Cardiovascular diseases are among the most frequent causes of death worldwide. The pathological basis of almost all cardiovascular diseases is atherothrombosis, i.e. atherosclerotic plaque rupture in combination with inflammatory and prothrombotic systemic and local changes in the circulation. Although the prognosis of these diseases has improved in the last decades, a halt of this trend and even a reversal of the trend in some population groups has been observed in the past decade. The long-term consequences of COVID-19 disease could also accelerate this trend. While the inclusion of meaningful risk factors and biomarkers holds enormous clinical potential for prevention and intervention, the state of knowledge here is very limited and there is a great need for action. CurATime, as a cluster project to combine expertise from science and industry, has set itself the goal of developing customised treatment and prevention concepts for cardiovascular diseases and their clinical application. As a sub-project of the overarching CurATime project, CurAIvasc stands for research into novel analyses of vascular imaging (carotid sonography) based on deep learning approaches. New developments in the field open up the possibility of exploiting the previously untapped potential of clinically relevant information from different vascular imaging modalities for in-depth clinical phenotyping. AI-assisted characterisation of vascular structure and function is an innovative approach to improve the precision of conventional measurements, both in terms of time efficiency and quality. Furthermore, the use of modern exAI (explainable AI) techniques also enables the identification of new, previously unknown, risk factors and digital biomarkers, which can subsequently be clinically evaluated.
Universitätsmedizin der Johannes Gutenberg-Universität Mainz