Automatic Epileptic Seizure Detection in EEG Signals Using Multi-Domain Feature Extraction and Nonlinear Analysis

Abstract

Epileptic seizure detection is commonly implemented by expert clinicians with visual observation of electroencephalography (EEG) signals, which tends to be time consuming and sensitive to bias. The epileptic detection in most previous research suffers from low power and unsuitability for processing large datasets. Therefore, a computerized epileptic seizure detection method is highly required to eradicate the aforementioned problems, expedite epilepsy research and aid medical professionals. In this work, we propose an automatic epilepsy diagnosis framework based on the combination of multi-domain feature extraction and nonlinear analysis of EEG signals. Firstly, EEG signals are pre-processed by using the wavelet threshold method to remove the artifacts. We then extract representative features in the time domain, frequency domain, time-frequency domain and nonlinear analysis features based on the information theory. These features are further extracted in five frequency sub-bands based on the clinical interest, and the dimension of the original feature space is then reduced by using both a principal component analysis and an analysis of variance. Furthermore, the optimal combination of the extracted features is identified and evaluated via different classifiers for the epileptic seizure detection of EEG signals. Finally, the performance of the proposed method is investigated by using a public EEG database at the University Hospital Bonn, Germany. Experimental results demonstrate that the proposed epileptic seizure detection method can achieve a high average accuracy of 99.25%, indicating a powerful method in the detection and classification of epileptic seizures. The proposed seizure detection scheme is thus hoped to eliminate the burden of expert clinicians when they are processing a large number of data by visual observation and to speed-up the epilepsy diagnosis.

Biosignal Generation and Latent Variable Analysis With Recurrent Generative Adversarial Networks

“…More specifically, generator G receives a sequence with z t and c t simultaneously at each time point t, where z t and c t are combined and input as a vector. At each time point t, z t is independently sampled from a uniform random Algorithm 1 Training procedure of the proposed method Require: (1) , z (2) , · · · , z (m) } from noise prior p z • Randomly generate m class-label sequences {c…”

“…if first update at this iteration then • Save weights of the discriminator θ {z (1) , z (2) , · · · , z (m) } from noise prior p z • Randomly generate m class-label sequences {c…”

An Amplitudes-Perturbation Data Augmentation Method in Convolutional Neural Networks for EEG Decoding

“…Traditional feature extraction methods mainly include frequency band analysis [8], multiscale radial basis functions [9], independent component analysis [10], continuous wavelet transform [11] ， and common spatial pattern algorithm [12] etc. In the classification stage, many traditional algorithms such as support vector machine (SVM) [13] and Bayesian classifier [14] have been employed. However, these methods heavily rely on handcrafted features, and the feature selection steps are timeconsuming even for experts in this domain.…”

Multiresolution Analysis in EEG Signal Feature Engineering for Epileptic Seizure Detection

“…The estimated values of the histogram bins are the features for classification of an EEG segments as a seizure or non-seizure. Lina Wang, et al developed a comprehensive approach of feature engineering by combining all three domain features including statistical features in time domain and obtained higher average accuracy [5]. Neural Network based BCI model proposed by Ankita Mazumder , et al, [6] used time varying adaptive autoregressive algorithm (TVAAR) for extraction of features in time domain.…”

“…According to Lina Wang et al [5], multiresolution analysis (MRA) of feature engineering produced better EEG signal processing results. Wavelet Energy and entropy are considered as the prime features for wavelet analysis as reported by Yatindra et al [8].…”