A Computational Investigation for Metal Artifact Reduction in Cone-beam X-Ray Computed Tomography

Due to low radiation dose, cone-beam computed tomography is growing in importance. In image reconstruction, a stack of cross-sectional images can be reconstructed from a series of X-ray radiographs, served as projections. The presence of metals in the object, however, has deteriorated the quality of the reconstructed image as well as the 3D modeling results. The consequence of metal artifact can be demonstrated through the occurrence of streak effects on the reconstruction volume. In this paper, the attempt has been conducted to solve the problem of metal artifacts. The proposed technique is achieved by segmenting the metal shadowgram and filling it with the averaged intensity prior to reconstruction. The result of computer simulation shows the potential of such method for clinical use