A three-dimensional weighted cone beam filtered backprojection (CB-FBP) algorithm for image reconstruction in...

Area of Science:

• Medical Imaging
• Computed Tomography (CT)
• Image Reconstruction

Background:

• The original FDK algorithm for cone beam (CB) CT reconstruction using a circular trajectory suffers from artifacts that worsen with increasing cone angle.
• These artifacts arise from the circular trajectory not meeting the data sufficiency condition (DSC) and conjugate ray inconsistency.
• Circular trajectories are advantageous in various clinical CT applications.

Purpose of the Study:

• To propose a novel 3D weighted axial CB-FBP algorithm for volumetric CT image reconstruction.
• To reduce CB artifacts under a circular source trajectory by addressing conjugate ray inconsistency.
• To maintain the efficiency and structural advantages of the original FDK algorithm.

Main Methods:

• A 3D weighting strategy is applied to projection data before 3D backprojection.
• The weighting suppresses contributions from conjugate rays with larger cone angles, dependent on the reconstruction plane's distance from the central plane.
• The algorithm is implementable in native CB or cone-parallel geometries.

Main Results:

• The proposed 3D weighted axial CB-FBP algorithm substantially suppresses CB artifacts.
• Significant artifact reduction is observed up to moderate cone angles (e.g., detector dimension of 64 x 0.625 mm) in cone-parallel geometry.
• The algorithm preserves the core structure, 1D ramp filtering, and data manipulation efficiency of the original FDK algorithm.

Conclusions:

• The 3D weighted axial CB-FBP algorithm effectively reduces cone beam artifacts in volumetric CT.
• This method offers an improvement over the standard FDK algorithm while retaining its practical benefits.
• The proposed algorithm provides a viable solution for high-quality CT imaging with circular trajectories.