| Popis: |
A detailed and experimentally verified methodology is outlined on how to properly process a long tibia bone starting from computed tomography (CT) scans to finite element modeling (FEM). For pre-processing of the bone, CT scan in the form of Digital Imaging and Communications in Medicine (DICOM) files are segmented using Mimics. Next comes assigning gray value Hounsfield Units (HU) of the bone constituents into their cortical and cancellous regions. To have the FEM model arrives at the same mass of that measured experimentally, it was found that cut-off density, cut-off HU, and the utilized number of sub-materials must be considered as varying parameters. The values of these parameters had to be adjusted to properly demarcate cancellous regions from those of the cortical resulting in heterogeneous medium. Next, prior to generating the FEM mesh from the generated 3D model, volume and surface meshes had to be produced. In order to validate the methodology, the modal frequencies of a long tibia bone were experimentally measured. The FEM values of the properly processed CT scans compared favorably with those found experimentally. |
