Quantitative analysis of reconstructed 3-D coronary arterial tree and intracoronary devices

Traditional quantitative coronary angiography is performed on two-dimensional (2-D) projection views. These views are chosen by the angiographer to minimize vessel overlap and foreshortening. With 2-D projection views that are acquired in this nonstandardized fashion, however, there is no way to know or estimate how much error occurs in the QCA process. Furthermore, coronary arteries possess a curvilinear shape and undergo a cyclical deformation due to their attachment to the myocardium. Therefore, it is necessary to obtain three-dimensional (3-D) information to best describe and quantify the dynamic curvilinear nature of the human coronary artery. Using a patient-specific 3-D coronary reconstruction algorithm and routine angiographic images, a new technique is proposed to describe: (1) the curvilinear nature of 3-D coronary arteries and intracoronary devices; (2) the magnitude of the arterial deformation caused by intracoronary devices and due to heart motion; and (3) optimal view(s) with respect to the desired "pathway" for delivering intracoronary devices.

I. Introduction

Quantitative coronary angiography (QCA) remains the classical and most commonly used tool to assess the results of coronary pharmacological or mechanical intervention. In the 1970s [1], it was developed to quantify vessel geometry and regression or progression of coronary artery disease. In the mid-1980s, digital systems were introduced into the catheterization laboratory to support the clinician during interventional procedures. This on-line QCA has been used for the selection of interventional device size, and for the assessment of the efficacy of the procedure. Currently, the QCA is performed on two-dimensional (2-D) projection views in which the vessel overlap and foreshortening have been subjectively minimized by the interventionist in a trial-and-error way. In Fig. 1(a), a right coronary artery (RCA) image was utilized for QCA and the 30% narrowing measurement is shown in Fig. 1(b). With 2-D projection views, however, there is no way to know how much error occurs in the QCA process due to foreshortening of the stenotic arterial segment. For example, in Fig. 1(c) and (d), the same vessel segment between two bifurcation points has been marked by two dots. These angiograms depict 77% and 52% foreshortening based on the three-dimensional (3-D) reconstruction results. Therefore, the existing QCA methods based on angiography [2] [3]–[12] allow limited assessment of the severity of coronary narrowings. Fig. 1.

(a) An example of RCA image with a lesion at the mid-segment as indicated by two arrows. (b) The results of 2-D QCA with 30% narrowing. Two typical examples of vessel foreshortening where the LCA segments indicated between two dots represents 77% foreshortening in (c) acquired at viewing angle (LAO 31°, CAUD 35°) and 52% foreshortening in (d) acquired at viewing angle (LAO 35°CRAN 15°)