Deformable Known Component Model-Based Reconstruction for Coronary CT Angiography

Area of Science:

• Medical imaging
• Biomedical engineering
• Radiology

Background:

• Coronary CT angiography (CCTA) is crucial for atherosclerosis detection.
• Cardiac implants like pacemakers and defibrillators cause metal artifacts in CCTA, hindering plaque visualization.
• Existing metal artifact reduction methods are insufficient for visualizing plaque near implant components.

Purpose of the Study:

• To develop and evaluate a novel reconstruction method (STF-dKCR) for reducing metal artifacts in CCTA.
• To improve the visualization of coronary artery plaque in patients with cardiac implants.
• To specifically enhance the imaging of plaque located near pacemaker or defibrillator leads.

Main Methods:

• The STF-dKCR method incorporates a deformable model for metal leads, including novel parameterization and 3D-2D preregistration.
• A polyenergetic forward model estimates x-ray propagation and spectral properties through the metal component.
• The method was validated using physical data from a cardiac phantom with a simulated vessel, metal wire, and Teflon sphere, compared against Filtered Back Projection (FBP) and FBP with metal artifact reduction (FBP-MAR).

Main Results:

• STF-dKCR significantly reduced metal artifacts compared to FBP and FBP-MAR.
• Only STF-dKCR substantially improved the visibility of a simulated plaque (Teflon sphere) near the metal wire.
• The attenuation of the Teflon bead improved with STF-dKCR, closely matching the expected value, unlike FBP and FBP-MAR.

Conclusions:

• The proposed STF-dKCR reconstruction method effectively reduces metal artifacts in CCTA.
• This technique shows significant potential for improving plaque visualization in the challenging clinical scenario of patients with cardiac implants.
• STF-dKCR offers a promising solution for enhanced atherosclerosis detection in the presence of wire-shaped metal components.