Fig. 5: Comparison of automated and manual CAC scoring on non-gated chest CT exams using internal and MESA datasets.
From: Automated coronary calcium scoring using deep learning with multicenter external validation
Confusion matrices for Stanford (a) and MESA (b) test sets comparing automated scoring non-gated chest CT exams to ground truth scores. Ground truth scores are on the y-axis and model prediction are on the x-axis of each matrix.
