A CFD-Based Approach to Validating Flow in a Prototype Dynamic Perfusion Phantom for Dynamic Contrast Enhanced (DCE) Imaging
A Thomas*, S Hollister, J Balter, University of Michigan, Ann Arbor, MISU-C-218-5 Sunday 1:30:00 PM - 2:15:00 PM Room: 218
Purpose: A dynamic phantom for quality assurance of dynamic contrast enhanced (DCE) imaging has been designed to achieve perfusion (in terms of flow rates) comparable to capillary exchange that occurs in human tissue. The validation of this physical phantom is critical to establishing its usefulness as a quality assurance tool. This study seeks to evaluate the use of computational fluid dynamics (CFD) to validate perfusion properties within the prototyped dynamic DCE phantom.
Methods: The phantom was evaluated in two parts, the vascular network, followed by the porous media compartments. CFD simulation software (Fluent) approximates fluid motion assuming the polymer phantom to be a solid, rigid object, receiving constant laminar flow. The fluid simulated in this model was a blood analog with a non-Newtonian fluid viscosity fitted to the Carreau model. The velocity across the system was visualized and the flow rates across the porous compartments were estimated.
Results: The flow across the perfusion phantom was qualitatively and quantitatively evaluated. The flow rates for highly permeable compartments demonstrating high rates of perfusion were quantified to estimate the differential perfusion and flow rate. Less permeable compartments showed decreased velocity profiles and flow rates across the scaffold, in some cases resulting in low flow or backflow, which was also observed for less porous compartments under experimental conditions. The compilations of simulations offer a range of flow rates representative of normal and diseased tissue (in the liver).
Conclusions: CFD is an effective tool for validation of design of a DCE phantom. The results of CFD analysis have been applied to both demonstrate the potential for the phantom to simulate a wide range of realistic perfusions characteristics, as well as to establish limits in the phantom design that would otherwise result in non-physiological flow and perfusion values.
Sponsored by NIHP01CA59827 and F31 EB012436-02.