Computational models and data analysis to tackle cardiovascular problems
My overall research interests are in the use of mathematical models and scientific computing to address complex biomedical problems. During my PhD, I have focused on the development of ad hoc software to efficiently simulate cardiovascular hemodynamics and electrocardiology problems. Recently, I have been working on the analysis of medical data of patients with congenital heart defects (such as single ventricles, double outlet right ventricles and bicuspid aortic valves) and simulation of hemodynamics in the interested districts.
You can find some examples here below!
You can find some examples here below!
Effect of wall compliance on hemodynamics of TCPC
The Total Cavopulmonary Connection (TCPC) is the result of a surgical operation done to palliate a congenital heart defect, called "Single Ventricle". Kids born with this anomaly cannot count on the pumping action of both ventricles, to pump blood to the systemic and the pulmonary circulation and are affected by mixing of oxygenated and deoxygenated blood. The TCPC restores the separation of systemic and pulmonary circulations, but the hemodynamics in this district is critical for the patient's condition. We are studying models and method to accurately simulate the blood flow in patient-specific TCPC, taking also into account the motion of the vessels wall, to predict the outcome of different types of treatment and help the clinicians to find the best one.
To take into account the patient-specific motion we developed a methodology to analyze time-resolved 3D MRI images, segment them to extract the position of the lumen in time and then register these position to obtain the displacement of each surface nodes. With this information we can solve the Navier-Stokes equations in a moving domain, using a custom code based on the C++ finite element library LifeV.
Analysis of MRI data to understand the link between morphology and hemodynamics in BAV patients
The Bicuspid Aortic Valve (BAV) is a congenital defect in which the aortic valve is composed of only 2 cuspids instead of 3. This anomaly generates further complications in some cases. We are developing a protocol to identify relationships between anatomy of the valve and hemodynamics in the aortic root, as a first step to understand the pathway to further complications.
Electrocardiology
During my PhD, I worked on modeling and simulation of the action potential propagation in the myocardium.
In particular our group is focusing on the electrical propagation in the left ventricle wall.
Starting from medical images, such as MRI or Spect, we reconstruct the geometry of the ventricles, using VMTK software, and we generate the volume computational grid. |
Numerical simulations of the action potential propagation are carried out using the C++ finite element library LifeV in which we implemented an algorithm to solve the Bidomain and the Monodomain equations for the electrical propagation in the myocardium, coupled with some of the cell ion models available in literature.