Hi All,
I am having problems using data that has been eddy corrected using FSL. When I run DKE on the raw nifti files, DKE works fine, but when I try using the eddy corrected version, I get an error message  saying:
"Reading input images... Undefined function or variable 'niftifile'.
Error in read_nii (line 52)
Error in dke_estimate>read_img_set_nii (line 1824)
Error in dke_estimate (line 157)
Error in dke (line 182)
MATLAB:UndefinedFunction"
Any ideas?
Thanks,
Colleen

Hi Colleen, 

I am not familiar with how eddy correction changes the structure of the nifti file. Could you send me an example of the 4D nifti you are trying to use? 

Best, 
Emilie

Hi Emilie,
I have attached an example of the 4D nifti file that has been eddy corrected in FSL.
Thanks,
Colleen

Hi Emilie,
Were you able to download and test the eddy corrected 4D nifti file I sent?
Thanks,
Colleen

Yes! Can you send your dke_parameters file as well and your gradient table? This will allow me to run DKE in debug mode from this side so I can see what is going on.
 
Thanks,
Emilie

Here is the gradient table. This data was acquired on a Phillips 3T.

Here is the dke_parameters file.
Thank you for your help!
Colleen

Hi Emilie,
Were you able to get the files I sent? Any luck with trying to figure out the problem with my data?
Thanks,
Colleen

He Emilie,
I haven't heard from you in a while. Have you had a chance to test my data?
Thanks,
Colleen

Hi Colleen! 

I haven't had the chance to run your data just yet, I will let you know as soon as I ran it. 

Emilie

Hi Colleen,

I just processed your data. I think the problem was that you were trying to feed in a compressed nifti (i.e., nii.gz).  You just need to change the file type like this:

fslchfiletype NIFTI input.nii.gz

I also noticed a few other things:

(1) your data is non-isotropic, but you apply isotropic smoothing with fwhm_img= [3.375 3.375 3.375]

(2)  if you plan to coregister to an atlas after DKE I think you should turn off map_interpolation

(3) the default setting for median_filter_method is quite strict and I think you should trying changing from strong (2) to weak (1)

Cheers,

Kirk