EEG-based Classification of Epileptic and Non-Epileptic Events using Multi-Array Decomposition

  title={EEG-based Classification of Epileptic and Non-Epileptic Events using Multi-Array Decomposition},
  author={Evangelia Pippa and Vasileios G. Kanas and Evangelia I. Zacharaki and Vasiliki Tsirka and Michalis Koutroumanidis and Vasileios Megalooikonomou},
  journal={Int. J. Monit. Surveillance Technol. Res.},
In this paper, the classification of epileptic and non-epileptic events from EEG is investigated based on temporal and spectral analysis and two different schemes for the formulation of the training set. Although matrix representation which treats features as concatenated vectors allows capturing dependencies across channels, it leads to significant increase of feature vector dimensionality and lacks a means of modeling dependencies between features. Thus, the authors compare the commonly used… 

Figures and Tables from this paper

Multi-channel EEG epileptic spike detection by a new method of tensor decomposition.
The proposed SMLRAT method can provide a practical solution to distinguish epileptic spikes from artifacts in real-life EEG datasets and compared the proposed feature extraction method with the state-of-the-art methods.
Nonnegative Tensor Decomposition for EEG Epileptic Spike Detection
A new tensor-based approach to detect epileptic spikes in EEG data based on estimating the so-called “eigenspikes” was proposed, and experimental results showed that the proposed method is efficient in detecting epilept spikes.
New Feature Selection Method for Multi-channel EEG Epileptic Spike Detection System
A new feature selection method is proposed that combines the Fisher score and p-value feature selection methods to rank the features by using the longest common sequences (LCS) to separate epileptic and non-epileptic spikes.
Wavelet transform-based feature extraction approach for epileptic seizure classification
The primary results show that, in the proposed approach, the performance of ANN is better than NN, SVM, and CNN.
A comparative study of machine learning algorithms for epileptic seizure classification on EEG signals
  • E. Imah, Arif Widodo
  • Computer Science
    2017 International Conference on Advanced Computer Science and Information Systems (ICACSIS)
  • 2017
The classification algorithms compared in this study are Generalized Relevance Learning Vector Quantization (GRLVQ), Backpropagation, SVM, and Random Forest, combined with Wavelet and PCA feature extraction, which shows the best performance in recognizing the five classes in EEG epileptic seizure dataset.
Epileptic seizures classification in EEG using PCA based genetic algorithm through machine learning
A Principal Component Analysis with Genetic Algorithm based Machine Learning approach is developed for the binary classification of epileptic seizures from the EEG dataset and is applied to the existing machine learning models to evaluate the model performance in comparison to the higher number of features.
  • Anjali Vasant Kadwe
  • Computer Science
    International Journal of Engineering Applied Sciences and Technology
  • 2019
The proposed method is based on the classification of EEG signal with the less number of sample and more accurately by using the Matlab software, which provides an overall classification accuracy of 99%.
A Machine Learning Model to Predict Seizure Susceptibility from Resting-State fMRI Connectivity
A novel method of assessing seizure susceptibility using data from 49 patients enrolled in the Epilepsy Bioinformatics Study for Antiepileptogenic Therapy (EpiBioS4Rx) is presented, employing a machine learning paradigm that utilizes a Random Forest classifier trained with resting-state functional magnetic resonance imaging (fMRI) data to predict seizure outcomes.
Classification task-driven efficient feature extraction from tensor data
The Efficient Greedy Feature Extraction method (EGFE) is proposed that avoids solving optimization problems of very high dimension and can be stopped when the extracted features are deemed to be sufficient for a proper discrimination of the classes.


Improving classification of epileptic and non-epileptic EEG events by feature selection
Classification of epileptic and non-epileptic EEG events
  • Evangelia Pippa, E. Zacharaki, M. Koutroumanidis
  • Computer Science
    2014 4th International Conference on Wireless Mobile Communication and Healthcare - Transforming Healthcare Through Innovations in Mobile and Wireless Technologies (MOBIHEALTH)
  • 2014
The non-epileptic class consists of two types of paroxysmal episodes of loss of consciousness namely the psychogenic non epileptic seizure (PNES) and the vasovagal syncope (VVS).
Patient-specific seizure prediction using a multi-feature and multi-modal EEG-ECG classification
A patient-specific classification algorithm is proposed to distinguish between preictal and interictal features extracted from ECG-EEG recordings and it is demonstrated that the classifier based on a Support Vector Machine (SVM) can distinguish preictsal from interictsal with a high degree of sensitivity and specificity.
Diagnostic Test for the Discrimination between Interictal Epileptic and Non-Epileptic Pathological EEG Events using Auto–Cross-Correlation Methods
A novel algorithm that produces auto-correlated coefficients from an appropriate, selected EEG segment based on a prototype spectral Variation of an extended spectrum in which alpha, beta, and gamma activities are included may be characterized as accurate for recognition purposes.
Anticipating epileptic seizures in real time by a non-linear analysis of similarity between EEG recordings.
A new method is described to analyze long-term non-stationarity in the EEG by a measure of dynamical similarity between different parts of the time series to track in real time spatio-temporal changes in brain dynamics several minutes prior to seizure.
Emotion Recognition From EEG Using Higher Order Crossings
A novel emotion evocation and EEG-based feature extraction technique is presented, in which the mirror neuron system concept was adapted to efficiently foster emotion induction by the process of imitation, justifying the efficiency of the proposed approach.
Tensor decompositions for feature extraction and classification of high dimensional datasets
This work proposes algorithms for feature extraction and classification based on orthogonal or nonnegative tensor (multi-array) decompositions, and higher order (multilinear) discriminant analysis (HODA), whereby input data are considered as tensors instead of more conventional vector or matrix representations.
Ictal heart rate differentiates epileptic from non-epileptic seizures
Video-EKG-EEG recordings of 67 epileptic seizures and 38 psychogenic nonepileptic seizures were reviewed and it was concluded that documenting an increase in HR during a spell of unresponsiveness usually can distinguish between epileptic and psychogenic causes.
Time‐frequency mapping of the rhythmic limb movements distinguishes convulsive epileptic from psychogenic nonepileptic seizures
The potential for time‐frequency mapping of data from a movement‐recording device (accelerometer) worn on the wrist as a diagnostic tool to differentiate between convulsive epileptic seizures and PNES is examined.