Epilepsy detection from electrical characteristics of EEG signals obtained from the brain of undergone subject is a challenge task for both research and neurologist due to the non-stationary and chaotic nature of EEG signals. As epileptic EEG signals contain huge fluctuating information about the functional behavior of the brain, it is hard to distinguish the fundamental dynamic, complex network of EEG signals without considering the strength among the nodes as they are connected with each other on the basis of these strengths. The prior research on natural visibility graph did not consider this issue in epileptic seizure, although it is a very important key point to have representative information from the signals. Hence, this paper aims to introduce a new idea for epilepsy detection using complex network statistical properties by measuring different strengths of the edges in natural visibility graph theory, which is considered as weight. Thus, the proposed method is named “weighted visibility graph”. In this proposed method, first, the epileptic EEG signals are transformed into complex network and then two important statistical properties of a network such as modularity and average weighted degree used for extracting the imperative characteristics from a network of EEG signals. After that, the extracted features are evaluated by two modern machine-learning classifiers such as, support vector machine with a different kernel function and k-nearest neighbor. The experimental results demonstrate that the combined effect of both features is valuable for network metrics to characterize the EEG time series signals in case of weighted complex network generating up to 100% classification accuracy.
1 February 2021 in IEEE Reviews in Biomedical Engineering
Epilepsy is one of the most chronic brain disorder recorded from since 2000 BC. Almost one-third of epileptic patients experience seizures attack even with medicated treatment. The menace of SUDEP (Sudden unexpected death in epilepsy) in an adult epileptic patient is approximately 8-17% more and 34% in a children epileptic patient. The expert neurologist manually analyses the Electroencephalogram (EEG) signals for epilepsy diagnosis. The non-stationary and complex nature of EEG signals this task more error-prone, time-consuming and even expensive. Hence, it is essential to develop automatic epilepsy detection techniques to ensure an appropriate identification and treatment of this disease. Nowadays, graph-theory has been considered as a prominent approach in the neuroscience field. The network-based approach characterizes a hidden sight of brain activity and brain-behavior mapping. The graph-theory not even helps to understand the underlying dynamics of EEG signals at microscopic, mesoscopic, and macroscopic level but also provide the correlation among them. This paper provides a review report about graph-theory based automated epilepsy detection methods. Furthermore, it will assist the expert's neurologist and researchers with the information of complex network-based epilepsy detection and aid the technician for developing an intelligent system that improving the diagnosis of epilepsy disorder.
12 October 2020 in Health Information Science and Systems
Manual staging of sleep data from EEG signals is time consuming, error prone, and resource intensive. These limitations lead to continuous interests of researchers and clinicians toward the automatic sleep stages classification of EEG signals. This paper proposes a scheme for automatic classification of sleep stages using “edge strength of visibility graph technique from single-channel EEG signals. To discover the validity of the proposed study with noise, we perform simulation analysis with two different time series named Lorenz and Rossler time series. The proposed technique accomplishes better performance than other related work for the two standard groups of sleep stages: Rechtschaffen and Kales standard and American Academy of Sleep Medicine.
Automatic diagnosis of epilepsy using electroencephalogram (EEG) signals is a hot topic in medical community as traditional diagnosis relies on tedious visual screening by highly trained clinicians from lengthy EEG recording. Hence, a new methodology to automatically detect epilepsy from EEG signals considering complex network as the principal dynamics of the epileptic EEG signals can be perfectly described by complex network is introduced. A novel edge weight method for visibility graph in the complex network for detection of epilepsy syndrome is presented. The effect of new edge weights for one key characteristic (such as, average weighted degree ) of complex network is investigated. Finally, the extracted feature set is evaluated by two popular machine learning classifiers: support vector machine (SVM) with several kernel functions and linear discriminant analysis. The experimental results on Bonn University datasets show that the proposed approach is able to characterise the epilepsy from EEG signals generating up to 100% classification performance by SVM with polynomial kernel.