Hi Nikhil,
we developed DICOM2FEM [1] to make meshes from CT scans. Tissue segmentation is based on the graph-cuts method and the resulting volumetric grid (voxel data) is transformed into a finite element mesh using the marching cubes algorithm and smoothing procedures.
Regards, Vladimir
[1] http://sfepy.org/dicom2fem/
On 2016-08-22 11:11, Robert Cimrman wrote:
Hi Nikhil,
I do not know much about this, but Vladimir also uses CT scans for making meshes - I will ping him to see if he has some tips.
r.
On 08/22/2016 09:49 AM, Nikhil Vaidya wrote:
Hi Robert,
Thanks for your reply! I went through the mesh file in example you pointed to. I now understand how group definition works.
I do have information about the material of each cell. I now understand that the mesh generation tools should assign group numbers to elements in the mesh file. My next question would be about a suitable meshing tool. The CT scan geometry data that I have is in the form of stl files.
Could you suggest some tool (preferably open source) that can take these stl files and generate a conformal mesh with group numbers assigned to cells?
I should mention here the following points: the avs-ucd format and the hermes3D tool.
I tried to look for medit online, but could find only dead links to it.
The tetgen manual shows the possibility of region definition, but it is only if the input geometries are in .poly or .smesh formats. I am not familiar with these formats and could not find a way to convert stl files to these formats.
I also could not find too much information about generating meshes in
Any help would be greatly appreciated! :) Best regards, Nikhil
On Sunday, August 21, 2016 at 10:10:13 PM UTC+2, Robert Cimrman wrote:
Hi Nikhil,
On 08/15/2016 01:20 PM, Nikhil Vaidya wrote:
Hello!
I am trying to solve a problem using sfepy. It is a multiphysics problem with weakly coupled convection and conduction. There are three regions in my domain which need to be separately defined. I went through the user guide and saw the format for defining regions in the domain. The regions in my geometry are not simple enough to be described by mathematical conditions. I usually use Starccm+ for my multiphysics simulations. There I define the regions using the GUI. One region in my domain is a real blood vessel observed in a CT scan. Can anyone tell me how I should define my regions in this case?
The mesh comes from a CT scan, right? Do you have information about the material of each cell (element)?
I have seen in the user guide, that regions can be defined using "groups" of cells and vertices. I would like to know how these groups are defined.
The groups (aka material ids) are defined using integer numbers assigned to each cell (element) in a mesh. The mesh has to contain those numbers for sfepy to use them - you probably need to define them at the time of the mesh creation
- during the CT scan segmentation and meshing.
Check, for example, the example [1] and the mesh used there to see how the groups are defined in the mesh file and in the example.
r.
[1] http://sfepy.org/doc-devel/examples/multi_physics/thermal_electric.html