An Information theoretic representation of brain connectivity for cognitive state classification using functional magnetic resonance imaging

Download
2013
Önal, Itır
In this study, a new method for analyzing and representing the discriminative information, distributed in functional Magnetic Resonance Imaging (fMRI) data, is proposed. For this purpose, a local mesh with varying size is formed around each voxel, called the seed voxel. The relationships among each seed voxel and its neighbors are estimated using a linear regression equation by minimizing the expectation of the squared error. This squared error coming from linear regression is used to calculate various information theoretic criteria. Then, the optimal mesh size, which represents the connections among a voxel and its neighbors, is estimated by minimizing these information theoretic criteria with respect to mesh size. The optimal mesh size is used to represent the degree of connectivity such that if the optimal mesh size is small, then the voxel is assumed to be connected with a small number of neighbors. On the other hand, high optimal mesh size indicates that voxels are massively connected. The proposed method shows that the local mesh size with the highest discriminative power depends on the participants, samples in the experiment, and voxels. The results indicate that the local mesh model with optimal mesh size can successfully represent discriminative information.
Citation Formats
I. Önal, “An Information theoretic representation of brain connectivity for cognitive state classification using functional magnetic resonance imaging,” M.S. - Master of Science, Middle East Technical University, 2013.