Abstract

Info

Author(s):: LIANG Peiwen; ZHENG Mengxun; TANG Long; ZHANG Hua; Guangdong Provincial Key Laboratory of Medical Image Processing/School of Biomedical Engineering, Southern Medical University,Guangzhou 510515, China

Abstract:: Objective To propose a conditional diffusion model based on template priors for correcting streaky metal artifactscharacterized by alternating bright and dark patterns that appear in reconstructed computed tomography (CT) images due tothe presence of metallic implants. Methods After isolating metal region using image segmentation technology to generatemetal trajectory projection data, the projection data of metal trajectory and the corresponding projection data of template priorimages were excluded. Subsequently, a conditional diffusion model was developed to recover the missing portions of theprojection data in the metal-affected regions. Finally, the restored projection data were reconstructed using filtered backprojection to obtain the corrected image. Results Compared with artifact disentanglement network and diffusion model, theproposed approach improved peak signal to noise ratio by 3.44 and 0.749 dB, and increased structural similarity index by0.079 and 0.015, respectively. Conclusion The proposed approach outperforms traditional metal artifact correction methodsand exhibits superior performance in reducing both streaky and shadow artifacts.