Artifact reduction in short-scan CBCT by use of optimization-based reconstruction

Area of Science:

• Medical Imaging
• Computational Imaging
• Image Reconstruction

Background:

• Cone-beam computed tomography (CBCT) widely uses the Feldkamp-Davis-Kress (FDK) algorithm for image reconstruction.
• FDK algorithm reconstructions often exhibit artifacts, limiting diagnostic accuracy and application scope.
• Optimization-based reconstruction methods show promise for artifact reduction in CBCT.

Purpose of the Study:

• To investigate optimization-based reconstruction for artifact reduction in short-scan CBCT data.
• To compare the performance of optimization methods against the standard FDK algorithm.
• To assess the efficacy of specific optimization techniques, including image-total-variation (TV) constrained reconstruction.

Main Methods:

• Employed optimization programs, including image-total-variation (TV)-constrained data-divergency minimization.
• Utilized data-weighting matrices, such as the Parker weighting matrix.
• Evaluated reconstruction performance on objects of practical interest to demonstrate artifact reduction.

Main Results:

• Optimization-based reconstruction demonstrated significant artifact reduction in short-scan CBCT data.
• Image-TV-constrained reconstruction proved effective in mitigating artifacts.
• Results indicate superior performance over standard FDK reconstruction in artifact suppression.

Conclusions:

• Appropriately designed optimization-based reconstruction effectively reduces artifacts in CBCT.
• Image-TV-constrained methods offer a viable solution for improving short-scan CBCT image quality.
• Optimization techniques enhance the reliability and applicability of CBCT imaging.