Research On The Nerve Fibers Of DTI Imaging Tracer System

Nowadays, neurological diseases and brain tissue pathological diseases, that has influenced people health index, are an important factor of life. So these health problems urgently need to be solved. The deepened study of modern medicine, not only helps to improve human cognition to the mystery of the nervous system, but also provides accurate diagnosis and cure of brain disorders. Therefore, the physiological characteristics and pathological changes of brain research has very urgently theoretical and practical significance.Magnetic Resonance Imaging(MRI)’s principle is applying magnetic field of objects, testing the emit electromagnetic wave, and describe internal structure by electromagnetic wave size. MRI develop rapidly, and its non-invasive technology features make itself widely used in the field of medical image. Diffusion tensor imaging(DTI) is on the basis of the MRI, extending out another kind of imaging technology. Imaging through the water molecules diffusion characteristics of the differences in the structure of brain tissue, and it can reflect the information such as the location of the nerve fiber structure. It’s safety and reliable, thus this technology is a very extensive application in the medical field, and it’s an important basis for clinical diagnosis and nervous system research.Because of the imperfect of the existing technology and disadvantages in the process of image, it’s become very important to filter noise out. It can not only ensure to get the precision of diffusion weighted image(DWI), but also improve the accuracy of the diffusion tensor. So as to ensure that the results of nerve fiber tracking and part of the degree of anisotropy degree of reliable(FA). Existing DTI nerve fiber tracer system, such as DTIStudio, Track Vis, etc., and there is no perfect image filtering process, which can lead to nerve tracking false of the results.To improve the accuracy of nerve fiber tracking results, this paper proposes a multi-level filter based on anisotropy filter. It uses anisotropic filter protective properties of structure and the efficient filter performance to filter DWI noise. After that, the diffusion tensor field is decomposed into different 2D images of tensor field noise reduction processing. After calculation reducing the noise of diffusion tensor field, finally getting the nerve fiber tracking required parameters. Second, by the improved anisotropic filter, controlled by the diffusion tensor of anisotropic diffusion direction of the filter, not only avoid the traditional isotropic filter fuzzy image detail, and it also can guarantee the nerve fibers to related pixels are not smooth, and effective. The filtering algorithm of this paper, through the simulationand validation of the clinical data, effectively improve the accuracy of nerve fiber tracking. Finally, the above research results, they are on the basis of the existing implementation of a system with complete functions, and completing the processing of data, neural tracing visualization parameters and results of analysis.