Hi Emmanuel,

sorry for the delay answering but I wanted to check everything again before.

I started it againg and now the error is different.

Briefly, I had:

1. Transformed the DICOM images into FSL (analyze) with MRI convert (I got in this step the bvals and bvecs files)

2. I have transformed with dcm2 the hdr images with the option "Modify Nifti" to SPM 8 4D NIFTI

3. I have changed the names of the file 0 into nodif and the 33 into trace.

4. I have create a 4D nifti image with the files from 1 to 32.

5. i have done the eddy correction with FDT in fsl5.0

6. I have made a brain mask with bet from the DTI corrected file.

7. I tryed tu run DTIFIT RECONSTRUCT DIFFUSION TENSOR, selecting manually the files.

Then I got this error:

"Errors: data and bvals/bvecs do not contain the same number of entries"

So my guess is that there is a problem with the data fromat from the begining. I am sending you attached the bvals and bvecs files.

Thank you very much for your help, I am sure that it is a stupid preprocessing step but I am not able to solve it.

2017-01-27 23:09 GMT+01:00 Emmanuel Vallee <[log in to unmask]>:

Hi Angela,

Can you post the content of the bvecs and bvals files, and also the output of the command: fslinfo <your_4D_image>

Cheers,

Manu

On 27 Jan 2017, at 12:35, Angela Bernabeu Sanz <[log in to unmask]> wrote:

Dear FSL experts,

I am new with FSL software. I am trying to analyze DTI data. I got DICOM data from a 3T Philips achieva.

I have converted them into 3D nii.gz images with dcm2 software which I have in the FSL package, I have labelled the trace image and the B0, and with the rest I have built up a 4D nii.gz image.

I am able to perform the eddy current correction but I am not able to perform the DTIFit reconstruction diffusion tensors. When I try it, it appears an error message saying:

“Errors: Could not open matrix file…../bvecs, Could not open matrix file…/bvals

Error: bvecs must be either 3xN or Nx3

I think it is a problem with the images format, or the image conversion from the beginning, as my bvecs and bvals files are in txt format.

I have tried also to transform the DICOM images with MRIConvert software but I have the same problem. I have read that this software is useful with fMRI images but not adequate with DTI.

Can anyone suggest me which software I may use to get the correct images files to work with FSL??, how can I get the 3xN bvecs and bvals files?

Thank you