Title
Component Selection In Blind Source Separation Of Brain Imaging Data
Abstract
Brain imaging technology has been wildly used in neuroscience field. Because original imaging data usually have high dimensionality and a high noise level, dimensionality reduction is generally required to remove noises and retain signals of interest. However conventional dimensionality reduction methods always carry the risk of discarding valuable signals and retaining useless noises. Here we propose a method for component identification to retain only the valuable components. This method is based on the physiological phenomenon in which the intensity of the signal of interest is changed after stimulus onset and is evaluated using stimulated data. The results indicate that the proposed method is valid to distinguish valuable components from all components, retain only the valuable components and improve the signal-to-noise (SNR) of raw imaging data.
Year
DOI
Venue
2017
10.1007/978-3-319-67777-4_53
INTELLIGENCE SCIENCE AND BIG DATA ENGINEERING, ISCIDE 2017
Keywords
Field
DocType
Blind source separation, Component identification, Dimensionality reduction
Dimensionality reduction,Pattern recognition,Computer science,Noise level,Curse of dimensionality,Signal of interest,Artificial intelligence,Stimulus (physiology),Neuroimaging,Blind signal separation,Physiological Phenomenon
Conference
Volume
ISSN
Citations 
10559
0302-9743
0
PageRank 
References 
Authors
0.34
7
4
Name
Order
Citations
PageRank
Xue Wei100.34
Ming Li25595829.00
Lin Yuan300.34
Dewen Hu41290101.20