Abstract
Area stenosis (AS) is emerging as new index to assess the severity of coronary artery disease (CAD). Coronary artery lumen area and its derived area stenosis can be obtained noninvasively from computed tomography angiographic (CTA) images. In this study, we developed a novel approach with 3D centerline-centric straightening method, which consisted of original segmentation, smoothing, fast marching (FM) method and central cutting, to quantify coronary lumen area. We tested the differences of area stenosis in a human patient with suspected CAD by varying the combination of methods (Group 2: original segmentation + no smoothing + FM +central cutting; Group 3: Original segmentation + smoothing + thinning + central cutting; Group 4: denser segmentation + smoothing +FM + central cutting; Group 5: original segmentation + smoothing + FM + one-sided cutting). The data with our method demonstrated patient had area stenosis of 37.5% with low intra-observer variability (1.1%). Group 2 to Group 4 underestimated the area stenosis (AS = 32.3%, 36.8% and 32.0%, respectively), however, Group 5 overestimated the area stenosis (AS = 45.6%). All groups 2 to 5 had much higher intra-observer variability than our method. The computational time per data set is approximately 25 minutes, which demonstrates our methods clinical potential as a real-time cardiac assessment tool.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call