Hi Mateo,

 

It looks like mri_convert terminated prematurely when you tried to convert those 512 dicom files.

 

Those error messages are printed when mri_convert wasn’t able to retrieve DTI information from dicom. If you don’t need the DTI values, you can ignore those error messages.

 

Also, looking at the codes, I think if you set environment variable FS_LOAD_DWI to 0 before running mri_convert, it should finish convert all 512 dicom files. To set the environment variable, you do ‘export FS_LOAD_DWI = 0’ in bash, and ‘setenv FS_LOAD_DWI 0’ in csh.

 

Can you give it a try?

 

Thanks.

 

Yujing

 

 

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> On Behalf Of Mateo Jean
Sent: Tuesday, April 26, 2022 7:51 AM
To: freesurfer <freesurfer@nmr.mgh.harvard.edu>
Subject: [Freesurfer] Mri Convert Bruker MGRE files dicom to nii error

 

        External Email - Use Caution        

Hello FreeSurfer Developers

 

I hope i don't spam, if anyone has a clue...

 

 

I need to convert 512 dicom files from Bruker into .nii . I firstly add .dcm to my Bruker multi gradient echo phase files then enter the following command :

mri_convert MRIm001.dcm phaseMGE.nii

 

Here is the first message :

 

reading from MRIm001.dcm...

Starting DICOMRead2()

dcmfile = /mnt/c/Users/Sahadatou/Documents/Stage_JEAN_2022/data/fantomes/tests/bug_test2/MRIm001.dcm

dcmdir = /mnt/c/Users/Sahadatou/Documents/Stage_JEAN_2022/data/fantomes/tests/bug_test2

ERROR: don't know how to get DWI parameters from --Bruker BioSpin MRI GmbH --

ERROR: GetDICOMInfo(): dcmGetDWIParams() 1

DICOM File: /mnt/c/Users/Sahadatou/Documents/Stage_JEAN_2022/data/fantomes/tests/bug_test2/MRIm001.dcm

break /home/rd521/space_freesurfer/build/docker_ubuntu18/trunk/rd521-7.2.0/utils/DICOMRead.cpp:4930

Ref Series No = 4

Found 514 files, checking for dicoms

ERROR: don't know how to get DWI parameters from --Bruker BioSpin MRI GmbH --

ERROR: GetDICOMInfo(): dcmGetDWIParams() 1

DICOM File: /mnt/c/Users/Sahadatou/Documents/Stage_JEAN_2022/data/fantomes/tests/bug_test2/MRIm001.dcm

break /home/rd521/space_freesurfer/build/docker_ubuntu18/trunk/rd521-7.2.0/utils/DICOMRead.cpp:4930

ERROR: don't know how to get DWI parameters from --Bruker BioSpin MRI GmbH --

ERROR: GetDICOMInfo(): dcmGetDWIParams() 1

 

Then 508 error messages like the three last lines appear.

It ends with :

 

First Sorting

Computing Slice Direction

Vs: 0 0 1

Vs: 0 0 1

Second Sorting

IsDWI = 0, IsPhilipsDWI = 0

Counting frames

nframes = 8

nslices = 63

ndcmfiles = 508

ERROR: the number of frames * number of slices does

not equal the number of dicom files.

 

I did a test with only 448 files (I just copy-paste the first 448 files = 7 TE * 64 slices) and I obtained the same error messages but the nifti files are created and seem usable (quick test with Fiji). Here are the last lines :

 

First Sorting

Computing Slice Direction

Vs: 0 0 1

Vs: 0 0 1

Second Sorting

IsDWI = 1, IsPhilipsDWI = 0

Counting frames

nframes = 7

nslices = 64

ndcmfiles = 448

INFO: rescale not needed

PE Dir = COL (dicom read)

Loading pixel data

TR=50.00, TE=3.10, TI=0.00, flip angle=15.00

i_ras = (-1, -0, 0)

j_ras = (-0, -1, 0)

k_ras = (-0, -0, 1)

writing to phaseMGR.nii...

Saving bvals and bvecs

 

I'm pretty new with mri convert and I don't understand why it stops converting to nii with 512 files (8 TE) but not 448 files for Brucker files.

 

Best Regards,

 

                      Mateo

 

 

Version: freesurfer-linux-ubuntu18_x86_64-7.2.0-20210721-aa8f76b