A Completely Open-Source Finite Element Modeling Chain for Tubular Tissue-Engineered Constructs
Adrienne M. Madison, Mark A. Haidekker
Abstract
Adrienne M. Madison, Mark A. Haidekker
Abstract
Finite-element modeling (FEM), well-established to predict the mechanical behavior of mechanical systems, enjoys growing popularity in the study of the biomechanical behavior of biological tissues. Contrary to mechanical systems, which can often be described analytically or with geometric primitives, biomedical objects require discretization of their often irregular shape. In noninvasive studies, imaging methods are used to obtain the shape. Frequently, a relationship between image intensity and biomechanical properties is assumed. Commercial FEM toolchains exist, but we failed to obtain a satisfactory discretization from simple phantom images. Driven by the application to image and characterize tissue-engineered blood vessels noninvasively, we sought to establish a completely open-source FEM toolchain. The open-source feature gives users the ability to modify and extend the code, and thus offers additional flexibility over commercial systems. We demonstrate that the combination of a custom module to discretize the geometry (meshing) combined with the open-source FEM solver Tochnog and free visualization software (namely, Paraview and OpenDX) completes an open-source FEM toolchain. We demonstrate its ability to analyze tubular phantoms modeled after tissue-engineered blood vessels and compare results to a commercial toolchain. We conclude that a fully open-source toolchain is feasible, but the critical element is the meshing module.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Finite-element modeling (FEM), well-established to predict the mechanical behavior of mechanical systems, enjoys growing popularity in the study of the biomechanical behavior of biological tissues. Contrary to mechanical systems, which can often be described analytically or with geometric primitives, biomedical objects require discretization of their often irregular shape. In noninvasive studies, imaging methods are used to obtain the shape. Frequently, a relationship between image intensity and biomechanical properties is assumed. Commercial FEM toolchains exist, but we failed to obtain a satisfactory discretization from simple phantom images. Driven by the application to image and characterize tissue-engineered blood vessels noninvasively, we sought to establish a completely open-source FEM toolchain. The open-source feature gives users the ability to modify and extend the code, and thus offers additional flexibility over commercial systems. We demonstrate that the combination of a custom module to discretize the geometry (meshing) combined with the open-source FEM solver Tochnog and free visualization software (namely, Paraview and OpenDX) completes an open-source FEM toolchain. We demonstrate its ability to analyze tubular phantoms modeled after tissue-engineered blood vessels and compare results to a commercial toolchain. We conclude that a fully open-source toolchain is feasible, but the critical element is the meshing module.
Key concepts: Toolchain, Finite element method, Computer science, Discretization, Solver, Hexahedron, Open source, Computational science