Development of a cone angle weighted three-dimensional image reconstruction algorithm to reduce cone-beam artefacts

Area of Science:

• Medical Imaging
• Image Reconstruction
• Cone-Beam Computed Tomography (CBCT)

Background:

• The Feldkamp algorithm is standard for cone-beam reconstruction but limited by small cone angles.
• Cone-beam artifacts degrade image quality, especially with larger cone angles.

Purpose of the Study:

• To develop and validate an algorithm for reducing cone-beam artifacts.
• To improve image quality at equivalent radiation doses by increasing the effective cone angle.

Main Methods:

• Computer simulations and phantom studies were used to analyze artifact origins.
• A novel algorithm incorporating projection angle weighting and 3D back-projection was developed.
• The algorithm adjusts data weighting based on projection angles and detector positions.

Main Results:

• Simulations and human head studies demonstrated significant reduction in cone-beam artifacts.
• Artifact reduction was particularly effective for objects at the periphery of the reconstruction planes.
• The proposed method achieved satisfactory image quality with increased cone angles.

Conclusions:

• A projection angle weight-based algorithm effectively reduces cone-beam artifacts.
• The algorithm allows for a multi-fold increase in cone angle, improving image quality.
• This approach achieves high-quality reconstructions at the same radiation dose.