Error messages

Hi, There!

I am a clinician and beginner of brain imaging analysis. I have installed MRtrix and tried to analyze my own data using it.

the parameter of MRI data are below;
b value =600
32 directions

However, whenever, I use the command “dwipreproc”, I encounter the error messages.

command line >> dwipreproc -rpe_none AP dti_denoise.nii dti_preproc.nii
mrinfo: [ERROR] no diffusion encoding information found in image "series.mif"
dwipreproc: [ERROR] Number of volumes in gradient table does not match input image

Dear Expert, What did cause the error message? I guess that the low b-value is a possible cause. Is it right?

Please give me an answer, If my guess is right, I should change the MRI scan protocol.

Thank you for your consideration and asssistence.



p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 10.0px Monaco; color: #f4f4f4; background-color: #000000; background-color: rgba(0, 0, 0, 0.85)}
span.s1 {font-variant-ligatures: no-common-ligatures}

dwipreproc -rpe_none AP dti_denoise.nii dti_preproc.nii


Hi there,

The first error message is key (the second is slightly misleading):

mrinfo: [ERROR] no diffusion encoding information found in image "series.mif"

The script cannot operate unless it has access to the diffusion gradient table that was used when acquiring the image data; this information can be provided to the script either using one of the command-line options (-grad or -fslgrad), or embedded within the image header (which is not possible here as you have used the NIfTI file format).

If you have used software to convert to .nii from DICOM, then that software should also be capable of extracting and exporting the diffusion gradient table for you. This includes MRtrix3’s own mrconvert, which can export the diffusion gradient table in either MRtrix (-export_grad_mrtrix) or FSL bvecs / bvals (-export_grad_fsl) formats, or will embed these data within the image header if the MRtrix image format is used.

Hopefully that’s enough information for you to go off; admittedly our documentation is still lacking a page discussing handling of these data…

Please give me an answer, If my guess is right, I should change the MRI scan protocol.

Even though the low b-value isn’t the cause of the reported issue, I would strongly recommend changing the protocol to obtain more diffusion contrast.


Thank you so much for your kind direction!!!

I will try again with another raw data.



Hi there,

I had the same issue. By providing diffusion gradient table (-fslgrad bevc bval) the problem is solved.


This was a great solution. Thanks for sharing your knowledge. It fixed my problem as well.