According to the authors, the research has significant clinical potential and could contribute to improving the quality of cardiac care.

“The most important result of our research is the confirmation that image analysis using AI algorithms can precisely determine the risk of future adverse cardiovascular events, i.e., hospitalisation due to coronary artery disease in the future,” said Professor Tomasz Roleder from Wrocław University of Science and Technology.

“AI can support real-time decision-making in clinical practice and pave the way for more personalized cardiology,” he added.

The research, published in the European Heart Journal, was co-authored by Roleder in collaboration with scientists from Radboud University Medical Center in Nijmegen and the Department of Biomedical Engineering and Physics at the Amsterdam University Medical Center.

According to a press release from Wrocław University of Science and Technology, the interdisciplinary team of physicians, engineers, and computer scientists focused on invasive coronary artery diagnostics, specifically intravascular imaging with OCT, which is already used in cardiology in Poland.

OCT provides detailed assessment of coronary atherosclerosis, including identification of so-called thin-capped fibroatheromas (TCFA). “These plaques are characterized by a high lipid content and a thin fibrous cap, which is referred to as thin-capped fibroatheroma (TCFA),” the press release said.

The study found that AI applied to OCT scans can not only detect thin-capped fibromas and atherosclerosis but also predict future cardiovascular events. It offers an alternative to manual image analysis in the core laboratory.

“Comprehensive analysis of coronary vessels using AI in OCT scans can help clinicians identify high-risk patients earlier, plan preventive therapy, and use time more efficiently in the haemodynamics laboratory, which will translate into better treatment outcomes,” the press release said.

wl/ bar/

tr. RL