Hasanzadeh - Sharif University of Technology
Kasaei - Sharif University of Technology

Magnetic resonance imaging (MRI) techniques provide detailed anatomic information noninvasively and without the use of ionizing radiation. The development of nau pulse seql.ences in MRI has allotyed obtaining images with high clinical importance and thtts joint analysis (multispectral MN) rr required for interpretation of these images. Fuzzy rule-based systems can combine many inpuls from widely varying sources so that they can be useful for description of tissues in the muhispectral MN. In a fuzry system, an error-free and optimized classifier can be obtained by genetic algorithms. In this paper, we have utilized a geneticfuzzy system for modeling dffirent tissues in brain MN as fuzzy classifers and have segmented the MR images by a combination ofthese classifiers using the evidence theory and the Dempster rule. Experiments were performed using the simulated brain data (SBD) set. The numerical validation of the results demonstrates the strength of the proposed algorithm for medical image segmentation using either the evidence theory or a maximization process as the combination step.

Magnetic resonance imaging (MRI), multispectral MRI, fuzzy system, genetic algorithm, Evidence Theory, Dempster rule