A study published today in the JACC: Cardiovascular Imaging shows that artificial intelligence tools can more rapidly, and objectively, determine calcium scores in computed tomographic (CT) and positron emission tomographic (PET) images than physicians, even when obtained from very-low-radiation CT attenuation scans.
These calcium scores found within the heart provide an accurate measure of atherosclerosis—a buildup of fats, cholesterol and other substances found in the artery walls that can lead to serious cardiac conditions.
Assessment of coronary artery calcium (CAC) by CT imaging provides an accurate measure of atherosclerotic burden. Coronary artery calcium is also visible in CT attenuation correction scans, always acquired with cardiac PET imaging.
The novel deep learning model, originally developed for video applications, was adapted to rapidly quantify coronary artery calcium. The model was trained using 9,543 expert-annotated CT scans and was tested in 4,331 patients from an external cohort undergoing PET/CT imaging with major adverse cardiac events. Same-day paired electrocardiographically gated CAC scans were available in 2,737 patients.
The CT attenuation maps were obtained with PET/CT scans and could be processed by artificial intelligence techniques for rapid and objective determination of coronary calcium score without additional scan and radiation.
Using these artificial intelligence and deep learning techniques requires less imaging, less radiation and lower costs, says senior author of the study Piotr Slomka, Ph.D., a research scientist in the Smidt Heart Institute at Cedars-Sinai, director of Innovation in Imaging and professor of Cardiology and Medicine in the Division of Artificial Intelligence in Medicine.