[MINC-users] information loss after mincreshape

Vladimir S. FONOV vladimir.fonov at gmail.com
Thu May 18 10:49:34 EDT 2017 

It's a "feature" of mincreshape: 
http://www.bic.mni.mcgill.ca/pipermail/minc-users/2015-March/004020.html


It bites people at the rate of approximately once per year.


Your resampling problem might be something completely different. Like a 
bug in mincresample when treating 4D volumes.

On 2017-05-18 10:28 AM, Jérôme Redouté wrote:
> Thank you Alex,
>
> Ok, it's still a bit obscure for me:
>
> If I understand, one of the 3 dimensions is a non-image dimension?
>
> Does it mean that it can only have 2 "image dimensions"?
>
> I don't really understand what you mean by "the 2 fastest varying". How
> can I identify the "non image dimension"
>
> thank you
>
> Jerome
>
>
>
>
> Le 18/05/2017 à 15:38, Alex Zijdenbos a écrit :
>> Hi Jerome,
>>
>> Although I don't think this explains what you are seeing, I suspect that
>> this:
>>
>> mincreshape +xdirection + ydirection +zdirection dyn.mnc tmp.mnc
>>
>> may not do what you think it does, due to this, rather non-intuitive, bit
>> in the man page:
>>
>>         +direction
>>                Flip images to give positive step value for spatial axes.
>> Note that the flipping of spatial axes only applies to "image
>> dimensions".
>> These are the  two  fastest
>>                varying (non-vector) dimensions in the file. If you
>> want to
>> flip a non-image dimension, you can convert it to an image dimension with
>> -dimsize dimname=-1 (the -1
>>                means don't really change the size). Check out the
>> examples.
>>
>> -- A
>>
>>
>> On Thu, May 18, 2017 at 9:30 AM, Jérôme Redouté <redoute at cermep.fr>
>> wrote:
>>
>>> Hi all,
>>>
>>> We are facing a new problem with mincreshape while processing dynamic
>>> PET
>>> data.
>>>
>>> here is our process:
>>>
>>>      dcm2mnc ==> dyn.mnc
>>>
>>>      mincreshape +xdirection + ydirection +zdirection dyn.mnc tmp.mnc
>>>
>>> we want then to reorient the image:
>>>
>>>      param2xfm -rotations 90 0 180 dcmPET2mnc.xfm
>>>
>>>      mincresample -transformation dcmPET2mnc.xfm tmp.mnc final.mnc -like
>>> template.mnc
>>>
>>> this last step fails with the following warnings:
>>>
>>>      "Irregular dimension in file tmp.mnc contains non-monotonic or
>>> repeated values"
>>>
>>>      "substituting a step size of 1"
>>>
>>> and the resulting file contains no data
>>>
>>> If we don't do the mincreshape step ==>  no problem with mincresample
>>>
>>> Any ideas?
>>>
>>> (we are using the dev branch of Minc tools)
>>>
>>> Thanks for your help
>>>
>>> Jerome
>>>
>>>
>>>
>>>
>>>
>>> Le 30/06/2016 à 16:29, Jérôme Redouté a écrit :
>>>
>>>> Thank you!
>>>>
>>>> I 'm going to test it right now
>>>>
>>>> Best
>>>> Jerome
>>>>
>>>> Le 30/06/2016 16:18, Robert D. Vincent a écrit :
>>>>
>>>>> Hi Jerome,
>>>>>
>>>>> I believe I have fixed the issue. After thinking about it, I
>>>>> believe it
>>>>> is
>>>>> safe to preserve the information whenever it happens to be present, so
>>>>> that
>>>>> is now the behavior of mincreshape.
>>>>>
>>>>> I pushed the fix to the develop branch.
>>>>>
>>>>>       -bert
>>>>>
>>>>> On Mon, Jun 27, 2016 at 8:10 AM, Jérôme Redouté <redoute at cermep.fr>
>>>>> wrote:
>>>>>
>>>>> Hi Robert,
>>>>>> did you find how to fix the mincreshape bug?
>>>>>> thank you
>>>>>> Best regards
>>>>>>
>>>>>> Jerome
>>>>>>
>>>>>> Le 15/06/2016 19:19, Robert D. Vincent a écrit :
>>>>>>
>>>>>> One thing I'd like to check: is the file you're using a
>>>>>>> regularly-sampled
>>>>>>> file, i.e. is there a constant time duration for each image?
>>>>>>>
>>>>>>>
>>>>>>> On Wed, Jun 15, 2016 at 1:09 PM, Robert D. Vincent <
>>>>>>> robert.d.vincent at mcgill.ca> wrote:
>>>>>>>
>>>>>>> Hi Jérôme,
>>>>>>>
>>>>>>>> Thanks for the report, I've reproduced it here. It seems that
>>>>>>>> mincreshape
>>>>>>>> fails to copy the time dimension properly. I will fix it as soon as
>>>>>>>> possible.
>>>>>>>>
>>>>>>>>        -bert
>>>>>>>>
>>>>>>>> On Wed, Jun 15, 2016 at 9:53 AM, Jérôme Redouté <redoute at cermep.fr>
>>>>>>>> wrote:
>>>>>>>>
>>>>>>>> Dear all,
>>>>>>>>
>>>>>>>>> I'm processing dynamic PET data acquired on Siemens Biograph mMR.
>>>>>>>>>
>>>>>>>>> after conversion from DICOM to MINC using dcm2mnc, I can extract
>>>>>>>>> frames
>>>>>>>>> timings using:
>>>>>>>>>
>>>>>>>>> mincinfo -varvalue time xxx.mnc
>>>>>>>>>
>>>>>>>>> BUT,
>>>>>>>>>
>>>>>>>>> after mincreshape +xdirection +ydirection +zdirection xxx.mnc
>>>>>>>>> xxx_res.mnc
>>>>>>>>>
>>>>>>>>> mincinfo -varvalue time xxx_res.mnc outputs only "0"
>>>>>>>>>
>>>>>>>>> any ideas about this bug?
>>>>>>>>>
>>>>>>>>> Thanks
>>>>>>>>> Jerome
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> ==================================================================
>>>>>>>>> Jérôme Redouté
>>>>>>>>> Ph.D. - Ingénieur de Recherche - Université Claude Bernard - Lyon1
>>>>>>>>> CERMEP - Imagerie du vivant
>>>>>>>>> 59 Bd Pinel. 69677 Bron - FRANCE
>>>>>>>>> tel : 33 (0)4 72 68 86 18 (bureau)
>>>>>>>>> tel : 33 (0)4 72 68 86 00 (standard)
>>>>>>>>> fax : 33 (0)4 72 68 86 10
>>>>>>>>> ==================================================================
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> MINC-users at bic.mni.mcgill.ca
>>>>>>>>> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>> MINC-users at bic.mni.mcgill.ca
>>>>>>> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>> ==================================================================
>>>>>> Jérôme Redouté
>>>>>> Ph.D. - Ingénieur de Recherche - Université Claude Bernard - Lyon1
>>>>>> CERMEP - Imagerie du vivant
>>>>>> 59 Bd Pinel. 69677 Bron - FRANCE
>>>>>> tel : 33 (0)4 72 68 86 18 (bureau)
>>>>>> tel : 33 (0)4 72 68 86 00 (standard)
>>>>>> fax : 33 (0)4 72 68 86 10
>>>>>> ==================================================================
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> MINC-users at bic.mni.mcgill.ca
>>>>>> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>>>>>>
>>>>>> _______________________________________________
>>>>> MINC-users at bic.mni.mcgill.ca
>>>>> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>>>>>
>>>>>
>>>>
>>> _______________________________________________
>>> MINC-users at bic.mni.mcgill.ca
>>> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>>>
>> _______________________________________________
>> MINC-users at bic.mni.mcgill.ca
>> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>>
>
> _______________________________________________
> MINC-users at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users


-- 
Best regards,

  Vladimir S. FONOV ~ vladimir.fonov <at> gmail.com

More information about the MINC-users mailing list