﻿ Mathematical Sciences: Better Prediction of Heart Attacks and Strokes - WPI

# Better Prediction of Heart Attacks and Strokes

Dalin Tang Uses Computational Models and Medical Imaging Data to Attack a Major Killer
by Ami Albernaz

An arterial plaque—a fatty deposit made of cholesterol, fat, calcium, and other materials—may be a ticking time bomb: Under certain conditions, and without warning, it could rupture, sending debris and blood clots formed at the rupture site to the brain, causing a stroke. It takes years for these plaques to grow, and while small they usually go unnoticed. Once a plaque becomes large enough to block 70 percent of an artery, surgery is often recommended to remove it and prevent a possible rupture and stroke.

Yet, that surgery may be over-prescribed because surgeons want to err on the safe side, says Dalin Tang, professor of mathematical sciences and biomedical engineering at WPI. In fact, the literature indicates that of 20 surgeries, only one will actually prevent a stroke. All surgery entails risk, and an unnecessary endarterectomy may damage the artery and cause harm by releasing debris into the bloodstream. More precise tools to predict which plaques are likely to rupture will allow doctors to cut down on unneeded surgeries, Tang believes. He's made finding such tools his life's work.

With the help of a four-year, \$1.4 million NIH grant awarded earlier this year (a renewal of a previous \$1.1 million NIH award), Tang and colleagues at Washington University in St. Louis and the University of Washington in Seattle have been working on a model that integrates fluid dynamics, solid mechanics, and histological and medical imaging to improve prediction. Tang believes the likelihood of rupture depends on two general factors: the composition of the plaque itself and mechanical forces acting on it.

To get a detailed picture of a plaque, in vivo, MRI data were acquired from patients and segmented to get plaque morphology and tissue components; this data is then validated with histology. They surmise that plaques covered by a thick, protective "cap" may be less likely to rupture than plaques with caps that are thin or worn away, and that plaques with larger lipid-rich cores may be more vulnerable than plaques with smaller cores. Yet blood flow and stress on the plaque also play an important role, and Tang and his colleagues believe figuring out exactly what that role is will ultimately help doctors make better decisions. "If nothing acts on the plaque, it won't rupture," Tang says.

In one recent study of 12 patients, Tang and colleagues found that plaques that had previously ruptured had higher plaque wall stress and flow shear stress values compared with plaques that had not ruptured. In another study of 14 patients, the researchers found that these stresses seemed to correlate with advanced plaque progression. Though more research is needed for the team to refine its model, Tang, who has been working on computational modeling for cardiovascular diseases for 27 years, is both patient and perseverant, recognizing that the challenge may be difficult, but it's eminently worthwhile. "It's hard to predict if a plaque will rupture," he says. "If it were easy, people would not have died."

(From the 2010 edition of WPI Research, the university's research magazine)