open-discussion
open-discussion > RE: IRTK registration on abdominal CT
Nov 12, 2013 11:11 PM | Torsten Rohlfing
RE: IRTK registration on abdominal CT
Hi Zhoubing -
To be very clear: I have no experience whatsoever with IRTK.
But generally, when registering thresholded data, I would be wonder whether both images have been thresholded, or only either source or target. Also, what happens with the masked-out pixels? Are they treated as missing, or taken at face value (e.g., zero?)
Interestingly, from the "rreg" documentation:
To be very clear: I have no experience whatsoever with IRTK.
But generally, when registering thresholded data, I would be wonder whether both images have been thresholded, or only either source or target. Also, what happens with the masked-out pixels? Are they treated as missing, or taken at face value (e.g., zero?)
Interestingly, from the "rreg" documentation:
- Parameter name : "Padding value"
- Parameter value :
A number used to determine if voxels in the target image are used
during registration. All voxels with an intensity less than or equal to
the padding value are ignored. - If you were using this option, try doing without it. In your case, setting the non-bone pixels to zero and using them in the registration as such will likely help the alignment. More so than ignoring the data at least.
- That's pretty much all I can think of. Like I said, no experience specifically with IRTK here. Or with niftyreg or flirt, to be honest ;)
- Good luck!
- Torsten
Threaded View
Title | Author | Date |
---|---|---|
Zhoubing Xu | Nov 12, 2013 | |
Torsten Rohlfing | Nov 12, 2013 | |
Zhoubing Xu | Nov 12, 2013 | |
Torsten Rohlfing | Nov 12, 2013 | |
Zhoubing Xu | Nov 22, 2013 | |
ELISEE ILUNGA | Nov 13, 2013 | |
Jose Manjon | Nov 13, 2013 | |
Torsten Rohlfing | Nov 12, 2013 | |