Research‎ > ‎

Functional Imaging + Biomechanics

Our colleagues in Edinburgh have shown that by using PET/CT to image the aortic valve and vasculature, they can indicate both high risk patients as well as those that may have CVD-related incidents in the future. The image on the right shows fused PET/CT images of the calcified aortic valve (Fig-A) and coronary artery (Fig-B), PET (Fig-C) and PET/CT images of the spine (Fig-D) and aortic arch (Fig-E and F) (Dweck et al. Eur Heart J 2013).  

Imaging Inflammation with PET/CT

PET/CT relies on the use of radiotracers which highlight certain cellular activities. A common tracer is 18F-Fluorodexoxyglucose (18F-FDG) which is a glucose analogue. The tracer is taken up by cells that require high levels of glucose, such as macrophages, and as macrophages are pivotal to the inflammatory process; 18F-FDG is a widely used as a surrogate for inflammation. Importantly, vascular inflammation is known to be a central player in CVD.

Imaging Active Calcification
with PET/CT

Another useful radiotracer is 18F-Sodium fluoride (18F-NaF) which detects new areas of vascular calcification and calcium remodelling. As with inflammation, calcification is also an established marker of CVD, and so the ability to non-invasively image and measure this activity holds real promise. 18F-NaF uptake has been proven useful in aortic valve disease and coronary arterial disease.

Imaging Inflammation with USPIO-MRI

Besides PET/CT, MRI together with a contrast agent of ultrasmall superparamagnetic particles of iron oxide (USPIO) could be a turning point in AAA management. USPIO-MRI has shown the ability to detect rapidly-expanding AAAs and we are now testing the method in a clinical trial – the MA3RS Trial.

We can determine the relationship between biological processes and biomechanics. With the finite element method we can compute the wall stresses and strains in the vascular tissue, whereas with computational fluid dynamics (CFD) we can calculate the haemodynamics and wall shear stress (WSS).

This image shows some of our work correlating wall stresses in AAA with inflammation measured using USPIO-MRI. Here we have compared the 2D MRI data with the corresponding wall stress. The crosshair on the 3D model shows the location of the data. There is some overlap between inflammation and wall stress in regions of thin ILT, however, using the current computational methods to model the ILT (which we think can improved and are working on here), we cannot correlate pockets of inflammation and wall stress in areas of thick ILT.

Generally, inflammation measured with USPIO-MRI appears to co-locate in areas of elevated wall stress. We are currently investigating this as part of a much larger project.

We are also using this approach to model to the non-aneurysmal thoracic aorta in cases who have CVD. This particular image shows the distributions of Gaussian surface cuvature, von Mises wall stress, active calcification (18F-NaF) and vascular inflammation (18F-FDG).