Mixture of autoregressive modeling orders and its implication on single trial EEG classification. (15th December 2016)

• Record Type: Journal Article
• Authors: Atyabi, Adham; Shic, Frederick; Naples, Adam
• Is Part Of: Expert systems with applications. Volume 65(2016)
• Page Start: 164
• Page End: 180
• Publication Date: 2016-12-15
• Subjects: Autoregressive analysis -- Genetic algorithm -- Particle Swarm Optimization -- Electroencephalogram; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2016.08.044
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: Two methods for mixing AR features for EEG signal classification are proposed. Evolutionary and ensemble learning methods are considered. The results are assessed against a set of conventional order estimation methods. The feasibilities are investigated using several BCI competition datasets. Adequacy of Ensemble-based mixture and EA-based fusion methods are shown. Abstract: Autoregressive (AR) models are of commonly utilized feature types in Electroencephalogram (EEG) studies due to offering better resolution, smoother spectra and being applicable to short segments of data. Identifying correct AR's modeling order is an open challenge. Lower model orders poorly represent the signal while higher orders increase noise. Conventional methods for estimating modeling order include Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC) and Final Prediction Error (FPE). This article assesses the hypothesis that appropriate mixture of multiple AR orders is likely to better represent the true signal compared to any single order. Better spectral representation of underlying EEG patterns can increase utility of AR features in Brain Computer Interface (BCI) systems by increasing timely & correctly responsiveness of such systems to operator's thoughts. Two mechanisms of Evolutionary-based fusion and Ensemble-based mixture are utilized for identifying such appropriate mixture of modeling orders. The classification performance of the resultant AR-mixtures are assessed against several conventional methods utilized by the community including (1) A well-known set of commonly used orders suggested by the literature, (2) conventional order estimation approaches (e.g., AIC, BIC and FPE), (3) blind mixture of AR features originated from a range of well-known orders. Five datasets from BCI competition III that contain 2, 3 and 4 motor imagery tasks are considered for the assessment. The results indicate superiority of Ensemble-based modeling order mixture and evolutionary-based order fusion methods within all datasets.