[MINC-users] retrieve coregistration

Tue Sep 20 11:37:13 EDT 2016

Hi Jérôme,

Thank you for the files. So I tried just converting your original file to
ecat using minctoecat, and then converting it back with ecat to minc:

minctoecat CHAPO06620_F15_04_DYN_avg.mnc test.v
ecattominc test.v test.mnc

The resulting test.mnc has reordered dimensions, but the two MINC volumes
align properly. So it does seem as though there is an issue with the
imgratio tool, it is somehow changing the voxel-to-world transform in an
unexpected way.

I wrote a little Python 3 script that fixes this - it's enclosed (and
poorly tested at this point, so be careful).

Here's a link to the ratio file with the corrected transform:

https://www.dropbox.com/s/nwe7r2cqaik97vw/ratio_fixed.mnc?dl=0

The corrected file looks fairly good when overlaid on the original in
Display, so hopefully imgratio isn't changing the underlying sampling of
the data, but rather it is only failing to properly preserve the transform.

    -bert

On Fri, Sep 16, 2016 at 4:58 AM, Jérôme Redouté <redoute at cermep.fr> wrote:

> Hi,
> modifying zstart effectively corrects z mismatch
> unfortunately, x and y do not align files correctly and I keep a small
> mismatch
> Here is an example of dataset:
> https://dl.univ-lyon1.fr/iulbw8zre0
> Thanks
> Jerome
>
>
>
> Le 15/09/2016 à 14:58, Robert D. Vincent a écrit :
>
>> Hi,
>>
>> So the xspace and yspace values have changed slightly, but the zspace
>> origin has been changed dramatically.
>>
>> If you do the following command:
>>
>> minc_modify_header -dinsert zspace:start=48.3 ratio_original.mnc
>>
>> Do the files line up correctly?
>>
>>
>>
>> On Thu, Sep 15, 2016 at 8:24 AM, Jérôme Redouté <redoute at cermep.fr>
>> wrote:
>>
>> Hi Robert,
>>> here is the output of mincinfo
>>>
>>> file: CHAPO06620_F15_04_DYN_avg.mnc
>>>
>>>> image: signed__ short -32768 to 32767
>>>> image dimensions: xspace zspace yspace
>>>>      dimension name         length         step        start
>>>>      --------------         ------         ----        -----
>>>>      xspace                    176          0.6        -52.8
>>>>      zspace                    164    -0.601562         48.3
>>>>      yspace                    208    -0.601562      63.8531
>>>>
>>>>
>>>> file: ratio_original.mnc
>>>> image: signed__ short -32000 to 32000
>>>> image dimensions: zspace yspace xspace
>>>>      dimension name         length         step        start
>>>>      --------------         ------         ----        -----
>>>>      zspace                    164    -0.601562           -0
>>>>      yspace                    208     0.601562     -62.2617
>>>>      xspace                    176          0.6        -52.5
>>>>
>>>> PS: imgratio is a tool from Turku's PET center
>>>
>>> Jerome
>>>
>>>
>>>
>>>
>>> Le 15/09/2016 à 12:23, Robert D. Vincent a écrit :
>>>
>>> Hi Jérôme,
>>>>
>>>> I don't know the "imgratio" tool, is it possible it is changing or
>>>> deleting
>>>> the voxel-to-world transform? What is the output of mincinfo on the two
>>>> mnc
>>>> files?
>>>>
>>>> If the basic sampling grid hasn't changed, you should be able to use
>>>> minc_modify_header to put the original voxel-to-world transform back
>>>> into
>>>> the ratio.mnc header.
>>>>
>>>>       -bert
>>>>
>>>> On Thu, Sep 15, 2016 at 5:59 AM, Jérôme Redouté <redoute at cermep.fr>
>>>> wrote:
>>>>
>>>> Hi all,
>>>>
>>>>> I need your help concerning a coregistration loss problem during image
>>>>> processing.
>>>>>
>>>>> I have a mnc PET 4D volume that I need to process in ECAT format.
>>>>>
>>>>> My processing steps are as follows:
>>>>>
>>>>> DYN.mnc => ecat with minctoecat
>>>>>
>>>>> ecat processing (imgratio) to generate a ratio image => ratio.v
>>>>>
>>>>> convert back to minc using ecattominc => ratio.mnc
>>>>>
>>>>> my problem is that ratio.mnc is not in alignment with DYN.mnc
>>>>>
>>>>>
>>>>> How should I correct ratio.mnc header to bring this alignment back?
>>>>>
>>>>> Thanks for your help
>>>>>
>>>>> 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
>>
>>
>
> --
> ==================================================================
> 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
>