Hi Yari,

good to hear. Nonetheless If you happen to acquire data which can be shared and 
reproduces that error we would be happy to take a look at it and find the cause 
of the bug.

Best,
Caspar

-----Ursprüngliche Nachricht-----
Von: yari mattei [mailto:[email protected]] 
Gesendet: Dienstag, 16. Juni 2015 15:40
An: [email protected]
Betreff: Re: [mitk-users] Mirror image

Hi Caspar,

sorry I am freaking out. The flipping filter (which was always there) actually 
solve the problem.
After flipping across axis 1, I create the segmentation and then the smoothed 
polygon model which is finally at in the right orientation.

Thanks a lot.

Cheers,
Yari

Il 16.06.2015 15:18, Goch, Caspar Jonas ha scritto:
> Hi Yari,
>
> if you select an image and open the "choose an operation" dropdown box 
> there should be a scrollbar on the right. At the very bottom should be 
> a section "Misc" which, among others, contains a "Flipping" Filter
>
> Best,
> Caspar
>
> -----Ursprüngliche Nachricht-----
> Von: yari mattei [mailto:[email protected]]
> Gesendet: Dienstag, 16. Juni 2015 14:42
> An: [email protected]
> Betreff: Re: [mitk-users] Mirror image
>
> Hi Caspar,
>
> I am trying to do what you suggest but on the "basic image processing" I 
> cannot find anything to flip image (just denoising and morphological filters 
> and arithmetic to do on 2 images) I also tried to look around in the other 
> view but I didn't find anything.
>
> Am I making some mistake?
>
> Thanks
> Yari
>
> Il 16.06.2015 14:00, Goch, Caspar Jonas ha scritto:
>> Hi Yari,
>>
>> as far as I am aware there is no button for it in the MITK binary release. 
>> However... the MITK Diffusion 2014.10.02 binary release contains the "basic 
>> image processing" view, which is able to flip images (no surface though) so 
>> you could flip the image to match the surface. This might not be what you 
>> want, depending on your use case.
>>
>> Does this help?
>>
>> Best,
>> Caspar
>>
>> -----Ursprüngliche Nachricht-----
>> Von: yari mattei [mailto:[email protected]]
>> Gesendet: Dienstag, 16. Juni 2015 13:32
>> An: [email protected]
>> Betreff: Re: [mitk-users] Mirror image
>>
>> Hi Caspar,
>>
>> thanks for the answer, at the moment I have only real image that I cannot 
>> share but I asked for a sherable sample.
>> So there is no way to mirror the coordinate - of one of the three 
>> views
>> - directly in the binary release?
>>
>> Do you know perhaps another way to do that?
>>
>> Cheers,
>> Yari
>>
>> Il 15.06.2015 17:15, Goch, Caspar Jonas ha scritto:
>>> Hi Yari,
>>>
>>> I just tried the same pipeline on some image we had here using the 2014.10 
>>> binary release and I cannot reproduce that behavior. Would it be possible 
>>> for you to provide a test data set on which the error occurs so we can try 
>>> and narrow it down? The details look ok to me.
>>>
>>> Best,
>>> Caspar
>>>
>>> -----Ursprüngliche Nachricht-----
>>> Von: yari mattei [mailto:[email protected]]
>>> Gesendet: Montag, 15. Juni 2015 15:54
>>> An: [email protected]
>>> Betreff: Re: [mitk-users] Mirror image
>>>
>>> Hi Caspar,
>>>
>>> Yes the pipeline is exactely the one you describe. Yes the surface is 
>>> already mirrored in step 3 (after "create smoothed polygon model" a can see 
>>> that the surface is mirrored).
>>>
>>> Because I was not sure about the data you asked me I paste the whole image 
>>> details :
>>>
>>> Image (000000000A307E30)
>>> Dimension: 4
>>> Dimensions: 256 224 208 9
>>> Channel: Unnamed []
>>> PixelType:
>>> BytesPerElement: 1
>>> ComponentType: unsigned_char
>>> NumberOfComponents: 1
>>> BitsPerComponent: 8
>>>
>>> TimeGeometry: ProportionalTimeGeometry (000000000A7489E0)
>>> TimeSteps: 9
>>> FirstTimePoint: 0
>>> StepDuration: 1 ms
>>> Time Bounds: 0 - 9
>>>
>>> GetGeometryForTimeStep(0): SlicedGeometry3D (000000000A45BB00)
>>> IndexToWorldTransform:
>>> Matrix:
>>> 0.343466 0 0
>>> 0 0.346719 0
>>> 0 0 0.198195
>>> Offset: [0, 0, 0]
>>> Center: [0, 0, 0]
>>> Translation: [0, 0, 0]
>>> Inverse:
>>> 2.91149 0 0
>>> 0 2.88418 0
>>> 0 0 5.04553
>>> Scale : 1 1 1
>>> BoundingBox: ( 0,256 0,224 0,208 )
>>> Origin: [0, 0, 0]
>>> ImageGeometry: 1
>>> Spacing: [0.343466, 0.346719, 0.198195]
>>> EvenlySpaced: 1
>>> DirectionVector: [0, 0, 1]
>>> Slices: 208
>>>
>>> GetPlaneGeometry(0): PlaneGeometry (000000000A0882E0)
>>> IndexToWorldTransform:
>>> Matrix:
>>> 0.343466 0 0
>>> 0 0.346719 0
>>> 0 0 0.198195
>>> Offset: [0, 0, 0]
>>> Center: [0, 0, 0]
>>> Translation: [0, 0, 0]
>>> Inverse:
>>> 2.91149 0 0
>>> 0 2.88418 0
>>> 0 0 5.04553
>>> Scale : 1 1 1
>>> BoundingBox: ( 0,256 0,224 0,1 )
>>> Origin: [0, 0, 0]
>>> ImageGeometry: 1
>>> Spacing: [0.343466, 0.346719, 0.198195]
>>> ScaleFactorMMPerUnitX: 0.343466
>>> ScaleFactorMMPerUnitY: 0.346719
>>> Normal: [0, 0, 0.198195]
>>>
>>> At the moment I am using the build-in functionality and I would like to 
>>> know how to do it in the binary application.
>>>
>>> Cheers
>>>
>>> Yari
>>>
>>> Il 15.06.2015 14:45, Goch, Caspar Jonas ha scritto:
>>>> Hi Yari,
>>>>
>>>> just to be clear, your workflow is:
>>>> 1. Load a dicom file
>>>> 2. Do a segmentation
>>>> 3. create surface from segmentation 4. Save surface as stl 5. Load 
>>>> as stl
>>>>
>>>> Is the stl already mirrored in step 3 or do you need steps 4 and 5? What 
>>>> coordinate system is the dicom file in? Could you give the matrix/spacing? 
>>>> (right click on image -> Details) Are you using build-in MITK 
>>>> functionality for step 3 or do you create the surface yourself?
>>>>
>>>> Are you looking for ways to mirror the surface in the binary application 
>>>> or how to do it in code?
>>>>
>>>> Best,
>>>> Caspar
>>>>
>>>> -----Ursprüngliche Nachricht-----
>>>> Von: yari mattei [mailto:[email protected]]
>>>> Gesendet: Freitag, 12. Juni 2015 14:00
>>>> An: [email protected]
>>>> Betreff: [mitk-users] Mirror image
>>>>
>>>> Hi everybody,
>>>>
>>>> I am using the MITK workbench to produce some stl file from a DICOM file, 
>>>> but I have a problem.
>>>>
>>>> I can create the segmentation and save the stl file and it looks good 
>>>> besides the fact that is mirrored (just to be clear with mirrored I mean 
>>>> that if the image should be AAABBB it appears as BBBAAA).
>>>>
>>>> There is a way to mirror sagital, axial or coronal view in the workbench? 
>>>> Or any other way to solve this problem?
>>>>
>>>> I hope that my explanation is clear, if not just let me know.
>>>>
>>>> Best,
>>>>
>>>> Yari
>>>>
>>>>
>>>>
>>>> -------------------------------------------------------------------
>>>> -
>>>> -
>>>> -
>>>> -------- _______________________________________________
>>>> mitk-users mailing list
>>>> [email protected]
>>>> https://lists.sourceforge.net/lists/listinfo/mitk-users
>>> --------------------------------------------------------------------
>>> -
>>> -
>>> -------- _______________________________________________
>>> mitk-users mailing list
>>> [email protected]
>>> https://lists.sourceforge.net/lists/listinfo/mitk-users
>> ---------------------------------------------------------------------
>> -
>> -------- _______________________________________________
>> mitk-users mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/mitk-users
>
> ----------------------------------------------------------------------
> -------- _______________________________________________
> mitk-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/mitk-users


------------------------------------------------------------------------------
_______________________________________________
mitk-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mitk-users

------------------------------------------------------------------------------
_______________________________________________
mitk-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mitk-users

Reply via email to