EEG Of Epileptic Patients Research Articles

A hybrid Local Binary Pattern and wavelets based approach for EEG classification for diagnosing epilepsy

Epilepsy is one of the grave neurological ailments affecting approximately 70 million people globally. Detection of epileptic attack is commonly carried out by viewing and analysing long-duration multi-channel EEG records. To counter this time-consuming process, a hybrid Local Binary Pattern-Wavelet based approach, classifying EEG in epileptic patients, is adopted in this research. Epilepsy is characterized by multiple ictal patterns in the form of synchronous epileptiform discharge transients. This work attempts to classify seizure from normal EEG recordings using the low-frequency activity. In order to perform this classification, the EEG signal is filtered and then transformed using Local Binary Pattern (LBP) into a new signal. Discrete Wavelet Transform (DWT) is employed to decompose the obtained signal. Wavelet coefficients are calculated to 5 levels of decomposition. A combination of univariate and bivariate features forms the feature set for seizure detection. This feature set extracted from low-frequency band coefficients helps in bringing out the dispersion, symmetry, and peakedness present in the EEG signal. A novel LBP based Spatio-temporal analysis of the continuous EEG signal for epilepsy detection is carried out on 105 seizures from 14 randomly selected subjects of CHB-MIT EEG database. A sensitivity of 100% is achieved on the CHB-MIT database while long term EEG is being tested with Linear Discriminant Analysis (LDA) classifier. The algorithm works well to obtain a false detection rate (FP/Hour) of 0.59. The specificity of 99.8% is attained with a mean accuracy of 99.6% when tested on 498.9 h of EEG data.

EPILEPTIFORM SPIKE DETECTION VIA CONVOLUTIONAL NEURAL NETWORKS.

The EEG of epileptic patients often contains sharp waveforms called "spikes", occurring between seizures. Detecting such spikes is crucial for diagnosing epilepsy. In this paper, we develop a convolutional neural network (CNN) for detecting spikes in EEG of epileptic patients in an automated fashion. The CNN has a convolutional architecture with filters of various sizes applied to the input layer, leaky ReLUs as activation functions, and a sigmoid output layer. Balanced mini-batches were applied to handle the imbalance in the data set. Leave-one-patient-out cross-validation was carried out to test the CNN and benchmark models on EEG data of five epilepsy patients. We achieved 0.947 AUC for the CNN, while the best performing benchmark model, Support Vector Machines with Gaussian kernel, achieved an AUC of 0.912.

Detection of epileptic activity in absence of EEG interictal epileptic discharges

To assess if the EEGs of epileptic patients without any detectable spikes contain the voltage specific epileptic map. Fifteen patients with pharmaco-resistant focal epilepsy were included. Six minutes of resting state EEG with and without any detectable spikes were selected from long-term recording (31-channels, ref:FCz). Resting-state EEGs from 48 healthy control-subjects were also recorded. For each patient EEG, we calculated the averaged spike and its voltage map. We fitted the spike map on (i) EEG of patient with visible spikes (ii) EEG of the same patient without any visible spike and (iii) EEGs of the 48 controls. The presence of the voltage epileptic map was characterized by: mean-correlation and Global Explained Variance (GEV). For these criteria statistical differences between (1) controls and EEG with spikes, and (2) controls and EEG without spikes were evaluated. The patient-specific epileptic voltage map was significantly more represented in the spike-free EEGs of patients than in EEGs of healthy controls (GEV p = 0.029; mean-correlation p = 0.032). This difference was more accentuated for patient-EEGs containing spikes (GEV p = 0.001, mean-correlation p < 0.001). Scalp EEGs of pharmaco-resistant epilepsy patients contains the epileptic voltage map (index of epileptic activity), even in absence of any detectable interictal epileptic discharges.

V2. Detection of epileptic activity in absence of EEG interictal epileptic discharges

Rationale Aim: to assess if the EEGs of epileptic patients with any detectable spikes contain the voltage epileptic map, specific of the individual patient. Methods Fifteen patients with pharmaco-resistant focal epilepsy were included. Six minutes of EEG with spikes and six minutes without any detectable spikes were selected from long-term monitoring recording during wakeful resting state (EEG 31-channels, ref: FCz). Resting-state EEG from 48 healthy control subjects were also recorded and corrected for artifacts. For the EEG of each patient, we calculated the averaged spike and its voltage map. We fitted the spike map on (i) EEG of patient with visible spikes (ii) EEG of the same patient without any visible spike and (iii) EEGs of the 48 controls. The amount of presence of the voltage epileptic map was characterized using two criteria: mean correlation and Global Explained Variance (GEV). For these criteria statistical differences between (1) controls and EEG with spikes, and (2) controls and EEG without spikes were evaluated using z-scores. Results The patient-specific epileptic voltage map was significantly more represented in the spike-free EEGs of patients than in EEGs of healthy controls (GEV p = 0.029; mean correlation p = 0.032). This difference was more accentuated for the patient EEGs containing spikes (GEV p = 0.001, mean correlation (p Discussion Scalp EEGs of patients with pharmaco-resistant epilepsy contains the epileptic voltage map (index of epileptic activity), even in absence of any detectable interictal epileptic discharges. This finding suggests that epileptic voltage map could be a new epileptic bio-marker (SPUM grant 140332 and 141165).

Latency to the first epileptiform activity in the EEG of epileptic patients

Background: Approximately 56% of patients with epilepsy have interictal discharges. Increasing the diagnostic value of interictal electroencephalography (EEG) may help in increasing the ability to diagnose epilepsy especially in patients with low incidence of interictal discharges. Objective: To determine the latency of the epileptic activity on EEG in epileptic patients. Methods: Two hundred patients with epilepsy of different age groups were recruited in this study. We excluded those who had a definite seizure within the last 24 h and those with acute symptomatic seizures. Continuous EEG (for 2 h up to 24 h) was analyzed to find out the latency of the first clear epileptic activity. Results: Epileptiform activity was detected in 45.45% of the patients in the first 20 min of EEG recording. This percentage increases to 55% after 1 h and 64% after 2 h of recording EEG trace. After 24 h, 85.45% had epileptiform activity. Conclusions: Diagnostic yield of digital EEG increases steeply with increasing the duration of the trace and hence the time of routine EEG should be increased to 1–2 h. A 24 h EEG is of utmost value in diagnosis of epilepsy if the routine EEG is not helpful in this task.

An EEG Analysis Research For Epileptics Using Probabilistic Neural Network

Nowadays, diagnosis for epilepsy depends on many systems helping the neurologists to quickly find interesting segments from the lengthy signal by automatic seizure detection. However, we notice that it is very difficult, to obtain long-term EEG data with seizure activities for epilepsy patients in areas lack of medical resources and trained neurologists. Therefore, we propose to study automated epileptic diagnosis using interictal EEG data that is much easier to collect than ictal data. The research, therefore, aims to develop an automated diagnostic system that can use interictal EEG data to diagnose whether the person is epileptic. To develop such a system, we extract from the EEG data three classes of features which respectively are Petrosian fractal dimension, Higuchi fractal dimension and Hjorth parameters and build a Probabilistic Neural Network (PNN) fed with these features. Meanwhile, we also broach demand for data standardization by analysis with EEG of epileptic patients.

High-frequency (80–500 Hz) oscillations and epileptogenesis in temporal lobe epilepsy

High-frequency oscillations (HFOs), termed ripples (80–200Hz) and fast ripples (250–600Hz), are recorded in the EEG of epileptic patients and in animal epilepsy models; HFOs are thought to reflect pathological activity and seizure onset zones. Here, we analyzed the temporal and spatial evolution of interictal spikes with and without HFOs in the rat pilocarpine model of temporal lobe epilepsy. Depth electrode recordings from dentate gyrus (DG), CA3 region, subiculum and entorhinal cortex (EC), were obtained from rats between the 4th and 15th day after a status epilepticus (SE) induced by i.p. injection of pilocarpine. The first seizure occurred 6.1±2.5days after SE (n=7 rats). Five of 7 animals exhibited interictal spikes that co-occurred with fast ripples accounting for 4.9±4.6% of all analyzed interictal spikes (n=12,886) while all rats showed interictal spikes co-occurring with ripples, accounting for 14.3±3.4% of all events. Increased rates of interictal spikes without HFOs in the EC predicted upcoming seizures on the following day, while rates of interictal spikes with fast ripples in CA3 reflected periods of high seizure occurrence. Finally, interictal spikes co-occurring with ripples did not show any specific relation to seizure occurrence. Our findings identify different temporal and spatial developmental patterns for the rates of interictal spikes with or without HFOs in relation with seizure occurrence. These distinct categories of interictal spikes point at dynamic processes that should bring neuronal networks close to seizure generation.

Seizure frequency and bioelectric brain activity in epileptic patients in stable and unstable atmospheric pressure and temperature in different seasons of the year – a preliminary report

Background and purposeAn epileptic seizure is a sum of exogenous and endogenous factors affecting an epileptic focus. The aim of the study was to examine the influence of changes in atmospheric pressure and temperature on the increase in the frequency of seizures and changes in EEG in epileptic patients. Material and methodsThe study included 30 epileptic patients (aged 19–54) reporting the influence of changes in weather conditions on the increase in the frequency of seizures for at least 2 years. EEG was performed twice each season at the time of stable and unstable weather conditions. ResultsIn stable and unstable weather conditions, epileptic changes in EEG were most often found in winter (in 43.3% and 63.3% of patients, respectively). Unstable weather conditions increased the proportion of patients with epileptic changes in EEG also in the other seasons. Unstable weather conditions caused an increase in the frequency of seizures in 40% of patients in spring, 43.3% in autumn, 40% in winter and in approximately 7% in summer. ConclusionsIn spring, autumn and winter, unstable weather conditions cause an increase in the frequency of seizures in almost half of the epileptic patients but only in 7% in summer. The increase in frequency of seizures in unstable weather conditions did not correspond in all patients with increase of changes in EEG. The higher proportion of epileptic patients with changes in EEG in unstable weather conditions in all seasons suggests an impact of these conditions on subclinical seizure discharges in this period.

Effects of levetiracetam on generalized discharges monitored with ambulatory EEG in epileptic patients

Purpose Quantitative analysis of epileptiform discharges (EDs) before and after the initiation of an antiepileptic treatment is a useful tool to objectively documentate the efficacy of an antiepileptic drug (AED). Aim of this study was to evaluate the effect of levetiracetam (LEV) on EDs, monitored with ambulatory EEG (A/EEG), in a limited series of patients with generalized epilepsy. Methods We performed 24 h A/EEG recording in basal condition and at follow-up after LEV therapy in 21 adult epileptic patients. Eleven received LEV as monotherapy and 10 as add-on. For each patient we quantified total epileptic activity considering the following parameters: total number, total duration, maximal duration and median duration of EDs. Self-reported information on the effect of LEV on clinical seizures was also collected, to determine the electro-clinical correlation. Results A high variability of the response to LEV was observed in the monotherapy group, without statistical differences for all the parameters investigated. A significant reduction of the total number of seizures (113.6 vs. 41.2; p = .01) was observed in patients in add-on therapy. The modifications of epileptiform EEG abnormalities did not necessarily correlate with the self-reported clinical impressions. Discussion The quantification of EDs monitored by A/EEG provides a useful objective support for evaluating the neurophysiologic profile and the real efficacy of an antiepileptic treatment. In our patients LEV was able to significantly reduce the EDs only in add-on therapy. Further larger studies are necessary to clarify the effects of LEV on electro-clinical features of generalized epilepsy.

Permutation entropy to detect vigilance changes and preictal states from scalp EEG in epileptic patients. A preliminary study

Permutation entropy (PE) was recently introduced as a very fast and robust algorithm to detect dynamic complexity changes in time series. It was also suggested as a useful screening algorithm for epileptic events in EEG data. In the present work, we tested its efficacy on scalp EEG data recorded from three epileptic patients. With a receiver operating characteristics (ROC) analysis, we evaluated the separability of amplitude distributions of PE resulting from preictal and interictal phases. Moreover, the dependency of PE on vigilance state was tested by correlation coefficients. A good separability of interictal and preictal phase was found, nevertheless PE was shown to be sensitive to changes in vigilance state. The changes of PE during the preictal phase and at seizure onset coincided with changes in vigilance state, restricting its possible use for seizure prediction on scalp EEG; this finding however suggests its possible usefulness for an automated classification of vigilance states.

Do Interictal Spikes Sustain Seizures and Epileptogenesis?

Interictal spiking is seen in the EEG of epileptic patients between seizures. To date, the roles played by interictal events in seizure occurrence and in epileptogenesis remain elusive. While interictal spikes may herald the onset of electrographic seizures, experimental data indicate that hippocampus-driven interictal events prevent seizure precipitation. Even less clear than the role of interictal events in seizure occurrence is whether and how interictal spikes contribute to epileptogenesis. Thus, while plastic changes within limbic neuronal networks may result from ongoing interictal activity, experimental evidence supports the view that epileptogenesis is accompanied by a decrease in hippocampus-driven interictal activity.

Application of chaos theory to the analysis of EEG in epileptic patients

Application of chaos theory to the analysis of EEG in epileptic patients

Does folic acid exert a provocative action on the EEG of epileptic patients? A preliminary report

Does folic acid exert a provocative action on the EEG of epileptic patients? A preliminary report

Spike detection based on a pattern recognition approach using a microcomputer

A system for the detection of spikes and sharp waves (SSWs) in EEGs of epileptic patients is described. The system uses a pattern recognition technique and it is implemented in a hybrid microcomputer. It performs the on-line simultaneous analysis of 4 EEG channels and thus it is appropriate for long-term EEG recording. The performance of the system was considered satisfactory when compared with that of a panel of electroencephalographers.