Single-view geometric calibration for C-arm inverse geometry CT.

Area of Science:

• Medical Imaging
• Computed Tomography
• X-ray Technology

Background:

• Accurate tomographic image reconstruction relies on precise knowledge of imaging system geometry.
• C-arm inverse geometry CT (IGCT) presents unique geometric calibration challenges.
• Artifacts in IGCT can degrade diagnostic image quality.

Purpose of the Study:

• To propose and evaluate a projection matrix-based calibration method for C-arm IGCT.
• To assess the method's effectiveness for scanning-beam digital x-ray (SBDX) systems.
• To reduce artifacts in IGCT reconstructions caused by geometric uncertainties.

Main Methods:

• A helical fiducial phantom was imaged across multiple gantry angles.
• Digital tomosynthesis and composite image generation were performed at each angle.
• A 3D-to-2D projection matrix was constructed to determine C-arm, source, and detector geometry by minimizing fiducial coordinate deviations.

Main Results:

• Simulations showed a reduction in relative root-mean-square error from 7.7% to 0.4% with the calibration method.
• Phantom studies demonstrated the elimination of artifacts in SBDX reconstructions using the proposed IGCT calibration.
• The method effectively determined system geometry even with source and detector translations and rotations.

Conclusions:

• The proposed projection matrix-based calibration method is effective for C-arm IGCT.
• This technique significantly reduces image artifacts in SBDX systems with geometric uncertainties.
• Accurate geometric calibration is crucial for high-quality tomographic image reconstruction in IGCT.