4D flow MR imaging for blood velocity field validation in atrial fibrillation patients

Atrial fibrillation (AF) is associated to a five-fold increase in the risk of stroke and AF strokes are especially severe. Stroke risk is connected to several AF related morphological and functional remodelling mechanisms which favour blood stasis and clot formation inside the left atrium. Indexes routinely used in clinical practice to stratify stroke risk, such us the CHA2DS2-VASc score, are based on empirical information without considering specific hemodynamic implications of AF which might improve the predictive power of such scores. Computational fluid dynamics (CFD) represents a valuable non-invasive approach to determine and assess physically meaningful parameters and indicators in a complex fluid dynamics system. The goal of this thesis is to validate the simulated blood velocity field in LA with 4D flow MR imaging data of three AF patients. CFD simulation was validated with two different patient-specific motion models one obtained from the 4D flow MRI data and the other obtained from the CT data and the results in terms of mitral valve (MV) flowrate were compared. These initial results show that the flow rate at the MV in the case of the simulation with 4D flow model is temporally synchronised with that measured by the 4D flow, unlike the result of the simulation with CT model, where the flow at the MV is out of phase with the one measured from 4D flow.