Have you ever tried dinifti (http://cbi.nyu.edu/software/dinifti.php)?
I use Ubuntu and it is also available in repos (using e.g. apt-get).
As far as I experienced it works for me the same way as mri_convert (from
FreeSurfer); and I have never had any problem.
Auer, Tibor M.D.
Biomedizinische NMR Forschungs GmbH
am Max-Planck Institut für Biophysikalische Chemie
Am Fassberg 11
Mail: [log in to unmask]
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf
Of Mark Jenkinson
Sent: Wednesday, March 18, 2009 1:06 PM
To: [log in to unmask]
Subject: Re: [FSL] From dicom to nii
They don't use the file order for our dicom images, so I guess they
However, there are lots of variations of dicom around, depending on
of the scanner and/or the programmer of the sequence. Is it a
standard sequence on
a common scanner that you are trying to convert, or something a little
There maybe people out there who have solved your exact problem if you
a little more info.
All the best,
On 18 Mar 2009, at 10:20, Hallvard Røe Evensmoen wrote:
> Yes I have:) I am not sure exactly what kind of information from the
> dicom files that these programs use. But based on the problems I
> get, I suspect that they don`t sort the images based on header info,
> but instead depend on the sequence that the dicom images are stored
> in. Do you know anything about that?
> Siterer Mark Jenkinson <[log in to unmask]>:
>> We have had good results from using mri_convert (from FreeSurfer)
>> or dcm2nii (from mricron). Have you tried these?
>> All the best,
>> P.S. You can only attach very small files to emails that go to the
>> whole FSL list, to avoid everyone getting full mailboxes.
>> On 18 Mar 2009, at 09:05, Hallvard RÃ¸e Evensmoen wrote:
>>> I have tried to convert a series of 3d dicom images into nifti using
>>> dicom2nii, MRIConverter and LONI debabeler. The best I get is all
>>> slices from the different volumes added together with dim3=1 (image
>>> attached). Anyone got a better suggestion than writing a scipt
>>> that first
>>> uses fslmerge to merge the 32 slices along the z axis to get the
>>> volumes and
>>> finally use fslmerge to merge the volumes in time.
>>> Yours sincerely
>>> Hallvard Røe Evensmoen