Coronary Flow Estimation for the Computational Assessment of Fractional Flow Reserve

Nowadays, fractional flow reserve (FFR) is considered the gold standard technique to assess risk of myocardial ischemia in the presence of coronary artery disease. Moreover, FFR is an invasive procedure, which requires specialized cardiologist and dedicated medical instrumentation, i.e. it is far from being risk free and it is expensive. In this context, a tool to estimate FFR from computational fluid dynamics non-invasively could impact positively the patient experience, reducing economic costs and providing a new diagnostic tool for physicians. Although there are some studies proposing computational solutions for the estimation of FFR, they generally lack of sensitivity analysis of the hemodynamics parameters. In this work we are interested in assessing the effect of coronary flow reserve (CFR) in the outcomes of the numerical simulations of coronary blood flow. To this end we make use of a set of 24 coronary computed tomography angiography (CCTA) images from which the arterial network is segmented and utilized to perform blood flow simulations. The blood circulation is modeled using lumped mathematical representations in a steady state regime, with geometrical features retrieved from the CCTA images. At least one measurement of fractional flow reserve (FFR) is available for each patient, totaling 35 measurements. Some hemodynamic parameters for the simulations were found to be patient specific while others are calibrated with a single general value for all patients. The study focuses on the estimation of CFR that minimizes the difference between the in-vivo FFR measurements and the computational estimations. This strategy may shed light on the underlying mechanisms ruling territorial myocardial resistance.