There is a growing need for patient-specific cardiac models for intervention planning, outcome prediction or assessment of cardiac disease progression. However, most of the recent work in cardiovascular simulation relies on generic heart models built from at most one cardiac phase with simplified motion, driven by fluid dynamics equations. We propose to advance the state-of-the-art by exploiting a comprehensive, patient-specific left heart model extracted from 4D Computed Tomography (CT) data. Explicit physiological constrains are captured in the modeling of the left ventricle (including outflow tract), left atrium (including pulmonary veins), mitral valve, and aortic valve (including ascending aorta). By using this patient-specific model as an input to a 3D Navier-Stokes solver we derive realistic hemodynamics, constrained by the local anatomy, along the entire heart cycle. We present a differential assessment of the flow dynamics corresponding to specific heart conditions. The simulation results shed light upon the functional differences between one normal and two diseased hearts - one with a dilated aortic root and one with a bicuspid aortic valve.     «

There is a growing need for patient-specific cardiac models for intervention planning, outcome prediction or assessment of cardiac disease progression. However, most of the recent work in cardiovascular simulation relies on generic heart models built from at most one cardiac phase with simplified motion, driven by fluid dynamics equations. We propose to advance the state-of-the-art by exploiting a comprehensive, patient-specific left heart model extracted from 4D Computed Tomography (CT)...     »