| T O P I C R E V I E W |
| D.Smith |
Posted - Jun 15 2006 : 2:57:58 PM
We draw ~16 ROIs onto each subject's anatomical scan. We scan each subject twice. Is there a way to coregister the ROIs from the first scan to the second anatomical file, so that we don't have to draw all 16 ROIs a second time? I realize that their head is not in the exact same position during the second scan- but if we can determine the tilt and movement then shouldn't we also be able to apply that tilt and movement to our previously drawn ROIs?
-Drew |
| 2 L A T E S T R E P L I E S (Newest First) |
| josh.bizzell |
Posted - Jun 15 2006 : 5:14:37 PM
Chris is right. In fact, all you really need is an affine transformation (instead of a higher-order warping) since the brains are from the same subject. Many packages can do this; SPM, AIR, RView, FSL, et al. SPM is probably the easiest to use, and will let you apply the transformation matrix to the ROI image. Make sure to use nearest-neighbor interpolation on the ROI image so the values will remain integers.
- Josh |
| petty |
Posted - Jun 15 2006 : 4:44:27 PM
you could coregister the 1st scan to the 2nd scan, then just apply the saved transformation matrix to the ROI file. |
|
|
