Advancing cardiac MRI: from Diagnosis to Intervention

This thesis is based on the hypothesis that recent innovations in cardiac MRI (CMR) have the potential to significantly strengthen the role of this technique: from a primarily diagnostic tool to an integrated modality that supports both precision medicine and interventional applications. The studies presented across three domains — quantitative tissue characterization and myocardial deformation, artificial intelligence, and interventional imaging — were designed to test this hypothesis and to explore how CMR can evolve beyond its current boundaries. CMR has become an indispensable tool in cardiovascular medicine, valued for its noninvasive nature and high diagnostic accuracy. Over the past twenty years, significant advancements have further enhanced its capabilities. Innovations such as parametric mapping and strain imaging have improved myocardial tissue characterization and enabled the early detection of subtle functional abnormalities. Additionally, the integration of artificial intelligence has simplified image acquisition and analysis, while supporting more precise clinical decision-making. Finally, CMR-guided interventions offer promising advantages such as the absence of ionizing radiation and superior soft tissue contrast compared to standard fluoroscopy. This thesis explores novel strategies to advance the diagnostic, prognostic, and interventional applications of CMR, with a particular focus on emerging technologies and AI integration.