Exploiting Feature Selection and Neural Network Techniques for Identification of Focal and Nonfocal EEG Signals in TQWT Domain

Abstract

For drug resistance patients, removal of a portion of the brain as a cause of epileptic seizures is a surgical remedy. However, before surgery, the detailed analysis of the epilepsy localization area is an essential and logical step. The Electroencephalogram (EEG) signals from these areas are distinct and are referred to as focal, while the EEG signals from other normal areas are known as nonfocal. The visual inspection of multiple channels for detecting the focal EEG signal is time-consuming and prone to human error. To address this challenge, we propose a novel method based on differential operator and Tunable Q-factor wavelet transform (TQWT) to distinguish the focal and nonfocal signals. For this purpose, first, the EEG signal was differenced and then decomposed by TQWT. Second, several entropy-based features were derived from the TQWT subbands. Third, the efficacy of the six binary feature selection algorithms, binary bat algorithm (BBA), binary differential evolution (BDE) algorithm, firefly algorithm (FA), genetic algorithm (GA), grey wolf optimization (GWO), and particle swarm optimization (PSO), was evaluated. In the end, the selected features were fed to several machine learning and neural network classifiers. We observed that the PSO with neural networks provides an effective solution for the application of focal EEG signal detection. The proposed framework resulted in an average classification accuracy of 97.68%, a sensitivity of 97.26%, and a specificity of 98.11% in a tenfold cross-validation strategy, which is higher than the state of the art used in the public Bern-Barcelona EEG database.

Exploiting the Performance of Marine Predators Optimization Algorithm in Combination with Neural Network Classifiers for Breast Mass Classification

“…The sum of the inputs' products and their weights is calculated in a FFNN. This is then fed to the output [39][40][41].…”

Recognizing seizure using Poincaré plot of EEG signals and graphical features in DWT domain

“…These phenomena can visually be evaluated via specialists in this fi eld. In long EEG records, the visual inspection can be a cumbrous and time-consuming action aimed at detecting the presence of epileptic seizures (1,13,14). Therefore, an automatic method would be required to classify seizure (S) and seizure-free (SF) EEG signals.…”

“…According to the literature, several methods have been developed for this purpose. Some of the methods are comprised of permutation entropy (15), horizontal visibility graph (HVG) (16), clustering technique (17), linear prediction error energy (18), fractional linear prediction (FLP) error (19), dual-tree complex wavelet transform (DT-CWT) (20), autoregressive modeling (21), tunable-Q wavelet transform (TQWT) (13,22), reconstructed phase space (RPS) (14), second-order difference plot (SODP) (23,24) , and improved eigenvalue decomposition of Hankel matrix and Hilbert transform (IEVDHM-HT) (25). Most of the latter methods work on the basis of nonlinear features extraction.…”

Accessing Artificial Intelligence for Fetus Health Status Using Hybrid Deep Learning Algorithm (AlexNet-SVM) on Cardiotocographic Data

“…DNNs worked effectively with persistent data using moving filters and max-pooling operations [ 16 ]. However, training DNNs for a converged solution is both time- and space-intensive and impedes their real-time implementation in clinical settings [ 17 , 18 , 19 , 20 , 21 , 22 ]. The resolve of this paper is to make progress toward a real-time clinical support system for all classes of CTG recordings.…”