An international group of interdisciplinary researchers, led by Charité – Universitätsmedizin Berlin and supported by the German Centre for Cardiovascular Research (DZHK) as well as funding from the German Research Foundation (DFG), has published new recommendations on the individualized medical treatment of arterial deposits (atherosclerosis). What is special: artificial intelligence (AI) is intended to help improve the evaluation of CT scans of the coronary arteries – thus enabling more personalized and effective therapy.
Every year, more than two million people in Europe with chest pain undergo so-called coronary CT angiography – an imaging procedure that can visualize and assess both calcified and non-calcified deposits in the coronary arteries. Such deposits can lead to heart attacks.
Until now, however, there have been no clear recommendations on how exactly the results of these examinations should be used for tailored medical treatment. According to the interdisciplinary experts, AI can help evaluate atherosclerotic plaques more precisely, making the results usable for individualized treatment recommendations. The statement calls for a paradigm shift towards personalized therapy in order to better prevent serious cardiovascular events (figure).
The researchers of the Quantitative Cardiovascular Imaging (QCI) Study Group recommend analyzing CT data with the help of AI to better assess the risk of heart attacks. The amount of plaque is classified quickly and precisely against age- and sex-specific average values using large datasets. The two first authors of the 4th QCI publication in Nature Reviews Cardiology are doctoral candidates in the DFG-funded BIOQIC Research Training Group: Kenrick Schulze and Anne-Marieke Stantien.
Treat early, prevent specifically
“If deposits are visible in CT, this should be a signal to begin medical treatment in addition to lifestyle optimization,” explains Prof. Dr. Dr. Marc Dewey of Charité, coordinator of the QCI Study Group. “If the plaque burden is significantly above a certain age- and sex-specific threshold, the interdisciplinary group recommends a more intensive drug regimen.” Such therapy may mean higher doses of cholesterol-lowering medications such as statins, or combining them with additional agents.
A new path to personalized medicine
The recommendations are based on a multi-stage expert consensus process and data from large studies. They mark a paradigm shift in cardiovascular medicine: moving away from generic treatment based on broad risk factors – toward individualized decisions grounded in actual vascular changes.
“These insights help us treat at-risk patients earlier and more precisely – and spare others from unnecessary therapy,” says Prof. Dewey. The next step is to test the effectiveness of the AI-supported treatment strategy in large randomized clinical trials comparing CT with AI to CT without AI.
Publication: Schulze K, Stantien AM, Williams MC, et al. Coronary CT angiography evaluation with artificial intelligence for individualized medical treatment of atherosclerosis: a Consensus Statement from the QCI Study Group. Nat Rev Cardiol. Published online August 1, 2025
Scientific contact:
Prof. Dr. Dr. Marc Dewey, Charité – University Medicine Berlin, marc.dewey(at)charite.de
Financial support:
The authors thank the DFG for funding the third QCI meeting on quantitative AI-supported assessment of coronary atherosclerosis (DE 1361/32-1).
