| Author |
 Topic  |
|
|
Dichter
BIAC Faculty
190 Posts |
 Posted - Oct 16 2007 : 10:43:23 AM
|
When running post-stats in a first-level analysis, we are getting the following error:
** ERROR (nifti_image_read): failed to find header file for 'stats/sigmasquareds'
** ERROR: nifti_image_open(stats/sigmasquareds): bad header info
Error: failed to open file stats/sigmasquareds
The nifti file is where it's supposed to be and the right size.
This error is new since upgrading to FSL 4. Thanks for any help. |
Gabriel S. Dichter, PhD
UNC Departments of Psychiatry & Psychology
http://www.can.unc.edu/ |
|
|
tankersley
BIAC Alum
USA
143 Posts |
Posted - Jan 10 2008 : 3:09:01 PM
|
Do you remember why you were getting this error?
Dharol |
 |
|
|
tankersley
BIAC Alum
USA
143 Posts |
Posted - Jan 14 2008 : 10:32:38 AM
|
I am getting the following error when running Feat Level 1:
WARNING: nifti_read_buffer(/mnt/users.q.shares/tankersley.2VInDua08vm9fedS6wsM/Control/Shapes/47072/47072_ica_nofBET/shapes1.ica/awesome_data.nii):
data bytes needed = 557056
data bytes input = 0
number missing = 557056 (set to 0)
Also, when i try to load the nifit file into MATLAB i get the following error:
??? Error using ==> readmr
Input size array doesn't match size of data!
A random internet forum recommended either saving in hdr/img format, or "You can use FSL's "correctheader" command to fix that. But ONLT
if you have 4 missing bytes and get the warning message you
posted. . .Without this correction, your data might be shifted by one voxel."
http://www.bic.mni.mcgill.ca/pipermail/minc-users/2007-October.txt
Has anyone encountered this problem, and does one of these solutions seem more advisable than the other?
Thanks,
dharol
|
|
|
|
dvsmith
Advanced Member
USA
218 Posts |
Posted - Jan 14 2008 : 11:06:00 AM
|
Can you load the original .bxh file into matlab (i.e., the one you got from the scanner)? If that works, verify that the length of the data.data is equal to t size listed in data.info.dimensions.size. I'm not sure if bxh2analyze would've worked if that were the case, though. If the original data is messed up, you'll probably have to get it re-reconed or either fix the .bxh file yourself.
-David
|
|
|
|
syam.gadde
BIAC Staff
USA
421 Posts |
Posted - Jan 14 2008 : 11:06:16 AM
|
Does the size of awesome_data.nii look unreasonable? i.e. is it zero-length? As an example, if the image is is 64x64x32x200, and it is 16-bit integers, then it should be of length 64x64x32x200x2 + 352 (for the NIFTI header) == 52429152 bytes.
|
|
|
|
tankersley
BIAC Alum
USA
143 Posts |
Posted - Jan 14 2008 : 11:14:27 AM
|
The original data/ header file are fine. awesome_data.nii is only 1KB, so i guess i'll try to remake it and see if it becomes a more reasonable size.
Thanks,
Dharol |
|
|
| |
Topic |
|
FSL 4 error message
