sMRI: Regional Tessellation and Surface Smoothing

Introduction

This script, tessellation_tutorial.py, demonstrates the use of create_tessellation_flow from niflow.nipype1.workflows.smri.freesurfer, and it can be run with:

python tessellation_tutorial.py

This example requires that the user has Freesurfer installed, and that the Freesurfer directory for ‘fsaverage’ is present.

See also

ConnectomeViewer

The Connectome Viewer connects Multi-Modal Multi-Scale Neuroimaging and Network Datasets For Analysis and Visualization in Python.

http://www.geuz.org/gmsh/

Gmsh: a three-dimensional finite element mesh generator with built-in pre- and post-processing facilities

http://www.blender.org/

Blender is the free open source 3D content creation suite, available for all major operating systems under the GNU General Public License.

Warning

This workflow will take several hours to finish entirely, since smoothing the larger cortical surfaces is very time consuming.

Packages and Data Setup

Import the necessary modules and workflow from nipype.

import nipype.pipeline.engine as pe  # pypeline engine
import nipype.interfaces.cmtk as cmtk
import nipype.interfaces.io as nio  # Data i/o
import os
import os.path as op
from niflow.nipype1.workflows.smri.freesurfer import create_tessellation_flow

tories

Set the default directory and lookup table (LUT) paths

fs_dir = os.environ['FREESURFER_HOME']
lookup_file = op.join(fs_dir, 'FreeSurferColorLUT.txt')
subjects_dir = op.join(fs_dir, 'subjects/')
output_dir = './tessellate_tutorial'

Inputs

Create the tessellation workflow and set inputs Here we will choose Gifti (gii) as the output format, because we want to able to view the surface in ConnectomeViewer.

In you intend to view the meshes in gmsh or Blender, you should change the workflow creation to use stereolithographic (stl) format.

tessflow = create_tessellation_flow(name='tessflow', out_format='gii')
tessflow.inputs.inputspec.subject_id = 'fsaverage'
tessflow.inputs.inputspec.subjects_dir = subjects_dir
tessflow.inputs.inputspec.lookup_file = lookup_file

We also create a conditional node to package the surfaces for ConnectomeViewer. Simply set cff to “False” to ignore this step.

cff = True
if cff:
    cff = pe.Node(interface=cmtk.CFFConverter(), name='cff')
    cff.inputs.out_file = 'Meshes.cff'

Outputs

Create a datasink to organize the smoothed meshes Using regular-expression substitutions we can remove the extraneous folders generated by the mapnode.

datasink = pe.Node(interface=nio.DataSink(), name="datasink")
datasink.inputs.base_directory = 'meshes'
datasink.inputs.regexp_substitutions = [('_smoother[\d]*/', '')]

Execution

Finally, create and run another pipeline that connects the workflow and datasink

tesspipe = pe.Workflow(name='tessellate_tutorial')
tesspipe.base_dir = output_dir
tesspipe.connect([(tessflow, datasink, [('outputspec.meshes',
                                         '@meshes.all')])])

If the surfaces are to be packaged, this will connect the CFFConverter node to the tessellation and smoothing workflow, as well as to the datasink.

if cff:
    tesspipe.connect([(tessflow, cff, [('outputspec.meshes',
                                        'gifti_surfaces')])])
    tesspipe.connect([(cff, datasink, [('connectome_file', '@cff')])])

tesspipe.run()

Example source code

You can download the full source code of this example. This same script is also included in Nipype1 Examples Niflow under the package/niflow/nipype1/examples directory.