CFD with STAR-CD is widely-known for its unique capabilities to simulate the complexities of gas motion and other in-cylinder processes in automobile engines. Recently, a group of researchers at the Imperial College of Science, Technology and Medicine in London have exploited some of those same capabilities to calculate the flow in another kind of engine – the human heart. It is intended that the simulations will be used initially to assist in clinical diagnosis and ultimately also as a component of virtual surgery, to help in planning the real thing.

The approach developed involves the combined use of CFD and Magnetic Resonance Imaging (MRI), which is the technique employed in Computer-Aided Tomography (CAT) scanning machines. The information obtained by MRI is in the form of thin two-dimensional image ‘slices’, like the example shown in Figure 1, on which the inner surface of one of the heart chambers selected for study, the left ventricle, has been traced.

    ,

Sets of these slices at various levels through the ventricle are obtained at discrete times spanning the filling and emptying phases of the heart cycle. Image processing and geometry reconstruction techniques are then used to determine the complete chamber topology at each time.

CFD, in the form of STAR-CD, then takes over. The time-varying ventricle volume is fitted with a moving mesh (Figure 2), including arbitrary sliding interfaces to accommodate the ‘inlet’ (left atrium) and ‘outlet’ (aorta) passages, at which the boundary conditions are applied. Calculations are performed over a number of heart ‘beats’, until a cyclically-repeating solution is obtained. A snapshot of the predicted flow in a vertical plane during the inflow (diastolic) phase, is shown in Figure 3. The agreement is, as they say, heartening!