Figure 1.

Case example of AI-guided coronary computed tomography angiography. Using a series of validated convolutional neural network models (including VGG19 network, 3D U-Net, and VGG Network variant) for image quality assessment, the machine learning algorithm (Cleerly, New York, NY) selects the best series, identifies and labels all of the major epicardial coronaries and their side branches, determines centerlines, performs coronary segmentation and labeling and then performs a rapid assessment of % stenosis, plaque volume and of adverse plaque characteristics. The data is then displayed in a graphical output to allow for clinical review.