Accurate delineation of coronary arteries from X-ray angiography supports early identification of narrowing or blockages, but is complicated by low contrast and thin, branching vessel structures. This study compares three U-Net-based segmentation architectures — U-Net, U-Net++ and U-Net 3+ — trained on 1,000 images from the ARCADE dataset and evaluated on a held-out set of 300 images. All three architectures achieved comparable performance (Dice scores of 0.748-0.752), with U-Net 3+ performing best overall, U-Net++ showing higher sensitivity to smaller vessels, and U-Net showing higher precision, demonstrating the potential of U-Net-based models for coronary artery segmentation. This work was presented at the 5th Serbian International Conference on Applied Artificial Intelligence (SICAAI 2026), Kragujevac, Serbia, and was carried out within the STRATIFYHF project.
Heart failure is one of the most life-threatening diseases of the modern era, with high global mortality and morbidity rates, motivating the need for long-term outcome prediction. One established tool is the MAGGIC Risk Calculator for Heart Failure, which predicts 3-5 year mortal…
Autonomous driving relies on computationally intensive perception pipelines to continuously detect and track objects in the surrounding environment. While some objects are key to plan safe and effective maneuvers, others may not be relevant and have no impact on the autonomous ve…
The STRATIFYHF project develops an AI-based decision support system integrating patient data with computational modelling to support risk stratification, diagnosis and progression prediction in heart failure. This study evaluates the cost-effectiveness of AI-enhanced ECG screenin…
Distinguishing heart failure with reduced ejection fraction (HFrEF) from heart failure with preserved ejection fraction (HFpEF) is clinically important but challenging. This paper presents an end-to-end deep learning pipeline for automated three-class classification (healthy, HFr…
Morbidity and disease severity in heart failure are commonly assessed using New York Heart Association (NYHA) classes. This study uses non-invasive data — physical examination, symptoms and disease history — to classify patients into four morbidity classes approximati…
The KeepA(n)I platform facilitates the auditing of computer vision systems that tag images, which aid visual communication on the Web and social media, from content moderation to the development of new apps and tools. In particular, KeepA(n)I enables a broad set of stakeholders t…