Vnitr Lek 2026, 72(1):60-66 | DOI: 10.36290/vnl.2026.009

Artificial intelligence in cardiology: Current clinical applications and regulatory framework in the EU

Lukáš Evin

Kardiovaskulární oddělení, Interní a kardiologická klinika, Fakultní nemocnice Ostrava

In recent years, artificial intelligence (AI) has become a standard part of modern cardiology, influencing diagnostics, interventional treatment, and patient follow-up. It is gaining ground most rapidly in areas with large volumes of structured signals or imaging data, particularly in ECG analysis, long-term ambulatory rhythm monitoring, support systems for electrophysiological procedures, and cardiovascular applications of computed tomography. At the same time, systems are being developed for the urgent triage of critical findings on CT/CTA (Computed Tomography Angiography) scans and tools that support standardized echocardiographic examinations even by less experienced users. This article summarizes selected AI-based medical devices that have declared conformity with the MDR 2017/745 and places their use in the context of the gradually introduced obligations under the AI Act 2024/1689, which supplements the MDR with specific requirements for high-risk AI systems. For each technology, the principle, available clinical validation, and practical impact on clinical workflow are discussed. Key issues for routine practice remain the quality of input data, interpretability of outputs, interoperability, and transferability of validation across populations and healthcare systems. While some tools are already supported by randomized or prospective multicenter evidence, others rely mainly on validation and implementation studies and will require further confirmation in long-term clinical endpoint trials.

Keywords: artificial intelligence, cardiology, ECG, atrial fibrillation, electrophysiology, CT angiography, MDR 2017/745, AI Act 2024/1689.

Accepted: February 3, 2026; Published: February 12, 2026  Show citation