Method and system for reconstructing cone beam CT image

Abstract

The invention relates to a method and system for reconstructing a cone beam CT image. The method comprises steps of: in a process that a ray source motions along a circular orbit, acquiring the projection data of an object irradiated by the ray source at intervals; performing filtered back projection reconstruction on the projection data to obtain a primary reconstructed three-dimensional image; acquiring the number of times that the pixels in the primary reconstructed three-dimensional image are subjected to back projection, and obtaining a secondary reconstructed three-dimensional image according to the number of times that the pixels are subjected to back projection. The method and the system may eliminate a pseudo shadow problem that an area inside the FOV is bright and an area outside the FOV is dark in a reconstructed fault plane in a three-dimensional FOV cone part, and improves the overall quality of the reconstructed image.

Description

technical field [0001] The invention relates to the technical field of computerized tomography, in particular to a method and system for realizing cone-beam CT image reconstruction. Background technique [0002] CT (Computed Tomography), that is, electronic computer tomography imaging, divides the projection body through which the X-ray beam passes into many small units (voxels, voxels), obtains the X-ray attenuation coefficient or absorption coefficient of each voxel, and then arranges them into a matrix , the numeric matrix. Through the digital / analog converter, each number in the digital matrix is ​​converted into a small square of gray scale ranging from black to white, that is, a pixel (pixel), and arranged in a matrix to form a CT image. [0003] With the development of flat panel detector technology, cone beam CT (ConebeamCT, CBCT) has been widely used. Such as figure 1 As shown, the principle is that the X-ray source S uses a cone-shaped X-ray beam with a relative...

AI Technical Summary

Problems solved by technology

[0007] Problem 1: In the circular orbit scanning mode, the FDK reconstruction algorithm does not meet the data completeness conditions for accurate reconstruction. When the X-ray beam cone angle increases, artifacts appear in the reconstruction results. The specific performance is: the reconstructed image is far away from the plane where the scanning orbit is located Pixel value drops;

[0008] Question 2: In the circular orbit scanning mode, the FDK reconstruction algorithm requires that the projection body can be fully irradiated by the X-ray beam at each projection angle (that is, the projection data of the projection body must be completely within the three-dimensional FOV), if the projection If the size of the body in the direction of the rotation axis of the circular orbit is too large, then, although the reconstruction fault plane based on the cylindrical part in the 3D FOV can be well reconstructed, the object plane outside the 2D FOV in the reconstruction fault plane based on the cone part in the 3D FOV The reconstruction quality is low, manifested as artifacts where the area inside the FOV is bright and the area outside the FOV is dark, resulting in the loss of some projection volume information

Images