The Cerebral Vein Extraction Based On The Susceptibility Weighted Imaging

The content of abstract.The cerebral venous system plays an important role in ensuring the stability of the brain blood supply. Studies of modern medicine on cerebral venous circulation disease are relatively small. Instead, the researches on the cerebral arterial system are some more. Except for the incidence of cerebral artery blood circulatory system diseases is relatively high, one of the reasons is closely associated with less test methods, more complex structure and more difficult diagnosis of cerebral venous circulatory system. At present, more mainstream and common imaging methods of intracranial venous system include digital subtraction angiography (DSA), magnetic resonance angiography (MRA), computed tomography angiography (CTA), and magnetic susceptibility weighted imaging (SWI). Among these, DSA, CTA, MRA, techniques require injection of contrast agents, and cannot fully distinguish arteries and veins. Susceptibility Weighted Imaging (SWI) which utilizes the differences of the magnetic sensitivity between different tissues, can show the tiny veins in the deep brain very clearly without contrast agents. It has obvious advantages compared with other methods. This paper presents an improved technique of the three-dimensional cerebral vein extraction. The cerebral venous quantification and visualization extraction can provide great help in the accurate positioning, diagnose brain venous diseases and the quantitative analysis of some neurological diseases. But it is difficult to extract the venous system from the original SWI images. There are three reasons, firstly, the veins are very tiny, the second one is the problem of background effects caused by the high frequency field inhomogeneity in the original SWI images, the third one is other paramagnetic materials may be confused with the vein because of the similar gray scales as the vein. In this study, in order to solve the second and third problems, we do the following research work:1. Remove the high frequency field inhomogeneity backgroundeffects. In the SWI after-treatment, by the addition of hop phase detection technology, this research remove the cranial base field.2. Extract the cerebral vein from the SWI image. From the post-treated SWI images, we can use the magnetic susceptibility difference between organizations to extract the cerebral vein. However, due to the interference from other paramagnetic organizations and uneven image background, it greatly increased the difficulty of this work To solve this problem, this study proposes an improved self-adaptive threshold method, which segments out the vein information in the SWI image accurately and noninvasively.3. Realize the 3D rendering of the veins. To show the 3D position information of the vein more vividly and clearly, this study realize the 3D rendering of the segmented vein based on VTK.