Magnetic resonance diffusion tensor tractography by searching for minimum curvature deviation near fiber crossing area

Recently, magnetic resonance diffusion tensor imaging (MR-DTI) has been used to analyze the structure of cerebral white matter and diagnose diseases in the human brain. Fiber tracking and tractography techniques using MR-DTI give us information regarding nerve fibers and allow us to diagnose disorders due to abnormalities in these fibers. In general, tractographies can be constructed continuously, unimpeded by the crossing of nerve fibers. When nerve fibers do cross over other nerve fibers, it is difficult to track them, due to the partial volume effect in the areas where they cross one another. In this paper, we propose a new method of fiber tracking that evaluates the deviation in the curvature of the areas of fiber crossing, which is based upon the trajectories of the areas before and after the where fibers crossed one another. The method´s performance was tested using simulation images that illustrate the bilateral visual cortices through the corpus callosum. The results demonstrate the usefulness of the proposed method.