Hi Adrienne,

If you haven't already done so, you should also send this to the cleaver mailing list (cleaver@sci.utah.edu, instructions at http://www.sci.utah.edu/software/cleaver.html under the Support tab).

Ayla

On May 29, 2013, at 4:30 PM, Adrienne Madison wrote:

Good Evening All,

I previously posted in attempts to get help meshing 3D brain geometries extracted from MRI data.  Segmentation was completed using Seg3D, and after complications with BioMesh3D, the Cleaver software option was suggested.

I am unfortunately now facing other dilemmas:

1. After meshing the models using Cleaver, I used a .ply display function written in MATLAB that provides a graphics display of the data in the .ply file.  My results are merely a black brick.  The function was tested with sample .ply data, and the accurate geometry was displayed.  I am attaching images of the sample file and my file (ktest) when utilizing the MATLAB function. 

1b.  I input 5 files representing 4 masks or materials and the background (Obtained from stage 2 of Biomesh3D) to be meshed by cleaver: air, mat1, mat2, mat3, mat4.  Is the "air" or background file responsible for the appearance of the black brick? When I attempt to not include the air file, cleaver fails.

2. Is there any way to "coarsen" meshes generated by cleaver. My models currently consist of  millions of vertices and tets, which overwhelm and freeze up FEM solvers and mesh viewers. 

Any suggestions or input you can provide to help solve these obstacles are greatly appreciated and urgently needed so that I can finish my research and defend my dissertation!  I am also attaching the .nrrd files which contain the segmented geometry of a normal brain with 4 materials (ammk.nrrd) and an abnormal brain with 4 materials (ammb.nrrd)

--

Adrienne Madison

CURO Graduate Engineering Fellow

604 Driftmier Engineering Center

Athens, GA 30602

706-542-4707

amadison@uga.edu

<ammb.nrrd><ammk.nrrd><ktest.png><sample.png>