Feature Selection and Classifier Parameters Estimation for EEG Signals Peak Detection Using Particle Swarm Optimization

Abstract

Electroencephalogram (EEG) signal peak detection is widely used in clinical applications. The peak point can be detected using several approaches, including time, frequency, time-frequency, and nonlinear domains depending on various peak features from several models. However, there is no study that provides the importance of every peak feature in contributing to a good and generalized model. In this study, feature selection and classifier parameters estimation based on particle swarm optimization (PSO) are proposed as a framework for peak detection on EEG signals in time domain analysis. Two versions of PSO are used in the study: (1) standard PSO and (2) random asynchronous particle swarm optimization (RA-PSO). The proposed framework tries to find the best combination of all the available features that offers good peak detection and a high classification rate from the results in the conducted experiments. The evaluation results indicate that the accuracy of the peak detection can be improved up to 99.90% and 98.59% for training and testing, respectively, as compared to the framework without feature selection adaptation. Additionally, the proposed framework based on RA-PSO offers a better and reliable classification rate as compared to standard PSO as it produces low variance model.

Vascular dementia classification based on hilbert huang transform as feature extractor

“…There are different techniques for extracting brain data, such as spatial filtering [17], temporal information [18], spectral information and event-related potential (ERP) [19]. Indeed, instead of using functional magnetic resonance imaging (fMRI), other researcher used spectral analysis as an alternative method [20].…”

Improving EEG signal peak detection using feature weight learning of a neural network with random weights for eye event-related applications

“…Several previous studies have relied upon the C3 and C4 channels to record the EEG activation in association with commanded gaze shift [3,30]. Also, the CZ channel is commonly used as a reference for EEG signals; we selected the C3, C4 and CZ EEG channels because they have relatively little contamination from artefacts due to eye blinking [31].…”

“…The utilization of peak detection algorithms has emerged as a useful tool in several physiological signal applications, such as the detection of epileptic activity [1], photoplethysmogram (PPG) monitoring [2] and the detection of eye gaze direction from activity in the frontal eye field [3]. In these applications, the peak detection algorithm is typically implemented in the first step of the signal classification process.…”

Distance Evaluated Simulated Kalman Filter Algorithm for Peak Classification of EEG Signals

“…The utilization of peak classification algorithm has become the most significant approach in several physiological signals applications such as the detection of epileptic EEG signals [8][9], the detection of P300 response in the EEG signals [10], photo-plethysmo-gram (PPG) monitoring [11], electrocardiogram (ECG) monitoring [12][13][14], the analysis of gastric electrical activity (ECA) [15], and the detection of eye gaze direction applications [16]. In those applications, peak classification algorithm is typically located in the first step of a classification process.…”