[Rtk-users] SPECT reconstruction issue

[Simon Rit]

Please stick to the mailing list. The DivideImageFilter is part of OSEM so
that doesn't explain it. I did a quick test and that works for me:
rtksimulatedgeometry -n 100 -o g
rtkprojectshepploganphantom -g g -o proj.mha --spacing 1.5
rtkosem -p . -r proj.mha -v -o osem.mha -n 5 -g g --nprojpersubset 100
and I obtain 5.7e6 in the osem.mha file and 5.9e8 in the proj.mha file.

On Tue, Jun 27, 2023 at 4:07 PM Nathan Sinsoilliez <n.sinsoilliez at gmail.com>
wrote:

> Hi,
> I'm not using any attenuation correction currently. The total counts of
> the reconstructed images is still too low compared to the total average in
> one projection. It looks like the pixel values are divided (which can
> explain why they're decimals). After checking the
> OSEMConeBeamReconstructionFilter on the wiki, I saw that there is a
> DivideImageFilter, maybe it can explain why but I don't really get how it
> works and why it is necessary so maybe I'm speaking nonsense here.
>
> Sinsoilliez Nathan
>
>
>
>
> Le mar. 27 juin 2023 à 10:18, Simon Rit <simon.rit at creatis.insa-lyon.fr>
> a écrit :
>
>> Hi,
>> I'm not sure we actually ever checked this but it's a good check. Are you
>> using attenuation correction? If not, I expect that the total number of
>> counts in one projection should be equal to the integral of the SPECT image
>> as this is the Radon transform model. So you may want to average the number
>> of counts and not sum it (i.e., divide 8E+6 by your number of projections).
>> Can you check this?
>> Simon
>>
>> On Tue, Jun 27, 2023 at 8:48 AM Nathan Sinsoilliez <
>> n.sinsoilliez at gmail.com> wrote:
>>
>>> Hello everyone,
>>>
>>> I'm currently using rtk to reconstruct SPECT images using OSEM and I
>>> have an issue after the reconstruction. The total count of the image is
>>> completely changed (like 8E+6 in the projections VS 2E+4 when
>>> reconstructed). Also most of the pixels now have decimal values.
>>> Graphically everything seems to be alright though.
>>>
>>> Do you have an idea about what could cause this problem ?
>>>
>>> Best Regards
>>> Sinsoilliez Nathan
>>> _______________________________________________
>>> Rtk-users mailing list
>>> rtk-users at openrtk.org
>>> https://www.creatis.insa-lyon.fr/mailman/listinfo/rtk-users
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.creatis.insa-lyon.fr/pipermail/rtk-users/attachments/20230627/298f6424/attachment.htm>