Triangular Surface Reconstruction of CT Images by Using Isosurface Construction

Abstract

With the development of the modern medical imaging technology and the computer technology, and their interaction as well, rapid prototyping manufacturing, and to design or product a 3D visualized accessory system have become reality. Currently, the information of human body obtained by various medical image equipments such as CT, MRI, is limited to 2D plane image of each layer. In order to ascertain the space relation, size, shape of pathologic body, it will have important theoretic and clinic practical meaning when using computer to take 3D reconstruction and rapid prototyping manufacturing from human organs and pathologic body. Aiming at the key technologies of 3D reconstruction from medical CT images, this paper analyzes and uses the isosurface construction method to construct its 3D models. The paper indicates the shortage of the normal MC algorithm that causes huge consumption of operation and is hard to work out, and applies SMC algorithm to take 3D reconstruction from medical spiral CT images. MC method mainly exists some problems, one is that the triangular facets are only approximate to the isosurface, two is that it is also approximate to connect two close edges’ intersection in the same cube with straight line, next, in the method, if the vertex with 0 value and the vertex with 1 value are at both ends of diagonal line respectively, there will be two kinds of possible connecting methods, thus ambiguity exists [1]. Among a few different methods to judge and eliminate ambiguity, the asymptote method [2] is in common use. Aiming at the shortage of MC algorithm and the characteristics of medical CT images, there is another special MC algorithm called SMC, it connects the triangular facets through estimating state value of each vertex of cube, avoids estimating ambiguity by calculating asymptotic line of hyperbolic curve, simultaneously reduces the calculation to the needless cubes, thereby simplifies the reconstruction method. The SMC method is proved to be simple and effective with two samples in this paper.