Accurate Lumen Segmentation and Stenosis Detection and Quantification in Coronary CTA

Accurate detection and quantification of coronary artery stenoses is fundamental in correct patient diagnosis and optimal treatment planning. Central to detection and quantification is the correct estimation of lumen in the coronary vessel. Here, we describe a method for accurate lumen segmentation in CTA and present results of an early prototype for stenosis detection and quantification for a variety of datasets. For the detection, a sensitivity of 72% and a PPV of 17% is obtained as compared to QCA, while a sensitivity of 57% and a PPV of 18% is achieved as compared to CTA. The stenosis degree is estimated with an absolute average difference of 49% as compared to QCA, and a weighted kappa value of -0.15 is obtained as compared to CTA. A Dice of 73% and 69% is reported for lumen segmentation of healthy and diseased vessel segments respectively.