Reducing computed tomography reconstruction and beam hardening artifacts by data fusion

Reconstruction and beam hardening artifacts can limit the quality of computed tomography (CT) data sets and cause problems for later steps which process these data sets. This paper presents an approach to combine additional X-ray images during CT reconstruction in order to reduce artifacts and to improve the accuracy of data extracted from the CT in subsequent processing steps. In order to achieve this, a modified version of filtered back projection is used. However, this technique can also be applied to other reconstruction algorithms like the maximum likelihood expectation maximization (MLEM). This approach has been investigated with both synthetic voxel data and physical X-ray computed tomography data.