[Insight-users] Interpolation Volume

[Luis Ibanez]

Hi Gupt,

We no longer make any difference between a sequence of 2D slices
and a 3D volume... unless.... they are not aligned, or... the spacing in
the inter-slice direction is more than 5 times the intra-slice spacing.


If you have segmentations for all the 2D slices, you can simply load them
all into a single 3D image using an ImageSeriesReader, and then you
can process it as a 3D volume.  More specifically, you can proceed to
compute its volume by using the LabelStatisticsImageFilter.


     Regards,


           Luis


--------------------------------
On Mon, Dec 7, 2009 at 12:10 PM, Aditya Gupta <adityargupta at gmail.com> wrote:
> Hello Luis,
>
> Thanks for the reply. In continuation to my previous question... So using
> the stack of 2d MRI slices, I've a certain segmentation on each of the
> slices and using the segmented output I've built a surface model and now
> I've incorporated the intensities of the slices in the surface model. Now,
> I'm interested to find the volume of the segmented region. So, using the
> segmented outputs from the various slices, I want to reconstruct a volume,
> compute and finally visualize the volume. Can you give me a start on how I
> can proceed? Is it possible to build a 3d volume (not surface model) from a
> stack of 2d slices (only one axial view not 3 views)? Are there any standard
> algorithms being used for construction of 3d volumes from 2d slices?
>
> Appreciate your help in this matter.
>
> Regards,
> Gupt
>
> On Wed, Dec 2, 2009 at 7:07 PM, Luis Ibanez <luis.ibanez at kitware.com> wrote:
>>
>> Hi  Gupt,
>>
>>
>> Yes, just do the following:
>>
>>
>> A) Load the 2D slices into a single 3D image using the
>>     itk::ImageSeriesReader,
>>
>>
>> B) Connect the image an ImageInterpolator
>>     The typical choice is the itkLinearImageInterpolateFunction class,
>>     (but you can also use NearestNeighbors, BSpline and
>>      WindowedSinc, it is a trade-off between computation time and
>>      interpolation precision)
>>
>>
>> C) Take all the points from your surface model and for every
>>     point call the method:
>>
>>       PixelType  pixelValue =    Interpolator->Evaluate( point )
>>
>>    and then assign that pixel value to the corresponding point
>>    in the surface model.
>>
>>
>>
>>  Please let us know if you run into any problem.
>>
>>
>>       Thanks
>>
>>
>>          Luis
>>
>>
>> ------------------------------------------------------------------
>> On Wed, Dec 2, 2009 at 2:35 PM, Gupt <adityargupta at gmail.com> wrote:
>> >
>> > I've a stack of 2d cardiac mri images in single view. I've written the
>> > code
>> > to do a surface reconstruction of the left ventricle wall of the heart
>> > (looks like a hollow cylinder). Is there a method/ function to
>> > incorporate
>> > the intensity information (in the 2d slices) in to the model, like some
>> > kind
>> > of interpolation to obtain intensity between the slices? Thanks for the
>> > help.
>> > --
>> > View this message in context:
>> > http://old.nabble.com/Interpolation-Volume-tp26614974p26614974.html
>> > Sent from the ITK - Users mailing list archive at Nabble.com.
>> >
>> > _____________________________________
>> > Powered by www.kitware.com
>> >
>> > Visit other Kitware open-source projects at
>> > http://www.kitware.com/opensource/opensource.html
>> >
>> > Kitware offers ITK Training Courses, for more information visit:
>> > http://www.kitware.com/products/protraining.html
>> >
>> > Please keep messages on-topic and check the ITK FAQ at:
>> > http://www.itk.org/Wiki/ITK_FAQ
>> >
>> > Follow this link to subscribe/unsubscribe:
>> > http://www.itk.org/mailman/listinfo/insight-users
>> >
>
>
>
> --
> Regards,
> Aditya Gupta, Ph.D.
> Post Doc Researcher
> UCF Computer Vision Lab
> Orlando, FL
> O: (407) 882 0130
> M: (407) 451 7735
>