
EEG Correlates of Visual Signal Processing: Spectral Decomposition using ICA
Author(s) -
Vamsi Krishna Vadla*,
Ramesh Naidu Annavarapu
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.d7315.118419
Subject(s) - electroencephalography , artifact (error) , independent component analysis , computer science , noise (video) , signal (programming language) , pattern recognition (psychology) , artificial intelligence , signal processing , brain activity and meditation , speech recognition , psychology , neuroscience , image (mathematics) , digital signal processing , programming language , computer hardware
Studies on signal processing of biological activities is a way to ascertain the anatomical and functional behavior of complex systems. Many biomedical signals represent their significance for understanding the systems, like ECG signals of heart and EEG, MEG signals of brain’s electro- and magnetobehavior reveals their importance. Availability of enormous amount of EEG data, with unwanted noise and artifacts, from complex systems, is challenging to uncover the underlying dynamics of signals sources. Functional analysis of this data requires various methods to remove the artifacts and noise present in it and further investigation of the system dynamics. In this paper, we discussed the removal of artifacts and noise from brain’s EEG activity to achieve artifact rejections of continuous EEG data and to apply Independent Component Analysis (ICA) to analyze the event-related tasks. Cluster decomposed signal components allow us to visualize the independent components of any number of subjects and subject groups. Differentiating vast electrical activity of an abnormal subject in comparison with a normal subject is possible with the decomposing signal. ICA approach helps in correlating the event-related EEG signals for a subject in two or more conditions of the same experiment; hence, it suggests in creating the data sets of epochs for every condition.