Fast and Robust Detection of Epilepsy in Noisy EEG Signals Using Permutation Entropy (original) (raw)

Detection of epileptic seizure based on entropy analysis of short-term EEG

PloS one, 2018

Entropy measures that assess signals' complexity have drawn increasing attention recently in biomedical field, as they have shown the ability of capturing unique features that are intrinsic and physiologically meaningful. In this study, we applied entropy analysis to electroencephalogram (EEG) data to examine its performance in epilepsy detection based on short-term EEG, aiming at establishing a short-term analysis protocol with optimal seizure detection performance. Two classification problems were considered, i.e., 1) classifying interictal and ictal EEGs (epileptic group) from normal EEGs; and 2) classifying ictal from interictal EEGs. For each problem, we explored two protocols to analyze the entropy of EEG: i) using a single analytical window with different window lengths, and ii) using an average of multiple windows for each window length. Two entropy methods-fuzzy entropy (FuzzyEn) and distribution entropy (DistEn)-were used that have valid outputs for any given data leng...

Epileptic EEG activity detection for children using entropy-based biomarkers

Neuroscience Informatics

Seizures, which last for a while and are a symptom of epilepsy, are bouts of excessive and abnormally synchronized neuronal activity in the patient’s brain. For young children, in particular, early diagnosis and treatment are essential to optimize the likelihood of the best possible child- specific result. Electroencephalogram (EEG) signals can be inspected to look for epileptic seizures. However, certain epileptic patients with severe cases show high rates of misdiagnosis or failure to notice the seizures, and they do not demonstrate any improvement in healing as a result of their inability to respond to medical treatment. The purpose of this study was to identify EEG biomarkers that may be used to distinguish between children with epilepsy and otherwise healthy and normal subjects. Savitzky-Golay (SG) filter was used to record and analyze the data from 19 EEG channels. EEG background activity was used to calculate amplitude-aware permutation entropy (AAPE) and enhanced permutation entropy (impe). The hypothesis that the irregularity and complexity in epileptic EEG were decreased in comparison with healthy control participants was tested statistically using the t-test (p < 0, 05). As a method of dimensionality reduction, principle component analysis (PCA) was used. The EEG signals of the patients with epileptic seizures were then separated from those of the control individuals using decision tree (DT) and random forest (RF) classifiers. The findings indicate that the EEG of the AAPE and impe was decreased for epileptic patients. A comparison study has been done to see how well the DT and RF classifiers work with the SG filter, AAPE and impe features, and PCA dimensionality reduction technique. When identifying patients with epilepsy and control subjects, PCA with DT and RF produced accuracies of 85% and 80%, respectively, but without the PCA, DT and RF showed accuracies of 75% and 72.5%, respectively. As a result, the EEG may be a trustworthy index for looking at short-term indicators that are sensitive to epileptic identification and classification.

Quantitative EEG based on Renyi Entropy for Epileptic Classification

Journal of Electrical and Electronics Engineering, 2019

Analysis on Electroencephalogram (EEG) signal can provide important information related to the clinical pathology of epilepsy. Detecting the onset, prediction and type of seizures based on EEG signals is very important to determine an appropriate treatment for the patients. However, EEGs have the high complexity with non-linear and non-stationary characteristics; hence, an analysis will be very difficult to do through a visual inspection. Signal processing applications are, therefore, needed to make the interpretation easier. In this study, we proposed a method for EEG analysis based on signal complexity for the epileptic EEG classification. The Renyi entropy was used to extract the data of EEG features, which consist of seizure, interictal and normal features. Then, these features became the input to a classification algorithm. SVM (Support vector machine) classifier was applied to determine the type of that epileptic EEG signal and achieved accuracy of 85 %. This study can be a reference for neurology as an efficient method for epileptic EEG classification

Classification of 5-S Epileptic EEG Recordings Using Distribution Entropy and Sample Entropy

Frontiers in physiology, 2016

Epilepsy is an electrophysiological disorder of the brain, the hallmark of which is recurrent and unprovoked seizures. Electroencephalogram (EEG) measures electrical activity of the brain that is commonly applied as a non-invasive technique for seizure detection. Although a vast number of publications have been published on intelligent algorithms to classify interictal and ictal EEG, it remains an open question whether they can be detected using short-length EEG recordings. In this study, we proposed three protocols to select 5 s EEG segment for classifying interictal and ictal EEG from normal. We used the publicly-accessible Bonn database, which consists of normal, interical, and ictal EEG signals with a length of 4097 sampling points (23.6 s) per record. In this study, we selected three segments of 868 points (5 s) length from each recordings and evaluated results for each of them separately. The well-studied irregularity measure-sample entropy (SampEn)-and a more recently propose...

Study of Learning Entropy for onset detection of epileptic seizures in EEG time series

2016 International Joint Conference on Neural Networks (IJCNN), 2016

This paper presents a case study of non-Shannon entropy, i.e. Learning Entropy (LE), for instant detection of onset of epileptic seizures in individual EEG time series. Contrary to entropy methods of EEG evaluation that are based on probabilistic computations, we present the LE-based approach that evaluates the conformity of individual samples of data to the contemporary learned governing law of a learning system and thus LE can detect changes of dynamics on individual samples of data. For comparison, the principle and the results are compared to the Sample Entropy approach. The promising results indicate the LE potentials for feature extraction enhancement for early detection of epileptic seizures on individual-data-sample basis.

Epileptic Seizure detection with Permutation Fuzzy Entropy using robust machine learning techniques

IEEE Access, 2019

The automatic and accurate determination of the epileptogenic area can assist doctors in presurgical evaluation by providing higher security and quality of life. Visual inspection of electroencephalogram (EEG) signals is expensive, timeconsuming and prone to errors. Several numbers of automated seizure detection frameworks were proposed to replace the traditional methods and to assist neurophysiologists in identifying epileptic seizures accurately. However, these systems lagged in achieving high performance due to the anti-noise ability of feature extraction techniques, while EEG signals are highly susceptible to noise during acquisition. The present study put forwards a new entropy index Permutation Fuzzy Entropy (PFEN), which may delineate between ictal and interictal state of epileptic seizure using different machine learning classifiers. 10-fold cross-validation has been used to avoid the over-fitting of the classification model to achieve unbiased, stable, and reliable performance. The proposed index correctly distinguishes ictal and interictal states with an average accuracy of 98.72%, sensitivity of 98.82% and a specificity of 98.63%, across 21 patients with six epileptic seizure origins. The proposed system manifests the fact that lower PFEN characterizes the EEG during seizure state than in the Interictal seizure state. The study also helps us to investigate the more profound enactment of different classifiers in term of their distance metrics, learning rate, distance, weights, multiple scales, etc. rather than the conventional methods in the literature. Compared to other state of art entropy-based feature extraction methods, PFEN showed its potential to be a promising nonlinear feature for achieving high accuracy and efficiency in seizure detection. It also show’s its feasibility towards the development of a real-time EEG-based brain monitoring system for epileptic seizure detection.