Implementation of Contextual Clustering Method for Statistical Parametric Maps in Functional Magnetic Resonance Images

Senthamarai Kannan1 and C. Vijayalakshmi2
1Department of Statistics, Manonmaniam Sundaranar University, India
2Department of Mathematics, Sathyabama University, India

Abstract: A contextual clustering procedure for Statistical Parametric Maps is calculated from time varying three-dimensional images. The algorithm can be used for the detection of neural activations from functional Magnetic Resonance Images. Ogawa et al. (1990) have discussed about the brain magnetic resonance imaging with contrast dependent on blood oxygenation concepts. Subsequently, the processing strategies for time-course data sets in functional magnetic resonance imaging of the human brain have been analyzed by Bandettini et al. (1993). By using the voxel by voxel testing technique, the neighborhood information is utilized and this is achieved by using a Markov random field prior concept and Iterated Conditional Modes algorithm. The simulation results and human functional magnetic resonance imaging experiments using visual stimulation demonstrate that a better sensitivity is achieved with a given specifications in comparison with the voxel-by-voxel thresholding technique.  

Keywords: Brain imaging, structural anatomy, auditory signal processing, statistical parametric mapping functional magnetic resonance images, fMRI.