The Dynamics of Heart Valves
IIHR Research Engineer K.B. Chandran, who retired in 2014, has an impressive collection of mechanical heart valves, gathered during his 30-plus years of study in biomedical engineering. He dedicated his career to understanding the unique fluid mechanics of the human circulatory system, and the dynamics of heart valves. Chandran served as the Battershell Chair Professor of Biomedical and Mechanical Engineering and former chair of biomedical engineering at the University of Iowa.
The Heart of the Matter
Chandran and his team contributed to the continuing evolution of mechanical heart valves. It’s a complex area of study, and one of the complicating factors is blood itself, which doesn’t behave like most fluids. As anyone who has ever had a cut knows, blood clots to prevent blood loss. A clot, or thrombus, can also form around a mechanical heart valve of any type and cause problems. The researchers want to know why, and how it could be prevented.
What makes blood unique among fluids? “Forty-five percent by volume of blood is blood cells, like red blood cells, white blood cells, and platelets,” Chandran explains. “We cannot just look at blood as a simple fluid. We have to study the interaction of the red blood cells and platelets as blood flows past a mechanical valve to determine why the clots form.”
Flow past a heart valve includes constant interaction between flowing blood and the moving leaflets, so simulations must involve the development of a fluid-structure interaction analysis. Variations in anatomy and other factors often play a role. “That creates a lot of challenges in the development of this fluid-structure algorithm,” Chandran says. “It’s much more complicated than normal engineering flow problems.”
Biomedical engineers use computer simulations (known as computational fluid dynamics, or CFD) to predict the flow field around a mechanical valve, but again, blood complicates things. “In a small 1-mm cube of blood, there are millions of red blood cells,” Chandran explains. “Currently the computer cannot handle computing all that.”
So engineers are developing multi-scale analysis, which begins by performing the simulation at the large scale, and then zooms in to see the detailed mechanics within a small region. In addition to the CFD work, IIHR researchers also perform a variety of experiments dealing with flow around a mechanical heart valve. As an example, using micro-PIV (micron-resolution particle image velocimetry), researchers can measure fluid and blood cell dynamics in a microscopic flow. The results from the experimental work can then be used as a benchmark to validate the CFD simulations.
But the simulations are still more than any one computer — even a super computer — can process. For such large-scale computations, they use a cluster of more than 200 computers. The software “delegates” part of the computation to each processor. The computers do the work simultaneously, much faster than a single computer could. “That’s called parallelizing the computation,” Chandran says. “As technology develops and we get more understanding, we can get better and better simulations.”
For Chandran, the opportunity to improve the quality of life for others made the work especially rewarding. “It’s a problem close to my heart,” he says.
Note: Chandran and two of his engineering faculty colleagues are the editors of a book: Image-based Computational Modeling of the Human Circulatory and Pulmonary Systems.