Hi,
Using BET (v1.2) on axial (transversal) T2 stacks I get very bad results: only part of
the brain is extracted. Using different settings with the -f and -g options does not
produce much better results (btw, no other options were used).
Therefore, a few questions arise:
1.
The stacks measures 512 * 512 pixels with 24 slices.The voxel size is
0.4492 * 0.4492 * 5.0 mm.
The fact that the voxels are not cubic could that be a problem?
2.
If I inverse Y (flip the stack vertically) the results become even worse.
Is BET sensitive to spatial orientation of the stack?
Ed Gronenschild, PhD
Maastricht University
Dept. of Medical Informatics
Maastricht, The Netherlands
|