A method to reduce the data-redundancy artifacts for arbitrary source trajectories in CT imaging

摘要

Irregular source trajectory sometimes is used in CT imaging. This includes the intended ones that are often used for improving the reconstructed image quality, and the unintended ones which are resulted from a motion correction process. These irregular trajectories can sometimes cause a problem called data-redundancy, i.e. different voxels of the object may be measured for different times. Failing to handle the data-redundancy may create artifacts in a reconstructed image. We proposed an approach which can correct these artifacts effectively within an acceptable amount of time. The method are based on two assumptions: (1) an (converged) iterative reconstruction is free from artifacts due to the redundancy; (2) The artifacts due to the redundancy are mostly in low-frequency. Based on these assumptions, we are able to extract the artifacts-only image and, with that to correct a normal analytical reconstructed image with artifacts. The results in a real scan study (which is contaminated by patient motion) showed promising results. Most of the artifacts were successfully removed. Further evaluations are required for more trajectories.