Font Size: a A A

Three-dimensional Reconstruction Of Angiography Based On Tomograms

Posted on:2008-10-15Degree:MasterType:Thesis
Country:ChinaCandidate:F YuFull Text:PDF
GTID:2178360212974288Subject:Signal and Information Processing
Abstract/Summary:PDF Full Text Request
Digital Subtraction Angiography (DSA) is one of the most important methods in the vessel deseases examination, with which the bones and the parenchymas in the mask images are removed by the computer while the blood vessels are preserved. It is of great importance to study the three-dimensional reconstruction of digital subtraction angiography in order to provide the accurate structure informations of the three-dimensional blood vessels for the clinical diagnosis.Traditional three-dimensional reconstruction of DSA focuses on the direct cone reconstruction, which only suits for the data with small rotation angle. In this paper a novel method is presented, which the rotation DSA data is transformed to the tomogram slice data and then reconstructed to get the imformation of blood vessels. Firstly, the filtered back projection algorithm is introduced based on the Radon transform and the Fourier theorem. This method leads to accurate tomogram slice data. However, Due to the classical filters have little effect on the suppression of the noise in DSA, the motion artifacts are noticeable in the slices. According to the characteristics of the blood noise, the weighted rhombus median filter is employed to remove the motion artifacts existing in the original algorithm. Better image quality are obtained by such scheme. Finally the structure of three-dimensional blood vessels is reconstructed by direct volume rendering algorithm.Experiments with the clinical data from GE DSA modality demonstrate that it is feasible to reconstruct three-dimensional blood vessels structure of DSA based on the tomograms. It provides a novel approach to the rotation DSA three-dimensional reconstruction.
Keywords/Search Tags:Digital Subtraction Angiography, Filtered Back Projection, Radon Transform, Direct Volume Rendering
PDF Full Text Request
Related items