Development of a phantom network for optimization of coronary artery disease imaging using computed tomography

Abstract

Computed Tomography provides a practical means of diagnosing coronary artery disease in patients, but requires improvements in image quality through optimization of scan parameters and development of new image reconstruction software. In this work, the design of an anthropomorphic phantom coronary artery network with appropriate CT attenuation and mechanical properties is described. The static and dynamic mechanical properties of fresh porcine coronary arteries are characterized at physiological loading conditions through uniaxial tensile testing and dynamic mechanical analysis. The Young’s modulus was determined to be 0.16 ± 0.04 MPa, and the storage and loss moduli at 1.25 Hz were 0.02 ± 0.01 MPa and 0.004 ± 0.002 MPa. The static and dynamic mechanical properties and CT attenuations of a platinum-cured silicone rubber, a tin-cured silicone rubber, a polyurethane rubber, and latex rubber previously used in soft tissue phantom applications were investigated and compared to those of coronary arteries. The platinum-cured silicone rubber was found to be the most suitable material for simulation of the arterial walls, with a physiological Young’s modulus of 0.13 ± 0.03 MPa (19% lower than that of the coronary arteries), storage modulus of 0.07 ± 0.03 MPa (273% higher), and loss modulus of 0.005 ± 0.002 MPa (43% higher). The CT Number of the platinum-cured silicone rubber was found to best suit the required range of values representative of the arterial walls. To manufacture the phantom coronary arteries, moulds of the arterial lumen were 3D printed on a commerical SLA printer, onto which the platinum-cured silicone rubber was dip/brush coated. The phantom coronary arteries were attached to a phantom heart model, filled with an iodine contrast and CT scanned. The average CT number of the phantom artery wall was found to be 65 HU, which is in agreement with the accepted range. The work presented in this study demonstrates the feasibility of in-house manufacturing of imaging phantoms.