[Gate-users] Image Reconstruction by MLEM

[Simon Stute]

Hi,

The reason is most likely the lack of normalization of your sinogram data.

In a realistic GATE simulation, even without independent crystal 
efficiencies, each LOR has a different sensitivity, mostly due to solid 
angles and energy threshold. These effects are due to the geometry of 
the scanner and reflect this geometry in the sensitivity image and in 
the reconstructions. So this should be corrected for through a 
normalization scan.

Check this IEEE MIC conference proceeding for more information: 
https://ieeexplore.ieee.org/document/6153804

All the best !
Simon

Le 01/01/2019 à 04:14, Yanchao Huang a écrit :
> Dear Gate users,
> Recently, I simulated a small animal PET scanner by GATE 8.0. The 
> scanner consists of 60 block-detectors in 5 contiguous rings. Each 
> detector consists of 13X13 LYSO crystals and  each pixel size is 
> 1.8X1.8X15 mm.
> Firstly，an uniform cylindrical source was simulated and the list-mode 
> data in ASCII format  was acquired.
> Secondly, I reorganized the data onto histgram. Then SSRB was applied 
> to gain a 2-D data set. Thirdly, I applied the 2-D ray tracing 
> algorithm to generate the system matrix simply. Each line of 
> response(LOR) was connected with the center of each crystal. And the 
> intersection length of the image pixel and the LOR was considered as 
> the detected probability. However, a high-light artifact showed in the 
> MLEM reconstructed image.
> I think this error may come from the sensitivity map which also shows 
> a spider-web shape distribution due to the scanner geometric 
> structure. But I don't know how to deal with it. Does any one has ever 
> meet such case? I‘m very much looking for your help and suggestions.
> By the way, Happy New Year to your all!
> Best wishes!
>
> _______________________________________________
> Gate-users mailing list
> Gate-users at lists.opengatecollaboration.org
> http://lists.opengatecollaboration.org/mailman/listinfo/gate-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengatecollaboration.org/pipermail/gate-users/attachments/20190104/d135bfff/attachment.html>