Science & Research

Model simulates blood disorders

Developed using groups of 10 atoms, the model gives a fuller picture of hematological disorders

By
Contributing Writer
Friday, October 18, 2013

A new model to represent blood circulation may offer a valuable tool in examining different hematological diseases’ effect on bloodflow.

The model, developed by Professor of Applied Mathematics George Karniadakis and collaborators, sheds light on the details of blood disorders at the cellular level and is a step toward visualizing the blood of individual patients to make clinical projections.

The technique uses computational modeling to create visual representations of different blood disorders, Karniadakis said. To give the full picture of blood, Karniadakis had to represent its various components, such as plasma and red blood cells. He then modeled these individual blood componenets in greater molecular detail.

“The bad news is that this takes several weeks to do one simulation like this, on a super computer. This involves millions, perhaps billions, of particles,” Karniadakis said. To solve this problem, the model was created by sorting atoms and molecules into groups of ten.

This “particle method” allowed Karniadakis to represent the blood in great detail, he said. The strategy let him “construct plasma” along with realistic red blood cells, hemoglobin and other blood components, he said.  In employing the model, “you capture correctly the various phenomena” in “extreme cases” such as disease, he said.

Since its development, the model has been applied to various diseases such as sickle cell anemia, in which abnormally shaped red blood cells impede blood flow.

“Researchers and hematologists have long observed that the stiff, sickled red cells interact with the less stiff, but more adhesive, non-sickled red cells during episodes of microvascular occlusion, which are the hallmarks of this dreaded disease,” wrote Wilbur Lam, assisant professor in Emory University’s Department of Pediatrics, in an email to The Herald. “The simulations in this study give us a reasonable explanation and valuable insight into how these interactions actually occur.”

“In traditional approaches, modelers have either performed large-scale simulation, or focused on highly localized phenomena,” wrote Roger Kamm, professor of biological and mechanical engineering at the Massachusetts Institute of Technology, in an email to The Herald. Using Karniadakis’ approach, “you can locally model what happens at scales down to single molecules, but embed those details in calculations performed at the macro-scale, encompassing complete arteries.”

“The calculations are only as good as the way we have to visualize them,” Kamm added.

Representing blood on both a circulatory and particle level requires working with multiple scales.

“The main challenges are the amount of data we might have and a large range of different space and time scales,” wrote Dmitry Fedosov, Karniadakis’ former PhD student, in an email to The Herald. “The involved lengths are from several nanometers to several centimeters, while time may vary from a sub-second to days,” he wrote. Anaylzing blood from multiple scales may help researchers to simulate bloodflow “in real time,” he added.

Real time simulation may make the model more patient-specific, Karniadakis said. Doctors could utilize this model to not only visualize the patient’s current blood condition, but also to monitor and predict the disease’s future trajectory, Karniadakis told American Scientist in a Sept. 30 article titled ‘Blood in Motion’ describing his work.

“This is a very nice step forward in the right direction,” said Paulo Arratia, assistant professor in Penn’s Department of Mechnical Engineering and Applied Mechanics. “We have to continue to improve this model through experimentation. Once we have everything worked together, then we can go to the next step — real applications.”