February 17, 2025
The Institute of Informatics & Telecommunications hosted invited speaker Maria Macarulla for a talk titled Non-invasive estimation of heart electrical activity using artificial intelligence on Monday 24 February 2025 at 12.30pm EEST.
The talk took place physically at the Lecture Room of the Institute of Informatics & Telecommunications, while participation was also possible via zoom.
Watch the video of the talk here:
About the talk: Atrial fibrillation (AF) is the most common cardiac arrhythmia, increasing the risk of stroke and heart failure and posing significant challenges in both diagnosis and treatment planning. Current clinical approaches rely heavily on invasive procedures, such as catheter-based electroanatomic mapping, to characterize atrial electrical activity and guide ablation therapy. These methods are time-consuming, costly, and associated with procedural risks. Non-invasive alternatives, such as electrocardiographic (ECG) recordings, provide a safer and more accessible option; however, interpreting surface ECG signals to infer underlying atrial activity remains a complex task. This talk explores the use of artificial intelligence, specifically graph neural networks, to bridge this interpretability gap.
Short Bio: Maria Macarulla Rodriguez is a PhD student specializing in biomedical engineering, with a focus on artificial intelligence for cardiology. She is part of the Cardiac Oriented Research (COR) group at the Universidad Politécnica de Valencia and collaborates with hospitals such as Hospital General Universitario Gregorio Marañón (Madrid), Hospital Universitario y Politécnico La Fe (Valencia), and Hospital Clínic de Barcelona. COR group research focuses on the non-invasive estimation of cardiac electrical activity for personalised treatments. Currently, María is working on atrial modeling and AI-driven analysis of atrial fibrillation simulations. Maria obtained her bachelor’s degree in Telecommunications Engineering in her hometown, Valladolid, and later specialized in signal processing during her master’s studies in Delft, the Netherlands. She is currently conducting a 3 months research stay at the Institute of Informatics and Telecommunications at NCSR Demokritos, where she is developing a GNN model to predict the outcomes of ablation therapies for atrial fibrillation.