期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2007
卷号：7
期号：11
页码：235-242
出版社：International Journal of Computer Science and Network Security
摘要：The activity patterns in fMRI data represent execution of different physical and mental tasks. Each of these patterns is unique and located in specific location in the brain. The main aim of analyzing these datasets is to localize the areas of the brain that have been activated in a given experiment. The multi stimuli fMRI data tends to appear scattered because of involvement of multiple activity patterns. The detection of prime activity in the scattered data leads to decision making on prime stimulus. The detection of prime activity is possible through identification of principal components in the multivariate data. The principle components always align along the principal axis with the component with maximum variance nearest to the origin and the minimum variance component at the other extreme end. Clustering of the principle components yields groups of components with most similar variance values. The average linkage clustering aids clustering of principle components. The cluster with maximum variance forms the principal component. This component represents the prime activity. The fMRI data set is huge and also the data size for different tasks is dimensionally dissimilar. Dimensionality reduction of high dimensional data helps reduce computational requirements for subsequent operations on the data, eliminates redundancies in the data, and, in cases where the feature data set dimensionality doesn’t match then a common dimension can be arrived at with the available data. All three reasons apply here, and motivate the use of Principal Component Analysis (PCA) a standard method for creating uncorrelated variables from best-fitting linear combinations of the variables in the raw depth data extracted using Statistical Parametric mapping (SPM). This approach is equivalent to finding an orthogonal basis such that the projection onto each successive vector (or\“principal component”) is of maximal variance (and uncorrelated with each previous vector). The templates comprising of principal components represent individual activity. These are then fed to the back propagation training algorithm. The trained network is capable of classifying the test pattern into the corresponding defined class.
关键词：fMRI, Scattered patterns, Prime activity, Principal component analysis, Pattern classification, Statistical Parametric Mapping, Back propagation neural network.Back propagation neural network
