Fig. 7: CAC scoring case examples.
From: Automated coronary calcium scoring using deep learning with multicenter external validation
Coronary artery segmentation (reference image on left, manual in middle; automated on right) in four different patients (a–d). Row a shows qualitatively similar performance between automated and manual methods of CAC scoring on non-gated chest CTs. Row b shows model prediction in a patient with significant aortic root calcification. Note that the model does not misclassify this as CAC. Row c is an example of false negative prediction by the model in the left circumflex. Row d is an example of a false positive in the left circumflex.
