PLoS One. 2017 Sep 8;12(9):e0184729. doi: 10.1371/journal.pone.0184729.

Fully-coupled fluid-structure interaction simulation of the aortic and mitral valves in a realistic 3D left ventricle model.

Wenbin Mao1 ¶, Andrés Caballero1 ¶, Raymond McKay2, Charles Primiano2, Wei Sun1 * 

  1. Tissue Mechanics Laboratory, The Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, Atlanta, GA, USA
  1. Cardiology Department, The Hartford Hospital, Hartford, Connecticut, USA

¶ These authors contributed equally to this work. 

* Corresponding author




In this study, we present a fully-coupled fluid-structure interaction (FSI) framework that combines smoothed particle hydrodynamics (SPH) and nonlinear finite element (FE) method to investigate the coupled aortic and mitral valves structural response and the bulk intraventricular hemodynamics in a realistic left ventricle (LV) model during the entire cardiac cycle. The FSI model incorporates valve structures that consider native asymmetric leaflet geometries, anisotropic hyperelastic material models and human material properties. Comparison of FSI results with subject-specific echocardiography data demonstrates that the SPH-FE approach is able to quantitatively predict the opening and closing times of the valves, the mitral leaflet opening and closing angles, and the large-scale intraventricular flow phenomena with a reasonable agreement. Moreover, comparison of FSI results with an LV model without valves reveals substantial differences in the flow field. Peak systolic velocities obtained from the FSI model and the LV model without valves are 2.56 m/s and 1.16 m/s, respectively, compared to the Doppler echo data of 2.17 m/s. The proposed SPH-FE FSI framework represents a further step towards modeling patient-specific coupled LV-valve dynamics and has the potential to improve our understanding of cardiovascular physiology and to support professionals in clinical decision-making.



Cardiovascular diseases (CVDs) are the leading global cause of death, accounting for >17.3 million deaths per year, or 31.5% in 2013. The left heart (LH) is a key player in the cardiovascular system, as diseases of and associated with the LV-valve complex account for a large share of CVD-related deaths. Detailed investigation of heart function has been actively pursued clinically in the last decade, with cardiac computational modeling emerging as a potential approach to gain an in-depth understanding of the biomechanics of the LH under healthy and diseased states.


This study tackled an important problem in cardiac computational modeling: being able to model the 3D coupled dynamics of the aortic valve (AV), the mitral (MV) valve, the blood flow, and the cardiac wall during the entire cardiac cycle. This was done by developing a novel computational framework that allowed us to move from a simplified de-coupled modeling approach in which the structure (i.e. cardiac tissue) is modeled separately from the fluid (i.e. blood), to an integrative fluid-structure interaction (FSI) modeling approach.


The LH FSI model developed in this study included detailed valve structures, cardiac wall deformation and pulsatile hemodynamic loads by considering anisotropic hyperelastic material models and human leaflet material properties. Simulation results were compared to subject-specific in vivo measurements, clinical data and experimental results, which demonstrated the capabilities of the FSI framework to simulate the healthy LH coupled valves structural response and bulk ventricular hemodynamics during the cardiac cycle.


While significant progress is still required in the modeling procedures before patient-specific cardiac models can be used in clinical practice, their potential to aid understanding of CVDs, surgical/procedural decision-making and personalization of treatments is undeniable. Computational modeling can provide highly controlled and quantitative analyses to study the distinct effects of various patient-specific anatomies, material properties, and loading conditions. Once a computational model is created, it can be used to quantify the relative contribution of each mechanism in the human heart towards some phenomenon of interest. FSI computational tools, like the one developed in this study, are an enabling step towards patient-specific modeling for clinical translation and can give detailed insights into the flow and structural phenomena under healthy, diseased and repaired states that are difficult or impossible to measure in vivo or in experiments.



Figure 1. Blood velocity at a) peak systole, b) early diastole and c) late diastole.




Video 1. Blood and leaflet dynamics during the cardiac cycle.