This is a 3D Slicer extension for creating volumetric meshes from segmentation using Cleaver2 or TetGen.
Cleaver2 mesher is freely usable, without any restrictions. TetGen mesher is only free for private, research, and educational use (see license for details).
- Download and install a latest stable version of 3D Slicer (https://download.slicer.org).
- Start 3D Slicer application, open the Extension Manager (menu: View / Extension manager)
- Install SegmentMesher extension.
- Start 3D Slicer
- Load a volume: switch to "Sample Data" module and load MRHead image
- Switch to "Segment Editor" module
- Add a new segment (it will contain the entire head)
- Fill segment by thresholding: click "Threshold" effect set 30 as lower threshold, click "Apply"
- Smooth segment: click "Smoothing" effect, set kernel size to 6mm, click "Apply"
- Add a new segment (it will contain a spherical lesion)
- Paint a sphere in the brain (simulating a lesion): click "Paint" effect, enable "Sphere brush", set "Diameter" to 8%, and click in the yellow slice view
- Switch to "Segment Mesher" module (in Segmentation category)
- Select "Create new Model" for Output model (this will contain the generated volumetric mesh)
- Click Apply button and wait a about a minute
- Inspect results: open "Display" section, enable "Yellow slice clipping", move slider at the top of yellor slice view to move the clipping plane; enable "Keep only whole cells when clipping" to see shape of mesh elements
- Create more accurate mesh: open "Advanced" section, set scale parameter to 0.5, click "Apply", and wait a couple of minutes
- Open "Display" section to enable clipping with slices.
- Go to "Segmentations" module to hide current segmentation.
- Switch to "Models" module to adjust visualization parameters.
- To save Output model select in menu: File / Save.
Cleaver parameters are described at https://sciinstitute.github.io/cleaver.pages/manual.html. To make the output mesh elements smaller: decrease value of --feature_scaling. To make the output mesh preserve small details (at the cost of more computation time and memory usage): increase --sampling-rate (up to 1.0).
Input data:
-i [ --input_files ] arg material field paths or segmentation path
This argument is set automatically by SlicerSegmentMesher module.
-B [ --blend_sigma ] arg blending function sigma for input(s) to
remove alias artifacts.
Too low value will not remove staircase artifacts.
Too high value may shrink structures and remove relevant details.
Default: 1.0.
Output data:
-f [ --output_format ] arg output mesh format (tetgen [default],
scirun, matlab, vtkUSG, vtkPoly, ply
[surface mesh only])
This argument is set automatically by SlicerSegmentMesher module to vtkUSG.
-n [ --output_name ] arg output mesh name (default 'output')
This argument is set automatically by SlicerSegmentMesher module.
-o [ --output_path ] arg output path prefix
This argument is set automatically by SlicerSegmentMesher module.
Meshing mode (element size control):
-m [ --element_sizing_method ] arg background mesh mode (adaptive [default],
constant)
For constant mode:
-a [ --alpha ] arg initial alpha value, default: 0.4
-s [ --alpha_short ] arg alpha short value for constant element
sizing method, default: 0.203
-l [ --alpha_long ] arg alpha long value for constant element
sizing method, default: 0.357
For adaptive mode:
-F [ --feature_scaling ] arg feature size scaling (higher values make a
coarser mesh), default: 1.0.
Meaningful range is about 0.2 to 5.0.
Lower value makes the output mesh finer,
higher value makes the output mesh coarser and meshing faster.
-L [ --lipschitz ] arg maximum rate of change of element size (1
is uniform), default: 0.2
It specifies how quickly the sizing field may grow away from size-limiting
features (like corners or curved interfaces).
-R [ --sampling_rate ] arg volume sampling rate (lower values make a
coarser mesh), default: 1.0 (full sampling)
Meaningful range is 0.1 to 1.0.
Lower value makes meshing faster, higher value
preserves fine details.
Advanced:
-b [ --background_mesh ] arg input background mesh
-I [ --indicator_functions ] the input files are indicator functions (boundary is defined as isosurface
where image value = 0)
-z [ --sizing_field ] arg sizing field path (use precomputed sizing field for adaptive mode)
-w [ --write_background_mesh ] write background mesh
--simple use simple interface approximation
-j [ --fix_tet_windup ] ensure positive Jacobians with proper vertex wind-up
(prevents inside-out tetrahedra in the output mesh)
This flag is specified by SlicerSegmentMesher module, no need to specify it as additional option.
-e [ --strip_exterior ] strip exterior tetrahedra (remove temporary elements that are added to make the volume cubic)
This flag is specified by SlicerSegmentMesher module, no need to specify it as additional option.
Other:
-h [ --help ] display help message
-r [ --record ] arg record operations on tets from input file
-t [ --strict ] warnings become errors
-v [ --verbose ] enable verbose output
This flag is specified by SlicerSegmentMesher module (based on Verbose option).
-V [ --version ] display version information
TetGen parameters are described at http://wias-berlin.de/software/tetgen/1.5/doc/manual/manual005.html#sec%3Acmdline
meshNode = getNode('Model')
mesh = meshNode.GetMesh()
cellData = mesh.GetCellData()
labelsRange = cellData.GetArray("labels").GetRange()
for labelValue in range(int(labelsRange[0]), int(labelsRange[1]+1)):
threshold = vtk.vtkThreshold()
threshold.SetInputData(mesh)
threshold.SetInputArrayToProcess(0, 0, 0, vtk.vtkDataObject.FIELD_ASSOCIATION_CELLS, "labels")
threshold.ThresholdBetween(labelValue, labelValue)
threshold.Update()
if threshold.GetOutput().GetNumberOfPoints() > 0:
modelNode = slicer.mrmlScene.AddNewNodeByClass("vtkMRMLModelNode", "{0}_{1}".format(meshNode.GetName(), labelValue))
modelNode.SetAndObserveMesh(threshold.GetOutput())
modelNode.CreateDefaultDisplayNodes()Cleaver is an Open Source software project that is principally funded through the SCI Institute's NIH/NIGMS CIBC Center. Please use the following acknowledgment and send references to any publications, presentations, or successful funding applications that make use of NIH/NIGMS CIBC software or data sets to SCI: "This project was supported by the National Institute of General Medical Sciences of the National Institutes of Health under grant number P41 GM103545-18."
TetGen citation: Si, Hang (2015). "TetGen, a Delaunay-based Tetrahedral Mesh Generator". ACM Transactions on Mathematical Software. 41 (2): 11:1-11:36. doi:10.1145/2629697