[Insight-users] Algorithm to resample anisotropic volume

[Mauro Maiorca]

Dear Darren,

I randomly came through this old post, since I'm dealing with a
similar problem (and nobody replied to it) I'm just wondering if you
have already got a solution to it.

In my opinion (please, correct me if I'm wrong), "avoiding severe
anisotropy of pixel dimensions in 3D volumes" (by interpolation,
inpainting, etc) without prior knowledge of the structure to analyse
is an artifact driven process: if the slice thickness is 0.5um, then
you can only image 3D features above the ~ 2*0.5um size (because of
Nyquist). Anything below that range is a mere "guess". Well, I do
believe that when you got the XY resolution *way* greater then the
slice thickness the "guess" is quite accurate (whether a good
interpolation is used), however it is still a guess, and it's hard to
convince people that your guess actually makes sense.  Also, when you
are dealing with organelles' ultrastructure images it seems really
hard to provide a statistical/mathematical model due to the high
variability in size/shape (and the underlying lack of knowledge).

The only solution that makes sense to me is adding information by
using correlative/multimodal imaging techniques merged togethe
r(afterall, it is what they do in the article you pointed out).

I'm looking forward to hearing your opinion.

Cheers,
Mauro

On Fri, Oct 2, 2009 at 11:03 AM, Darren Weber
<darren.weber.lists at gmail.com> wrote:
>
>
> On Thu, Oct 1, 2009 at 5:42 PM, Darren Weber <darren.weber.lists at gmail.com>
> wrote:
>>
>> Dear Luis et al.,
>>
>> The ITK software guide has some good warnings about avoiding severe
>> anisotropy of pixel dimensions in 3D volumes (pp. 242-247 in printed book).
>> I have a situation where microscopy has been used to obtain digital images
>> of sections that were sliced by microtome at 0.5 micrometers thickness (Z),
>> with the XY pixel spacing at about 0.055 micrometers.  I'm told that it's
>> about the best that can be done with this method of imaging.
>>
>> In early work MRI volumes, there was an interesting method proposed by
>> Dale & Sereno (1993; see pp. 8-9, esp. Fig. 1, and Appendix B on p. 17).  Is
>> there an algorithm in ITK that implements this sort of linear algebra?
>>
>> Download a PDF of this paper here:
>> ftp://ftp.buckinstitute.org/dweber/MEG_EEG_source_DaleSereno1993.pdf
>>
>> Kind regards,
>> Darren
>>
>
>
> @ARTICLE{DaleSereno1993,
>   author = {Dale, Anders M. and Sereno, Marty I. },
>   title = {Improved localization of cortical activity by combining {EEG} and
>         {MEG} with {MRI} cortical surface reconstruction: A linear
> approach},
>   journal = {J Cogn Neurosci},
>   year = {1993},
>   volume = {5},
>   pages = {162--176},
>   number = {2},
> }
>
>
> _____________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.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
>
>