Image-based blood flow modeling to lessen invasive treatment
One of the many pluses of the virtual world is the ability to model before taking an irreversible action. Combining that modeling with medical imaging is a powerful combination of both worlds: the predictive powers of a virtual tool with the real-world images of the subject.
Take, for example, brain aneurysms — the local dilations of arteries in the brain that are estimated to affect 2 to 5 percent of the U.S. adult population. While the exact mechanisms of aneurysm progression are not fully understood, there is a consensus that blood flow plays a role in their initiation, progression and rupture.
Dr. Vitaliy Rayz’s Cardiovascular Flow Modeling Laboratory is combining computational fluid dynamics (CFD) modeling and medical imaging methods for patient-specific blood flow analysis in brain aneurysms to evaluate the benefit of using flow-related factors for risk analysis and to see if these flow factors can be calculated reliably from magnetic resonance imaging (MRI) data.
The ultimate goal of this research is to avoid invasive treatment in cases in which the treatment risk may exceed the risk of aneurysm growth and rupture. The vast majority of unruptured intracranial aneurysms (UIAs) are treated, even though most remain stable for years. That’s because of the grave consequences of an aneurysm rupture, which may result in a hemorrhage or stroke, with high rates of death and disability for survivors.
Current clinical guidelines for risk assessment are based on aneurysm location, size and shape, as well as a patient’s medical history, and do not account for flow-related factors. Rayz’s team is developing a methodology for improving aneurysm risk assessment by combining the risk factors derived from magnetic resonance imaging of blood flow (4D flow MRI) with currently accepted risk factors.
Identifying at-risk aneurysms before they grow will allow patients who have them to receive earlier treatment and avoid exposure to the risk of rupture. Furthermore, it will allow for less aggressive management of patients who are found to be at low risk for rupture.
Rayz’s team conducts CFD simulations based on MRI data. The image data are used to obtain 3D geometry of the aneurysmal arteries and prescribe patient-specific inflow and outflow conditions for the model. The blood flow equations are solved numerically to obtain the distribution of relevant variables — valuable information not available from imaging alone.
Rayz’s team collaborates closely with the team of Dr. Pavlos Vlachos, professor of mechanical engineering at Purdue University, to obtain high-resolution flow measurements in patient-specific aneurysm replicas generated with 3D printing. The team also conducts flow measurements on MRI scanners at the Purdue MRI Facility.
Rayz’s research is funded by the National Heart, Lung, and Blood Institute (NHLBI) and the National Institute of Neurological Disorders and Stroke (NINDS), both institutes of the U.S. National Institutes of Health (NIH). His studies are conducted in collaboration with neurosurgeons, radiologists and imaging scientists at Northwestern University, Barrow Neurological Institute, and the University of California San Francisco. The teams of clinicians and scientists at these medical institutions are conducting imaging studies of UIA patients, and then provide their MRI data for flow modeling and analysis at Rayz’s lab at Purdue.
The use of a data-assimilation approach, in which predictive computational models are informed by medical imaging, will continue to grow in power and importance, facilitating personalized, precision medicine.
Rayz believes that applying novel algorithms to enhance MRI flow measurements will enable reliable calculation of predictive blood flow metrics. His goal is to develop a modeling tool for predicting the risk of aneurysm growth from an analysis of MRI data and medical records so patients can get the optimal and least invasive treatment.
Vitaliy L. Rayz, PhD
Associate Professor, Weldon School of Biomedical Engineering
Associate Professor, Mechanical Engineering (by courtesy)
College of Engineering, Purdue University
