[Insight-users] Creating 3D images based on slices

[Wagner Sales]

Hi Luis,

1 - I have three series of slices. Considering a human head, for example:
       a) one serie starting at the top of the read, uniformly spaced.
       b) one serie from the left to the rigth, uniformly spaced.
       c) one serie from the left the front side to the back side, 
uniformly spaced.
       In other words, the z-axis will change.
2 - Because the spaces are to me, big (images 512x512, only 20 images ) 
I was thinking in use the three series to obtain more accuracy in the 
generated volume. The interpolation can be maked in this case, to have 
better accuracy or not?
Now I and my colleague are testing your tips using the reader whith just 
one serie.
One more thing. Some example or tip to put memory->Image<Image3D>?

PS: I don't now how you have time to develop and solve our questions. :)

thkns
Luis Ibanez wrote:

>
> Hi Wagner,
>
> If you already have the data in memory, then
> the only issue is to perform interpolation.
>
> (Assuming that the slices are actually not
>  spaced uniformly)... or are they uniformly
> spaced along Z ?
>
> If the space if uniform you don't need any
> interpolation...
>
> Otherwise, You may want to look at the BSpline
> interpolators in ITK as well as the KernelBase
> splines.
>
>
>    Regards,
>
>
>       Luis
>
>
> -------------------
> Wagner Sales wrote:
>
>> Hi Luis,
>>
>> Luis Ibanez wrote:
>>
>>>
>>> Hi Wagner,
>>>
>>> Are your slices being stored in independent files ? 
>>
>>
>>
>> Here is the problem... Are not! But I have the data in memory, but I 
>> can convert to every format. In last case, pixel by pixel... ;)
>> I have the position z-axis position too... Well, if have like do this 
>> withou using the serie readers, like in VTK ( vtk have one too ), 
>> better to me. But if not have, or are must difficult, 'll change my 
>> program... no problem.
>> Thks.
>>
>>>
>>>
>>> If so, you can simply use the ImageSeriesImageReader.
>>> You will find examples in the directory:
>>>
>>>        Insight/Examples/IO
>>>
>>> they are:
>>>
>>>        DicomSeriesReadImageWrite.cxx
>>>        ImageReadImageSeriesWrite.cxx
>>>        ImageSeriesReadWrite.cxx
>>>        RGBImageSeriesReadWrite.cxx
>>>
>>>
>>>
>>> This will create a 3D image with the collection of
>>> 2D image from your files.
>>>
>>> It is still *your* responsibility to setup the Z
>>> pixel spacing, since the reader has no other way to
>>> figure out this value. (DICOM probably can...).
>>>
>>> You can use the itkChangeInformationImageFilter for
>>> the purpose of setting such pixel spacing.
>>>
>>> There is no interpolation involved in this process.
>>> The 3D image will be just a concatenation of slices.
>>>
>>>
>>> If your slices do not correspond to a uniform
>>> sampling along Z.... then.... you are in trouble...  :-/
>>>
>>> Are they at least parallel  ? ...
>>>
>>> The brute force approach will be to use KernelSplines
>>> in order to interpolate intensity values... that will
>>> work... but is not going to be fast.
>>>
>>>
>>>    Regards,
>>>
>>>
>>>       Luis
>>>
>>>
>>>
>>> ------------------------
>>> Wagner Sales wrote:
>>>
>>>> Hi,
>>>>
>>>> First, thank's by the last help. like you all probabily know, I 
>>>> need to use a volume in ITK. You told me to use a itkImage with 
>>>> dimension = 3. I looked in the examples, and i'm convinced that 
>>>> that's the right way to do it. But, now I have a different 
>>>> problem... like I said before, I need to fill this 3-dimensional 
>>>> itkImage with n 2-dimensional images. So, my current problem is how 
>>>> o put n differents images into the 3D image, and, furthermore, how 
>>>> to specify the coordinates of the images in the 3D one (the 
>>>> z-axis). After that, I need to interpolate the images, because my 
>>>> slices aren't contiguous, I was looked how to set the space between 
>>>> the images. Are just that? I don't need to specify the interpolation?
>>>>
>>>> Thanks,
>>>>
>>>> Wagner
>>>>
>>>>
>>>> _______________________________________________
>>>> Insight-users mailing list
>>>> Insight-users at itk.org
>>>> http://www.itk.org/mailman/listinfo/insight-users
>>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Insight-users mailing list
>>> Insight-users at itk.org
>>> http://www.itk.org/mailman/listinfo/insight-users
>>>
>>
>>
>> _______________________________________________
>> Insight-users mailing list
>> Insight-users at itk.org
>> http://www.itk.org/mailman/listinfo/insight-users
>>
>
>
>
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users
>