Development of a novel computational framework to gain mechanistic insights of atrial fibrillation and optimize therapeutic strategies

Atrial fibrillation (AF) is associated with a fivefold increase in stroke risk and accounts for 15–18% of all strokes. Stroke prevention relies on the CHA₂DS₂-VASc score, which, based on general clinical factors, lacks patient-specific predictive accuracy. To overcome this limitation, computational fluid dynamics (CFD) simulations have emerged as a valuable tool, enabling detailed analysis of left atrial appendage (LAA) blood flow and identifying stasis-prone regions with high thrombotic risk. This research focuses on developing a digital twin model of the left atrium (LA), integrated with an advanced CFD framework to quantify patient-specific hemodynamic parameters in different AF types, including paroxysmal and persistent AF. Special attention is given to LAA flow dynamics, crucial for stroke risk assessment. The model incorporates contrast-enhanced CT and Doppler imaging to personalize LA simulations, enhancing clinical decision-making and guiding therapeutic strategies. Beyond CFD-based stroke risk assessment, this thesis explores AF-related mechanisms and interventions, particularly the role of pulmonary vein (PV) contractions in AF progression and left atrial appendage occlusion (LAAO) as an alternative stroke prevention strategy. Additionally, it investigates the use of 3D-printed models to simulate LAAO device placement, improving procedural outcomes. Findings highlight the potential of computational modeling in refining risk stratification, thromboembolic predictions, and treatment optimization. This comprehensive approach underscores the role of digital twin and CFD models in transforming AF management, offering personalized stroke prevention strategies, improved device testing, and patient-centered care.