Epileptic Seizure Detection Using Geometric Features Extracted from SODP Shape of EEG Signals and AsyLnCPSO-GA

Abstract

Epilepsy is a neurological disorder that is characterized by transient and unexpected electrical disturbance of the brain. Seizure detection by electroencephalogram (EEG) is associated with the primary interest of the evaluation and auxiliary diagnosis of epileptic patients. The aim of this study is to establish a hybrid model with improved particle swarm optimization (PSO) and a genetic algorithm (GA) to determine the optimal combination of features for epileptic seizure detection. First, the second-order difference plot (SODP) method was applied, and ten geometric features of epileptic EEG signals were derived in each frequency band (δ, θ, α and β), forming a high-dimensional feature vector. Secondly, an optimization algorithm, AsyLnCPSO-GA, combining a modified PSO with asynchronous learning factor (AsyLnCPSO) and the genetic algorithm (GA) was proposed for feature selection. Finally, the feature combinations were fed to a naïve Bayesian classifier for epileptic seizure and seizure-free identification. The method proposed in this paper achieved 95.35% classification accuracy with a tenfold cross-validation strategy when the interfrequency bands were crossed, serving as an effective method for epilepsy detection, which could help clinicians to expeditiously diagnose epilepsy based on SODP analysis and am optimization algorithm for feature selection.

A novel framework of MOPSO-GDM in recognition of Alzheimer's EEG-based functional network

“…From past studies, hybrid algorithms are commonly used as improved algorithms, such as GA-PSO (Gupta et al, 2019 ; Liu et al, 2020 ), BAPSO (Almadhor et al, 2021 ), PSO-GWO (Dahmani and Yebdri, 2020 ), and hybrid multi-objective optimization algorithms: NSGA-MOPSO (Shuaipeng et al, 2017 ; Xie et al, 2022 ). Nevertheless, most hybrid algorithms directly combined the whole of two algorithms, without considering the complexity of the algorithm (Wang et al, 2022 ). Therefore, in this paper, superfluous operators are eliminated and only mutation is added to simplify the algorithm flow.…”

“…Furthermore, substantia researches showed that network properties can capture the synergistic effect of abnormal paroxysmal firing of neurons more effectively than the characteristics of single channel signals (Wang et al, 2015b ; Busche and Hyman, 2020 ; Gao et al, 2021 ). In addition, the combination of multiple features can help to display EEG abnormalities from various angles and improve the recognition efficiency (Mahato and Paul, 2020 ; Pei et al, 2020 ; Narayan, 2021 ; Wang et al, 2022 ). For instance, Narayan ( 2021 ) extracted and combined the features of AAR parameters, Barlow parameters, Hjorth parameters, etc.…”

“…for the recognition of motor EEG signals. Wang et al ( 2022 ) applied the second-order difference plot to extract geometric features for epileptic seizure detection of EEG. These results all lead to the consistent conclusion that a combination of features can improve classification accuracy compared to a single feature.…”

“…Commonly used evolutionary computation algorithms include particle swarm optimization (PSO) (Kennedy and Eberhart, 1995 ), ant colony optimization (ACO) (Ke et al, 2010 ), artificial bee colony (ABC) algorithms (Ren et al, 2017 ), and genetic algorithm (GA) (Holland, 1992 ). PSO was proposed by Kennedy and Eberhart ( 1995 ), due to their fast convergence speed and stable particle cooperation, it is widely applied in feature selection and highly efficient in applications with high-dimensional data (Samal et al, 2007 ; Tran et al, 2018 ; Chen et al, 2021 ; Wang et al, 2022 ). However, it is noted that PSO has a problem of premature convergence (Deng et al, 2017 ).…”