• Specifications
• GeometryGeneration

Target Outcome

This specification defines a surface representation a tissue from a segmented image.

Protocols

Required Infrastructure

• Slicer is a free, open source software package for visualization and image analysis (BSD-style open source license, see http://www.slicer.org).

Input

Binary image of tissue volume aligned with original MRI coordinate system. For more details, see Image Segmentation.

Geometry Generation Procedures

Generating the geometry happens in two parts.

First the surface is directly extract from the label.

Slicer Procedures

1. If necessary, in Slicer, load image as data. It is important to check Show Options and then also check LabelMap.

2. From the modules drop down select Surface Models > Model Maker or simply search for Model Maker.

3. Within the IO section:

   1. Select the desired Label map from the Input Volume drop down. If the desired label is missing try reloading the data as in Step 1.

   2. Select create new ModelHierarchy( as...) or select an existing Hierarchy. Please note that all model can be generated under the same Hierarchy.

4. Within the Create Model section:

   1. Specify the desired Model Name for the selected Label Map.

5. In the Parameters section:

   1. Optionally select the Label(s) within the image to be added to the model.

   2. Set Smooth to zero.

   3. Set Decimate to zero.

6. Click Apply to generate the stl.

Next the raw stl from Slicer is processed in MeshLab

Mesh Lab Procedures

1. Load Stl generated in prior step.
2. Perform Surface Reconstruction. The goal is to remesh the surface with triangles that are approximately similar in size at the desired mesh density. There are several that can be methods to used. These are listed in order of usefulness.

   1. MeshLab's Iso Parameterization does an excellent job of remeshing, but it is very fickle and often won't work on a mesh. Try it first. Detailed instructions on Iso Parameterizing can be found here.

   2. Poisson Reconstruction does not produce the even triangle size that isoparameterization does, but it is more robust and the resulting surface can then be isoparameterized itself. Sometimes this reconstruction produces undesirable results. Detailed instructions on Poisson Reconsctruction can be found here.

   3. VCG Reconstruction this also does not produce even triangles, but like the poisson method can then be used as input for isoparameterization. VCG seems to be quite robust, but the result doesn't always seem to isoparameterize.Detailed instructions on VCG Reconstruction can be found here.

3. It is a good practice, but not always necessary to apply a smoothing alogrithm to the reconstructed surface mesh. From the menu Filters > Smoothing, Faring and Deformation > Taubin Smoothing using the default settings.

Output

Surface representation of tissue of interest in stl format in MRI coordinate system.

This Stl will be used in Geometry Morphing if necessary or Mesh Generation.