Intracranial CFD
A tool collection to perform vessel segmentation based on 3DRA and post contrast CBCT intracranial data. The repository also provides an OpenFOAM computational fluid simulation pipeline.
Workflow
- Batch conversion of DICOM to NIFTI format
- Image registration (external process, recommend to use 3D slicer if manual initialization is needed)
- Located the defected point
- Crop the image to smaller VOI
- Vessel segmentation with multiple Otsu thresholding/ CNN methods
- Refinement on the segmented data (optional)
- Extract surface as VTK file
- Vessel centerline extraction
- Crop the defected vessel (with Paraview or other 3D mesh processing software)
- Using fiducial point to locate the defected point
- Using
auto_vessel_seg.py
to automatically crop the defected vessel region
- Extend the inlet and outlet region
- Remeshing and volume meshing
- CFD simulation
- Probe CFD result along centerline
- Multiplanar reconstruction (MPR) with input image and the centerline to straighten the image data
Usage