A Method for Computer-Assisted 3D Reconstruction of Coronary Arteries Using Angiography Images (Scientific Report)

In this paper, we present a simple method for the 3D reconstruction and visualization of coronary arteries in angiography. We developed an algorithm to manually select vessel borders around the stenotic region. We select twenty cross sections around problematic region. Cross section points are used for 3D reconstruction. In 3D, each cross section defines a circle with a diameter of the length of the cross section and origin at the midpoint of two cross section corners. Relative information about a vessel diameter can be known from circle diameters around corresponding region. Therefore, obtained circles are used for visualizing the change in vessel diameter along the selected part of the vessel. Creating tubular structure by joining these circles in 3D was realized by a Matlab program. Created tubular structure represents the final 3D reconstruction of blood vessel. We applied selection and 3D visualization procedure for two images which belong before and after percutaneous coronary intervention. As shown in the result, the coronary artery was successfully reconstructed in 3D from angiography images.