This process determines cortical thickness by measuring the distance between each white matter mesh vertex and the intersection of the vertex normal with the gray matter mesh using the RicNormalThickness program. There is a check to make sure that the normal from the first mesh does not intersect itself before intersecting the other mesh.
The required input files are a gray matter mesh and a white matter mesh.
There are a number of optional inputs.
An output file directory can be specified. If not specified the output will go to the directory that the meshes came from. The default directory is the ric subdirectory of the subject
There are minimum and maximum allowable distances for cortical thickness. If gray matter vertex to white matter vertex distance is outside the limits, it will be discarded.
Surface curvature texture maps can be used to eliminate vertices in areas of high curvature which can be prone to measurement error. Minimum and maximum curvature values can be entered for each. If the method for measuring curvature for the texture uses an absolute value then the minimum curvature value should be entered as zero.
There are three methods available for the actual calculations. The Brute Force method compares each gray matter vertex to every white matter vertex. The Subdivision method breaks down the brain into a set of smaller overlapping volumes. This method is much quicker. The Threaded method extends the Subdivision method by processing each of the smaller volumes with an individual thread. This is a big advantage with multiprocessor computers.
There is an output summary displayed in the process window. There are also a number of output files as well.
There is a comma seperated variable output file (.csv) which contains the summary data as well as a histogram of the cortical thickness values.
There is also a texture map (_thick.tex) of the cortical thickness distributions which can be mapped to the white matter mesh
Additional files if verbose output selected:
There is a vector mesh (_vector.mesh) where each vector connects a white matter vertex to the nearest gray matter vertex. When displayed along with the white matter mesh, it gives a good verification of the process.
There are also two output meshes showing normals from both the gray matter and white matter mesh surfaces (_nor.mesh).
Download Instructions:
Click on the link below to download the file. Place it in the root BrainVISA directory. This directory should have subdirectories bin, brainvisa, include, lib, share. Run gunzip to uncompress the file. The run 'tar -xvf RicNormalThickness.tar' to put all the files in their proper places. When you run BrainVISA, there will be an additional catagory 'ric' that will contain the plugin.
Download NormalThickness plugin from NITRC (binaries for Linux, Mac and Windows and source code are available )