3D Reconstruction Algorithm For Medical Images

Medical image 3D reconstruction is to turn from medical imaging equipment such as CT,MRI sequence of two-dimensional tomography digital images,the method of using computer graphics through a draw or body surface rendering algorithm to reconstruct 3 d surface or internal images,which can visually see the three-dimension form of tissues and organs.Medical image 3D reconstruction plays an important role in human body simulation,prosthetics and plastic surgery,virtual surgery,robotic surgery,real-time surgical navigation,and radiotherapy planning.The 3D reconstruction method of medical image is divided into surface rendering and body rendering,and the surface rendering only draws the surface of the object,i.e.only the pixel points with the same gray value are extracted and the surface is prepared.According to the optical model,the color value and the opacity value of the sampling points need to be resampled,and then the image is drawn according to the algorithm flow of each body.This paper briefly describes the commonly used in surface rendering and volume rendering of several algorithms,including chip level cube reconstruction,moving tetrahedron,subdivision method,surface tracking method,throwing snowballs,shear deformation,such as3D texture mapping algorithm,discusses the moving cube and light projection algorithm principle.The work of this paper mainly has the following two parts.First of all,the typical method to draw the equivalent point coordinates in the moving cube algorithm is linear interpolation,which is reasonable for the points within the same organization.However,at the junction of two different tissues,due to the step change of the gray level of two vertices,the linear interpolation method will lead to a large error.In view of the interface of different draw the importance of accuracy in clinical diagnosis,based on the analysis of linear interpolation to calculate the equivalent point of coordinate error,on the basis of mathematical method proved in organization border adopt point method to calculate the equivalent point coordinates,the accuracy is better than the linear interpolation method.Further,in the interactive performance of reconstruction speed and system demand is higher,but the calculation accuracy without special requirement,can be used for all contour point point method in calculating coordinates,in order to improve the running speed of the system.The experimental results show that the improved algorithm is effective.Second,in the analysis of all kinds of surface rendering and volume rendering common reconstruction algorithm on the basis of the principle and process,as a part of the research program of science and technology,this article is based on C⁺⁺and VTK development environment,the development of medical image three dimensional reconstruction software package.This package is not only realized the moving cube,such as common light projection surface rendering,volume rendering reconstruction algorithm,as the research and comparison,the same reconstruction algorithm,developed multiple versions of different improvement optimization,and support the user according to clinical need,the algorithm module to do a variety of Settings.All the implemented algorithm modules have been tested,and the results show that the software package developed in this paper is not only convenient,but also has a high precision,which can meet the needs of clinical diagnosis.