Donate to Science & Enterprise

S&E on Mastodon

S&E on LinkedIn

S&E on Flipboard

Please share Science & Enterprise

Math Model to Gauge Heart Attack Risk from Plaques

Heart, circulation system

(Bryan Brandenburg, Wikimedia Commons)

22 Aug. 2019. An academic-industry collaboration is creating a non-invasive technique to detect rupture-prone plaque build-ups in arteries and predict one’s risk of heart attack. A team from the Computational Mechanics Group at University of Texas in Austin and the diagnostics company HeartFlow Inc. in Redwood City, California is developing the technology, funded by a three-year, $550,000 grant from National Science Foundation.

The UT-Austin and HeartFlow researchers are seeking better and more cost-effective methods to diagnose a type of coronary artery disease called vulnerable plaques.  Coronary artery disease, also known as atherosclerosis or hardening of the arteries, often results from a build-up of cholesterol plaques on the arteries feeding the heart, and is a major risk factor for heart attacks and other cardiac diseases. Vulnerable plaques containing lipids or fats and other substances build up inside the artery walls, and become inflamed. If these vulnerable plaques rupture, they release debris into the blood stream, causing blood clots and blockages, leading to a heart attack.

The Computational Mechanics Group at UT-Austin studies computational geometry, mathematical models for geometric analysis, considering variables such as surfaces and angles within designated spaces and times. While the lab studies pure mathematical questions, its researchers also apply computational geometry to practical problems, including those in biomedical engineering, such as designing nanoparticle drug delivery systems and algorithms to model growth of prostate tumors.

Working with HeartFlow, the Computational Mechanics Group plans to design a model that captures data from non-invasive computed tomography or CT scans and calculates a personalized risk profile of vulnerable plaques rupturing into the blood stream. Current techniques use invasive catheters to investigate suspected arteries for vulnerable plaques and collect data on temperature and acid levels to determine rupture risk.

The researchers plan to write a mathematical model that makes it possible to read CT scans, identify vulnerable plaques, and determine their composition without catheters. The model will use results of earlier CT scans and match the images to medical data, indicating lipid and calcium content in the fibrous caps found in arterial plaques that are prone to rupture.

Thomas Hughes, a professor of mathematics and engineering at UT-Austin and director of the group says in a university statement, “Recent breakthroughs have presented the opportunity to segment vulnerable plaques from CT scans and identify material constituents, including lipid and calcium content, and fibrous caps, from which patient-specific, computational models can be formulated and the biomechanical stress in the plaque calculated.”

HeartFlow Inc. is a designer of web and mobile software that analyzes CT images to detect and identify problems with blood flow through the heart leading to coronary artery disease. The company’s algorithms are based on data collected through its web and mobile apps, and stored in the cloud. HeartFlow returns a personalized graphic of a patient’s heart showing arteries at higher risk of forming clots.

In the collaboration with UT-Austin, HeartFlow is expected to take prototype software developed in the lab through clinical trials, regulatory approvals with FDA, and subsequent commercialization. Hughes has an equity stake in HeartFlow.

More from Science & Enterprise:

*     *     *

Comments are closed.