Coreform Cubit provided an intuitive platform for meshing the model geometry, especially in complex biological tissues with fine anatomical detail. Without Cubit’s capabilities, many important features would be lost in coarser, simplified anatomical models.
Nicholas Glover, Ph.D. Student, University at Buffalo
University at Buffalo case study (Nicholas Glover project)
Background
Nicholas Glover is a Ph.D. student in the Systems Biomedicine and Pharmaceutics Laboratory led by Dr. Ashlee N. Ford Versypt, where he works on the mechanisms of chemical and biomechanical damage in kidney tissues during diabetic kidney disease. During his masters and current PhD project in this research group, Nicholas studied the biomechanical properties of the glomerulus, the major anatomical structure of the kidney responsible for filtering wastes from the blood, which is the primary target for diabetic tissue damage. He sought to establish a connection between morphological changes in the anatomical structure as a result of disease and the resulting changes in the transport of filtrate as it moves through the glomerular filtration barrier to better understand the progression of diabetic kidney damage. To simulate these processes in a computational model, ANSYS fluent and FEbio have been used as dedicated solvers. Coreform Cubit provided an intuitive platform for meshing the geometry for the model, especially in the more complex biological tissues containing fine anatomical details. Without the capabilities of Coreform Cubit, many important details would be lost in rendering coarser, more simplified models of the anatomy. Structured and unstructured hex meshing, the availability of journal files paired with cubit command language, and the export options have been indispensable tools for his project.
The significance of the project is in connecting several processes that have been shown individually to contribute to injury in kidneys due to diabetes into a sophisticated computer simulation. This computational tool will aid in understanding how the processes interact and how diabetic kidney damage begins and changes over time. In the long-term, results from this project will help to predict the impacts of many competing factors on kidney health during diabetes management. The results of these simulations can give an estimation of the filtering capacity of each layer in a quantitative way. This information can be used to guide effective treatment plans for patients as well as predict and monitor disease states in a non-invasive manner.
Figure 1: Geometry of the kidney’s glomerular filtration barrier and the boundary conditions associated with each surface to be meshed in Coreform Cubit. The vertically stacked layers of the tissue represented here each have different porous media material properties.
THE PROBLEM
As the function of a kidney to filter wastes from the blood is dependent on the various physical filtering layers, a 3D representative geometry of the filtering structure is first needed to simulate the filtration process. To build such a geometry of the system, biological images from kidney glomeruli were used to construct multiple disease state geometries using structure parameters derived from the images. To have an accurate CFD model for the layered filtration barrier, meshing software that provides a high-quality mesh was needed. Some of the layers are the gaps between biological cells of different sizes and shapes, and others are more amorphous porous materials. So it is important to also have the ability to make element sets that correspond to the various biological materials present in the model for the different layers. Slight variations to the ultrastructure of the geometry are made to analyze hydrodynamic properties of individual structures in the filtration barrier. The anatomical variations on the filtration barrier were constructed using parametric CAD models, which lead to flexible CAD modeling that requires automated meshing. A meshing environment where CAD parameters are integrated with Coreform cubit will streamline the analysis of each layer of the barrier.
The limitations of other available software were the lack of meshing tools to accommodate the complex glomerulus structure and non-intuitive user interfaces. In the realm of CFD modeling, Coreform Cubit provided Nicholas with an excellent platform via the intuitive GUI paired with Cubit Journal files to mesh the desired complex biological tissue (Figure 1).
THE SOLUTION
Previously published anatomical measurements were used as the geometry for the solute transport study through the glomerular filtration barrier. These measurements were used to create a domain that filtrate (water, small molecules, and– in disease conditions– proteins) passes through. The domain consists of layers of fluid channels and porous media. Before running the multiphysics problem in the ANSYS Fluent software, it was necessary to create an appropriate mesh for the problem. Coreform Cubit was used to create a high-quality mesh (Figure 2) for the application, which captures the fine geometry of the layers and contains important information at the boundaries of the domains in the model, indicating contact between the filtrate and cell surfaces that form the channels of the glomerular filtration barrier.
Figure 2: High-quality mesh for the kidney’s glomerular filtration barrier produced by Coreform Cubit. The meshing captured fine features of the narrow channels, allowed biasing near the domain boundaries, and had coarser resolution in the middle of the large fluid domains.
Adapting previously published theoretical formulations for multiphasic transport in the kidney, physical properties were assigned to the layers, and interface conditions were selected that best simulate physiological conditions in the kidney. Within the model there are biphasic and fluid domains, a normal fluid velocity calculated from a single nephron glomerular filtration rate, and boundary conditions that dictate fluid behaviors on cell-filtrate interactions. As the research in the kidney project progresses, and the team implements more variations in biological structures, the use of the Coreform Cubit unique “sculpt” function will allow for easy-to-use and efficient hexahedral meshing options for the normal and diseased geometries.
In ANSYS, the team leveraged the ANSYS journal files to run CFD simulations on the Coreform Cubit generated mesh. Simulations were run to study hydraulic conductivity and wall shear stress (WSS) in normal and diseased states. The simulations yield the distributions of fluid velocity and shearing forces along the surface of the slit diaphragm, an essential filtering layer of the kidney (Figures 3 and 4).
Figure 3: Fluid velocity along the surface of the slit diaphragm (the inset region of Figure 2).
Figure 4: Wall shear stress along the surface of the slit diaphragm (the inset region of Figure 2).
CONCLUSION
Through support from the NSF CAREER Award, National Institute of General Medical Science (NIGMS) Award (T32 GM144920) to University at Buffalo (N.O.G. Trainee), NIGMS R35GM133763, and the University at Buffalo, Dr. Ford Versypt’s lab is developing a multiscale model of a virtual kidney during the onset and progression of diabetic kidney disease. Multiscale modeling considers chemical, biological, and physical phenomena that depend on multiple spatial and temporal scales. Before Nicholas’ part of the project, the focus was primarily chemical and biological. Nicholas’ project addressed the physical barriers for filtration and provides an important foundation for starting to couple the dynamic chemical processes that influence the morphological and material properties of the kidney anatomical layers to the biomechanics of those layers, which in turn affect filtration. Now with the Coreform Cubit meshes and the simulation workflow using these meshes, the team can simulate damage to cells due to wall shear stress, which dictates the detachment forces of a primary filtering layer, and effects on protein transport, which is the quantity that healthcare professionals use to track disease progression. This will enable the team to develop the next generation of software tools for predicting disease progression in advance of clinical detection of the protein leakage associated with permanent damage.
