity in drug-resistant focal epilepsy. Arterial Spin Labeling (ASL), a magnetic resonance imaging (MRI) technique for quantitative noninvasive measurement of cerebral blood flow (CBF), can provide a direct measure of variations in cerebral perfusion associated with the epileptic focus. In this study, we aimed to confirm the ASL diagnostic value in the identification of the epileptogenic zone, as compared to electrical source imaging (ESI) results, and to apply a template-based approach to depict statistically significant CBF alterations. Standard video-electroencephalography (EEG), high-density EEG, and ASL were performed to identify clinical seizure semiology and noninvasively localize the epileptic focus in 12 drug-resistant focal epilepsy patients. The same ASL protocol was applied to a control group of 17 healthy volunteers from which a normal perfusion template was constructed using a mixed-effect approach. CBF maps of each patient were then statistically compared to the reference template to identify perfusion alterations. Significant hypo- and hyperperfused areas were identified in all cases, showing good agreement between ASL and ESI results. Interictal hypoperfusion was observed at the site of the seizure in 10/12 patients and early postictal hyperperfusion in 2/12. The epileptic focus was correctly identified within the surgical resection margins in the 5 patients who underwent lobectomy, all of which had good postsurgical outcomes. The combined use of ESI and ASL can aid in the noninvasive evaluation of drug-resistant epileptic patients.     

颅内电极在感觉运动区皮质发育不良癫痫手术中的应用

目的：探讨长程颅内电极监测及电刺激方法,在感觉运动区皮质发育不良的难治性癫痫外科手术评估中的意义。方法：筛选MRI提示的皮质发育不良区域与重要功能区-感觉运动区位置关系密切的11例难治性癫痫患者,且头皮长程视频脑电监测及PET检查也初步提示癫痫发作与皮质发育不良所在脑区有关,在可疑脑区放置颅内电极,然后进行颅内电极长程视频脑电监测及电刺激检测,对癫痫起源位置及功能区定位,明确癫痫发作起源区域与感觉运动功能区的解剖学关系,在定位结果指导下进行切除术。结果：11例中3例位于左侧半球,8例位于右侧半球,11例感觉运动功能区皮质分布均存在不同程度变异,7例癫痫发作起源区域与感觉运动功能区一定范围重叠,其中5例与感觉区重叠,该5例切除了起源区域与发作有关的部分感觉区,2例部分致痫灶与运动区重叠,该2例仅切除了除与发作有关的运动区以外的癫痫起源区域,4例癫痫发作起源区域与感觉运动功能区相对独立,该4例完全切除癫痫发作起源区域;手术后6例患者发作消失,2例患者发作频率减少90%以上,1例癫痫发作控制无效,2例患者发生部分感觉缺失,但对生活无明显影响。结论：在皮质发育不良的癫痫患者中,有较高比例的病人伴有功能区皮层分布的变异,长程颅内电极监测及电刺激能够实现癫痫起源区域及功能区精确定位,明确功能区变异情况,对于指导病灶切除,避免损伤功能区皮质,减少术后并发症具有重要意义。     

颅内电极在感觉运动区皮质发育不良癫痫手术中的应用

目的:探讨长程颅内电极监测及电刺激方法,在感觉运动区皮质发育不良的难治性癫痫外科手术评估中的意义。方法:筛选MRI提示的皮质发育不良区域与重要功能区-感觉运动区位置关系密切的11例难治性癫痫患者,且头皮长程视频脑电监测及PET检查也初步提示癫痫发作与皮质发育不良所在脑区有关,在可疑脑区放置颅内电极,然后进行颅内电极长程视频脑电监测及电刺激检测,对癫痫起源位置及功能区定位,明确癫痫发作起源区域与感觉运动功能区的解剖学关系,在定位结果指导下进行切除术。结果:11例中3例位于左侧半球,8例位于右侧半球,11例感觉运动功能区皮质分布均存在不同程度变异,7例癫痫发作起源区域与感觉运动功能区一定范围重叠,其中5例与感觉区重叠,该5例切除了起源区域与发作有关的部分感觉区,2例部分致痫灶与运动区重叠,该2例仅切除了除与发作有关的运动区以外的癫痫起源区域,4例癫痫发作起源区域与感觉运动功能区相对独立,该4例完全切除癫痫发作起源区域;手术后6例患者发作消失,2例患者发作频率减少90%以上,1例癫痫发作控制无效,2例患者发生部分感觉缺失,但对生活无明显影响。结论:在皮质发育不良的癫痫患者中,有较高比例的病人伴有功能区皮层分布的变异,长程颅内电极监测及电刺激能够实现癫痫起源区域及功能区精确定位,明确功能区变异情况,对于指导病灶切除,避免损伤功能区皮质,减少术后并发症具有重要意义。     

An index of signal mode complexity based on orthogonal transformation

Irregular and complex signals are ubiquitous in nature. The principal aim of this paper is to develop an index, quantifying the complexity of such signals, which is based on the distribution of the strengths of its orthogonal oscillatory modes estimated by singular value decomposition. The index is first tested with simulated chaotic and/or stochastic maps and flows. Among neural data analysis, the index is first applied to a cognitive EEG data set recorded from two groups, musicians and non-musicians, during listening to music and resting state. In the gamma band (30–50 Hz), musicians showed robust changes in complexity, consistent over various scalp regions, during listening to music from resting condition, whereas such changes were minimal for non-musicians. Then the index is used to separate healthy participants from epileptic and manic patients based on spontaneous EEG analysis. Finally, it is used to track a tonic-clonic seizure EEG signal, and a conspicuous change was found in the complexity profiles of delta band (1–3.5 Hz) oscillations at the onset of seizure. We conclude that this index would be useful for quantification of a wide range of time series including neural oscillations.     

Patient-Specific Epileptic Seizure Prediction in Long-Term Scalp EEG Signal Using Multivariate Statistical Process Control

An accurate epileptic seizure prediction algorithm can alleviate the problem and reduce risks in the life of a patient suffering from epilepsy. The main motive of this work is to propose a model which can predict seizures well in advance of its occurrence. Multivariate statistical process control (MSPC) has been used for seizure predictions in long-term scalp EEG signal. It has been observed that excessive neuronal activity in the preictal period of seizure changes the electrical characteristic from chaotic to rhythmic behavior. These changes have been utilized for prediction. Eight temporal based features are used for predicting the seizures by using multivariate statistical process control, which is widely known as an anomaly monitoring method. 90 seizures from the CHB-MIT EEG data of ten patients are analyzed.ResultThe results of the proposed method demonstrated that 80 seizures out of 90 in preictal period were correctly predicted prior to the seizure onset, thereby giving a sensitivity of 88.89%. The false positive rate is observed to 0.39 per hour.ConclusionThis study proposed a temporal based patient-specific epileptic seizure prediction method using MSPC in long-term scalp EEG signals. It also provides the possibility of realizing an EEG-based epileptic seizure prediction system which requires less computational power.SignificanceThe proposed method does not require preictal data for modeling. The extracted features are computationally easy. The tested result shows good accuracy on the CHB-MIT data base.     

柴胡对癫痫模型电活动的调制

目的 :研究柴胡对癫痫发作的影响。方法 :以家兔和大鼠为实验对象 ,用毛果芸香碱致痫 ,采用脑电图和细胞外玻璃微电极记录技术 ,观察柴胡对癫痫模型大脑皮层放电及海马脑片场电位的影响。结果 :腹腔注射柴胡后可使癫痫发作次数及发作持续时间显著减少 ,发作间隔时间显著延长 ,(P <0 .0 5 ) ,脑片旁滴注柴胡后使致痫大鼠海马脑片诱发场电位幅度平均降低 2 0 .4 1% ,恢复时间平均为 6 .86min ,(P <0 .0 1)。结论 :柴胡注射液能明显抑制癫痫模型电活动 ,提示柴胡具有抗痫作用     

A Physiology-Based Seizure Detection System for Multichannel EEG

Background

Epilepsy is a common chronic neurological disorder characterized by recurrent unprovoked seizures. Electroencephalogram (EEG) signals play a critical role in the diagnosis of epilepsy. Multichannel EEGs contain more information than do single-channel EEGs. Automatic detection algorithms for spikes or seizures have traditionally been implemented on single-channel EEG, and algorithms for multichannel EEG are unavailable.

Methodology

This study proposes a physiology-based detection system for epileptic seizures that uses multichannel EEG signals. The proposed technique was tested on two EEG data sets acquired from 18 patients. Both unipolar and bipolar EEG signals were analyzed. We employed sample entropy (SampEn), statistical values, and concepts used in clinical neurophysiology (e.g., phase reversals and potential fields of a bipolar EEG) to extract the features. We further tested the performance of a genetic algorithm cascaded with a support vector machine and post-classification spike matching.

Principal Findings

We obtained 86.69% spike detection and 99.77% seizure detection for Data Set I. The detection system was further validated using the model trained by Data Set I on Data Set II. The system again showed high performance, with 91.18% detection of spikes and 99.22% seizure detection.

Conclusion

We report a de novo EEG classification system for seizure and spike detection on multichannel EEG that includes physiology-based knowledge to enhance the performance of this type of system.