Update the tutorial to use Segmentations module instead of Model Maker module to create the brain surface model.

See the discussion here:

https://discourse.slicer.org/t/wrong-number-of-cells-in-model-maker-model/29282

List of possible categories:

• Grids (new category): This can include meshes, grids, point clouds; e.g. UnstructuredGrid

• Surface models (already exists in 3D Slicer)

• Physics (new category)

  • Kinematics
  • Property
  • BCs, Load, Interaction???
  • Solution

• Visualization (new category)

TODO:

• Decide which categories to use.
• Put the modules into one of the above categories.
• Update README.md with modules listed in categories.
• Update categories in the following locations:
  • Module documentation
  • README.md
  • Source code for each module

• Make Input Tumor Volume (fuzzy classified) optional
• Remove nrrd, mvloader.nrrd and unused imports. See 2a5c393
• Use slicer.util.arrayFromVolume instead of writing and reading temporary files. See 2a5c393
• Check if this averaging is needed (why not use just the value from one voxel?):
  

  SlicerCBM/BrainMaterialProperties/BrainMaterialProperties.py

  Lines 261 to 269 in c0d16f0

  	for i in range(iv-1, iv+1):
  	for j in range(jv-1, jv+1):
  	for k in range(kv-1, kv+1):
  	material_props[idx, 1] += YM_parenchima *(1-brain_ventricle_membership[i, j, k])+YM_ventricles * brain_ventricle_membership[i, j, k]
  	material_props[idx, 2] += PR_B * (1-brain_ventricle_membership[i, j, k])+PR_V * brain_ventricle_membership[i, j, k]
  	material_props[idx, 1]/= 8
  	material_props[idx, 2]/= 8

• Use millimeters instead of meters for length. Also, convert related units (Pa to MPa, density, etc.). See 2a5c393

List of modules to add for EEG...

Module names are in bold font.

Image processing:

• SkullStrippingUsingFreeSurfer (not yet implemented as a 3D Slicer module)
  • input: MRI???
  • output: brain mask??? MRI without skull???

Brain Model Generation:

• TetrahedralMeshGenerator to generate both 2D and 3D brain models
  • input: brain surface model
  • output: 3D brain model (tetrahedral)
• ComputationalGridGenerator supersedes TetrahedralMeshGenerator
• FuzzyClassification Fuzzy tissue classification to classify tissue classes (ventricles, parenchyma and tumour (if present))
  • input: MRI, brain mask
  • output: fuzzy classified brain tissues (ventricles, brain parenchyma, tumor) as nrrd files
• MTLEDSimulator to generate integration point file with respect to the input geometry.
• BrainMaterialProperties Assign material properties to brain tissue using fuzzy tissue classification
  • input: ventricle, parenchyma and tumour classes (nrrd)
  • output: material properties files with each integration points assigned a material property (txt)

Electrode Sheet Generator:

• ElectrodesToMarkups (rename as SegmentsToFiducials???) Identification of electrode locations on the intra-operative CT
  • input: segmented electrode image from CT (nrrd)
  • output: 3D fiducial points where there are electrodes
• FiducialToModelDistance Projecting electrode locations onto brain surface model
  • input: 3D fiducials (original electrode locations identified from CT
  • output: 3D fiducials on the undeformed brain surface
• FiducialsToSurface Create an electrode sheet model from a set of fiducial points (similar to SheetFromPoints below)
  • input: fiducials
  • output: 3D electrode sheet model
• SheetFromPoints Create an electrode sheet model from a set of fiducial points (similar to FiducialsToSurface above).
  • input:
  • output:
• BrainMeshSurfaceCellsSelection Select brain surface nodes when electrode sheet is not a single rectangular piece
  • input: brain surface model, projected electrode locations as 3D fiducials
  • output: selected brain triangles, loaded nodes on brain surface
• BrainSurfaceNeighbouringCellsSelection Select neighbouring cells to fill in the gaps and select almost all the brain cells under different electrode sheet stripes ???strips???
  • input: selected brain triangles with respect to projected electrodes on brain surface
  • output: loaded nodes with respect to electrode stripes
• SurfaceTriangulation to generate a uniform triangulation of the surface model (electrode sheet model or brain model or any other surface model)
  • input: brain surface model generated by model maker
  • output: uniformly triangulated brain surface model
• NodeSelector Loaded node selection
  • input: electrode sheet triangulated 2D model, brain 2D model
  • output: 3D fiducial points on brain surface under electrode sheet representing selected brain nodes
• Scattered Transform get the BSpline transform by using original electrode locations and projected electrode locations as input and apply the transform to get the deformed surface nodal positions for each of the loaded nodes selected on the brain surface.
• MeshNodesToFiducials (need to change the name) Compute displacement loading based on loaded nodes selected on undeformed brain surface and corresponding node locations on the deformed brain geometry.
  • input: loaded node locations (undeformed brain geometry, deformed brain geometry)
  • output: displacement loading (load file)

Remaining steps for constructing brain biomechanical model

• SkullGenerator to regenerate the skull surface using the triangulated brain surface model, defining contacts definition and combining skull and brain.
  • input: 2D brain model, 3D brain model
  • output: file containing skull nodes and elements

Solver for calculating brain deformations:

• MTLEDSimulator
  • input: tetrahedral mesh; load file, material properties file
  • output: deformed model (.vtk file)
• ExplicitSim MTLED C++ code which is a dependecy for MTLEDSimulator module (upload to separate bitbucket or github repository)

Image Warping:

• ExplicitSimWarp Create image warping transform using ExplicitSim MLS shape functions (Ben to upload modifications)
  • input: output from MTLED simulation "brain7_LPS.vtu"???; image???; mask???
  • output: transform file "warp.nrrd"
• ResampleDTIVolume (built-in Slicer module not part of SlicerCBM, in this list only as a reminder to include in documentation)
• DTISegmentation Segment DTI using fuzzy c-means clustering
• Fusion (rename as CreateSkullAndScalpSegments) Construct skull and scalp segments
  • input: MRI, skull stripped brain mask
  • output: skull and scalp layers
• Fusion2 (rename as ???) Combine segments into a single label map (image fusion of tissue classes)
  • input: segmented tissues (wm, gm, csf, scalp, skull, electrode sheet model (can add or not))
  • output:
• ElectricalConductivity Assign electrical conductivity to head model
  • input: complete segmentation labelmap; DTI (for Tuch's method)
  • output:
• MVoxMeshGenerator Generate voxel-based mesh using MVox
  • input: attributes image (labelmap); conductivity tensors image
  • output: MFEM mesh; MFEM tensors grid function file
• MVox which is a dependency for MVoxMeshGenerator module (requires MFEM), see separate repository at https://github.com/benzwick/mvox
• EEGSolver Solve the EEG forward problem
  • input: mesh; conductivity tensors; dipole position; electrode positions
  • output:
• Visualisation Visualize the deformation
• EEGVisualization Visualize the electric potential

Other modules (not used for EEG)

• TumorResectionAndBRainRemodelling

Miscellaneous

• Check spelling and format of all module names (use US spelling and UpperCamelCase)
• Add README.md for each module with short description (one sentence is enough) that is shown when browsing GitHub.

Images use millimeters, whereas some meshes and MTLED integration point coordinates use meters. It would be better to use the same units everywhere.

• Do not save files to the source code directory (create new nodes in Slicer instead and allow user to save them). See c7355d7

• Setting the tissue classes (parenchyma and ventricles) is not robust. Different images might cause the classes to flip.

  See:

  • SlicerCBM/FuzzyClassification/FuzzyClassification.py

    Line 331 in c0d16f0

    brain_ventricle_membership[brain_mask > 0] = u[0]

  • SlicerCBM/FuzzyClassification/FuzzyClassification.py

    Line 335 in c0d16f0

    brain_parenchima_membership[brain_mask > 0] = u[1]

  Fixed in 7b756f4

• Add output selectors for membership scalar volumes

https://github.com/Slicer/ExtensionsIndex/blob/master/.github/PULL_REQUEST_TEMPLATE.md#todo-list-for-submitting-a-new-extension

• Extension has a reasonable name (not too general, not too narrow, suggests what the extension is for)
• Repository name is Slicer+ExtensionName
• Repository is associated with 3d-slicer-extension GitHub topic so that it is listed here. To edit topics, click the settings icon in the right side of "About" section header and enter 3d-slicer-extension in "Topics" and click "Save changes". To learn more about topics, read https://help.github.com/en/articles/about-topics
• Extension description summarizes in 1-2 sentences what the extension is usable (should be understandable for non-experts)
• Any known related patents must be mentioned in the extension description.
• If source code license is more restrictive for users than BSD, MIT, Apache, or 3D Slicer license then the name of the used license must be mentioned in the extension description.
• Extension URL and revision (scmurl, scmrevision) is correct, consider using a branch name (main, master, release, ...) instead of a specific git has to avoid re-submitting pull request whenever the extension is updated
• Extension icon URL is correct (do not use the icon's webpage but the raw data download URL that you get from the download button - it should look something like this: https://raw.githubusercontent.com/user/repo/main/SomeIcon.png)
• Screenshot URLs (screenshoturls) are correct, contains at least one
• Homepage URL points to valid webpage containing the following:
  • Extension name
  • Short description: 1-2 sentences, which summarizes what the extension is usable for
  • At least one nice, informative image, that illustrates what the extension can do. It may be a screenshot.
  • Description of contained modules: at one sentence for each module
  • Tutorial: step-by-step description of at least the most typical use case, include a few screenshots, provide download links to sample input data set
  • Publication: link to publication and/or to PubMed reference (if available)
  • License: We suggest you use a permissive license that includes patent and contribution clauses. This will help protect developers and ensure the code remains freely available. We suggest you use the Slicer License or the Apache 2.0. Always mention in your README file the license you have chosen. If you choose a different license, explain why to the extension maintainers. Depending on the license we may not be able to host your work. Read here to learn more about licenses.
  • Content of submitted s4ext file is consistent with the top-level CMakeLists.txt file in the repository (description, URLs, dependencies, etc. are the same)

See also

• https://slicer.readthedocs.io/en/latest/developer_guide/extensions.html]
• https://slicer.readthedocs.io/en/latest/developer_guide/extensions.html#distribute-an-extension

Cannot import gmsh Python package in ComputationalGridGenerator or TetrahedralMeshGenerator

This is happening with the 5.2.2 stable version of Slicer:

Python console user input: import gmsh
Python console user input: gmsh.initialize()
error: [/opt/slicer/Slicer-5.2.2-linux-amd64/bin/SlicerApp-real] exit abnormally - Report the problem.

See this discussion: https://discourse.slicer.org/t/cannot-import-gmsh-python-package/29344

See also this discussion on Gmsh: https://gitlab.onelab.info/gmsh/gmsh/-/issues/2474

• Remove unused variables and UI elements: inputSelector, outputSelector, imageThresholdSliderWidget, enableScreenshotsFlagCheckBox. See bb71f6a in #66
• Remove hardcoded path for ExplicitSim command:
  

  SlicerCBM/MTLEDSimulator/MTLEDSimulator.py

  Line 321 in c0d16f0

  command_line = ["/home/saima/explicitsim/build-bucketsearch/ExplicitSimRun", dir_path+"/demo.ini"]

• Change the the dir_path working directory to not use the source code directory:
  

  SlicerCBM/MTLEDSimulator/MTLEDSimulator.py

  Line 253 in c0d16f0

  dir_path = os.path.dirname(os.path.realpath(__file__))

• Fix GUI to get ctkCollapsibleButtons working properly.

Currently files that start with two hyphens -- will have the first hyphen removed:

Currently, to load ComputationalGridGenerator use the path:

https://github.com/SlicerCBM/SlicerCBM/tree/master/ComputationalGridGenerator/ComputationalGridGenerator

but it should be:

https://github.com/SlicerCBM/SlicerCBM/tree/master/ComputationalGridGenerator

Also, delete these files:

• https://github.com/SlicerCBM/SlicerCBM/blob/master/ComputationalGridGenerator/CMakeLists.txt
• https://github.com/SlicerCBM/SlicerCBM/blob/master/ComputationalGridGenerator/ComputationalGridGenerator.png

• Rename Fusion as CreateSkullAndScalpSegments or CreateSkullAndScalpFromBrainSegment
• Fix interface (rename labels etc.)
• Allow loading labelmap to be used as the input volume (instead of a scalar image volume)
• Allow user to create or rename the segmentations node before it is created automatically
• Check if we need to output both the segments and the labelmap (otherwise delete the temporary nodes when they are no longer required)

This module will remove skull from pre-operative MRI to create brain mask used for brain surface model.

Reasons to use VTK meshes:

• MFEM can read and write VTK mesh files:
  https://docs.mfem.org/html/classmfem_1_1Mesh.html#a05a1b852899371aeebf84e09cb7d3eac
• Slicer has native support for VTK meshes.
• MVox (MFEM-based) mesh generator can create VTK mesh files.

Other mesh formats that are used in this project:

• Gmsh .geo and .msh for mesh generation
• .inp for ExplicitSim

Currently, there are a few files (e.g., Abaqus .inp, Gmsh .msh, ExplicitSim .txt and .ini) that are loaded and saved by specifying the filename in the module. It would be better to store this data in MRML nodes so it does not need to be saved and loaded when switching between modules. The vtkMRMLTextNode might be suitable for some data, but it currently works only with file extensions txt, xml or json. See also this discussion:
https://discourse.slicer.org/t/loading-vtkmrmltextnode-with-arbitrary-file-name-extension/29747

Some Python packages are installed and not used.

Check that Python package installation is correct and done in the correct location in the code in all modules.

See here for how to install packages within a module: https://slicer.readthedocs.io/en/latest/developer_guide/script_repository.html#install-a-python-package

• Do not save files to the source code directory (create new nodes in Slicer instead and allow user to save them).

Create a new module that has only the "Displacement Loading" section of the existing MeshNodesToFiducials module:

3D Slicer's built-in Surface Toolbox now does surface remeshing with PyACVD:

https://github.com/Slicer/SlicerSurfaceToolbox/blob/3b556d10307c5c9ee9fd8da1ea873ddc1d48acf3/SurfaceToolbox/SurfaceToolbox.py#L396

The results appear to be the same with both modules.

• Check that results are the same when using Surface Toolbox
• Update tutorial
• Delete Surface Triangulation module

• Remove unused gmsh_api:
  

  SlicerCBM/ComputationalGridGenerator/ComputationalGridGenerator/ComputationalGridGenerator.py

  Lines 7 to 8 in bddc6b1

  	import gmsh_api
  	import gmsh_api.gmsh as gmsh

• Mesh 3D not 2D by uncommenting some code as described in the comment below:
  

  SlicerCBM/ComputationalGridGenerator/ComputationalGridGenerator/ComputationalGridGenerator.py

  Line 519 in bddc6b1

  gmsh.model.mesh.generate(2)

• Change "Input Folder" to "Output Folder" and move to "Outputs" in GUI.

• Remove unused invert threshold option.

• Check if "Wrap Solidify" effect has been installed before running this code, and prompt user to install SurfaceWrapSolidify extension:

      #use shrink wrapping to wrap the bone and get the full surface
      segmentEditorWidget.setActiveEffectByName("Wrap Solidify")
      effect = segmentEditorWidget.activeEffect()
      segmentEditorWidget.setCurrentSegmentID(segmentationNode.GetSegmentation().GetSegmentIdBySegmentName("cran"))
      effect.setParameter("region", "outerSurface")
  

  to avoid this error:

  Traceback (most recent call last):
    File "/home/ben/projects/SlicerCBM/SlicerCBM/ComputationalGridGenerator/ComputationalGridGenerator/ComputationalGridGenerator.py", line 273, in onApplyButton
      self.logic.process(self.ui.inputSelector.currentNode(), self.ui.fold_save.currentPath)
    File "/home/ben/projects/SlicerCBM/SlicerCBM/ComputationalGridGenerator/ComputationalGridGenerator/ComputationalGridGenerator.py", line 365, in process
      effect.setParameter("region", "outerSurface")
  AttributeError: 'NoneType' object has no attribute 'setParameter'
  

• Add SurfaceWrapSolidify extension as a dependency CMakeLists.txt Remove SurfaceWrapSolidify effect instead. It is not needed if thresholding is done correctly. See #59

• Fix error: "Failed to compute results: File (/home/ben/scratch/ComputationalGridGenerator/Segmentation_1_cran.stl) not found"

Module 'Fusion2' fail to instantiate due to missing python library 'pynrrd':

Implement try/except to fix.

Module 'FiducialsToSurface' fail to instantiate due to missing indentation in try/except block:

Indent line to fix.

Use ComputationalGridGenerator instead of TetrahedralMeshGenerator

Run Gmsh using external process instead of using Python interface to Gmsh to resolve #41 and to address licensing concerns around Gmsh and GPL inside 3D Slicer.

Use launchConsoleProcess to run external commands instead of using subprocess directly in, e.g., MVoxMeshGenerator

See:

• https://discourse.slicer.org/t/cannot-import-gmsh-python-package/29344/8
• https://github.com/Slicer/Slicer/blob/02a9ed37282a89c1e39d478080df0763a65f9889/Base/Python/slicer/util.py#L3445

See also this: https://slicer.readthedocs.io/en/latest/developer_guide/script_repository.html#launch-external-process-in-startup-environment

• Use readthedocs + sphinx + markdown for documentation and tutorials
• Use SlicerCBM wiki for draft tutorial