AI-adoptie in de cardiologie: regulering op een mondiaal kruispunt

<p>The integration of artificial intelligence (AI) and machine learning (ML) into cardiology represents one of the most profound opportunities to transform patient care. Automated diagnostics, predictive analytics and prescriptive treatment recommendations promise to reshape workflows, improve efficiency and get us closer to the much-heralded precision cardiovascular medicine. Yet, as the scoping review by <I>Hussain et al</I><cross-ref type="bib" refid="R1">1</cross-ref> makes clear, this promise is hampered by regulatory shortfalls that could inadvertently compromise both safety and innovation.</p> <p>Within just the past few years, nearly 300 AI-driven medical devices have been cleared by the United States Food and Drug Administation (FDA) for clinical use in cardiology and this number will almost certainly grow rapidly in the future.<cross-ref type="bib" refid="R2">2</cross-ref> Excitement in transformer architectures and foundation models suggests that reinventions and updates of