Abstract
Background: Atherosclerosis imaging-quantitative CT angiography (AI-QCT) guided by artificial intelligence has demonstrated high diagnostic accuracy for assessment of coronary atherosclerosis, stenosis and ischemia (AI-QCT ISCHEMIA ). Methods: CERTAIN was a multicenter study in 5 expert US centers prospectively recruiting 775 consecutive adult patients referred for CCTA. First, physicians were asked to complete an assessment of the patients’ diagnosis, additional imaging plan, intervention plan and medication management plan based on the conventional site assessment. Subsequently, coronary CT angiography (CCTA) exams were analyzed using AI-QCT, after which physicians were asked to repeat the assessment. Results: The 775 patients had a mean age of 64±12 years, 304 (39%) were female. Compared to site CCTA, AI-QCT analysis improved physician’s complete confidence two- to fivefold in every step of the care pathway (Figure). Overall, AI-QCT was associated with at least one clinical category change in over half (389; 50.2%) of patients compared to site CCTA interpretation, including CAD-RADS (304; 39.2%) and plaque burden (194; 25.2%). With implementation of AI-QCT, there was a 67% (43 vs. 14; p <0.001) reduction in patients planned for myocardial perfusion imaging and a 48% (163 vs. 85; p <0.001) reduction in patients referred for PCI or CABG compared to site CCTA. Conclusion: Implementing AI-QCT enabled a single-modality triple-aim assessment of stenosis, quantitative atherosclerosis and ischemia assessment through CCTA that improved physicians’ confidence, enhanced preventive medication uptake while it reduced the need for perfusion imaging and invasive testing.
Sign-in/Register to access full text options 
Free) 
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
