Application of hybrid GLCT-PICA de-noising method in automated EEG artifact removal. (July 2020)

• Record Type: Journal Article
• Authors: Jindal, K.; Upadhyay, R.; Singh, H.S.
• Is Part Of: Biomedical signal processing and control. Volume 60(2020)
• Publication Date: 2020-07
• Subjects: Electroencephalogram -- Artifact removal -- Independent component analysis -- Katz-Fractal sparsity -- General Linear-Chirplet Transform (GLCT); Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2020.101977
• Languages: English
• ISSNs: 1746-8094

Abstract

Graphical abstract: Highlights: A novel autonomous EEG artifact removal method using joint application of fpICA and GLCT is proposed. The proposed GLCT-PICA EEG de-noising method is implemented on simulated and experimentally recorded EEG activity. The criterion of Katz Fractal Sparsity is developed and assessed for automatic identification of artifactual ICs. The criterion of spike zone thresholding is developed and assessed for GLCT based artifactual ICs de-noising. The results of the study reveal that the proposed EEG de-noising method performed better than conventional ICA based methods. Abstract: The electrical activities associated with non-cerebral biological origins are usually having high amplitude (order of 230–350 micro-volts) and effect on-going cerebral activity (order of 7–20 micro-volts) adversely. The frequent occurrence of multiple artifactual origins makes it imperative to adopt adequate artifacts removal methodologies prior to feature estimation in Brain-Computer Interface applications. The present work proposes a novel artifact removal methodology using a joint application of Fast-Power Independent Component Analysis and General Linear Chirplet Transform for automatic identification and rejection of artifactual origins. After segregating on-going Electroencephalogram activity into Independent Components, Katz-Fractal Sparsity criterion is employed to identify artifactual components. The identified artifactual components are treated by General Linear Chirplet Transform-based EEG de-noising method to recover useful cerebral information leaked with artifactual origins. Thereafter, Inverse Independent Component Analysis yields artifact corrected clean Electroencephalogram activity for further analysis. The effectiveness of the proposed methodology is validated with simulated and empirical Electroencephalogram dataset. The experimental results establish the proposed method as a potential candidate for non-cerebral artifacts correction and noise suppression from Electroencephalogram records.