Our group recently attended the Helmholtz AI Conference 2025 in Karlsruhe. It was a highly insightful event, emphasizing critical advancements in AI for science, including foundational models, agentic AI, and energy consumption. Silvia and Julian presented their research via posters (PSII-43 and PSII-46), sparking valuable discussions. This conference offered a crucial opportunity to engage with the AI community and gather new perspectives for our work.

From April 23–26, the 91st Annual Meeting of the German Society of Cardiology (DGK) took place in Mannheim. Joshua and Silvia presented posters on their latest research (pacing frequency dependence of cardiac excitation conduction via fibroblasts and non-invasive atrial cardiomyopathy diagnosis), while Christian gave a talk on the automatic division of bi-atrial geometries. Pascal and Axel also attended, connecting with the cardiology research community.

From March 30 to April 1, the annual meeting of the European Heart Rhythm Association (EHRA) took place, bringing together scientists, healthcare professionals, and key stakeholders from around the world involved in arrhythmia management. Axel Loewe and Silvia Becker represented the IBT at the event. Additionally, Silvia gave a presentation on why it matters how the P-wave is measured in the ECG.

From March 12 to 14, the E-Science Days took place in Heidelberg with the theme “Research Data Management: Challenges in a Changing World.” Axel Loewe and Silvia Becker represented the IBT in the workshop “Information Infrastructures in the Era of Artificial Intelligence: Opportunities and Challenges,” organized by the Leibniz ScienceCampus DiTraRe. Additionally, Silvia Becker, together with Lea Singson from FIZ Karlsruhe, gave a presentation on the legal aspects of digital twins and synthetic ECGs.

A new paper investigating the interactions between reduced extracellular calcium levels and sympathetic stimulation as a compensatory mechanism, potentially contributing to sudden cardiac death in chronic kidney disease patients, has been published in the Journal of Physiology.

Our recent publication "Consensus Statement on the Credibility Assessment of Machine Learning Predictors" in the journal Briefings in Bioinformatics aims to ensure the reliability and credibility of machine learning models used in clinical and biomedical contexts. Together with other leading experts, we outline a structured framework based on 12 foundational statements and propose rigorous guidelines for evaluating and enhancing the credibility of ML predictors, particularly when used in high-stakes healthcare decisions.

A new paper introduces the latest version of the DREAM for simulating cardiac arrhythmias. Published in Engineering with Computers, it provides an efficient approach to reentry simulations in cardiac electrophysiology.
Read the full paper here.