<html><head></head><body><div style="color:#000; background-color:#fff; font-family:Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif;font-size:16px"><div id="yui_3_16_0_ym19_1_1487327948470_27929">Hello,</div><div id="yui_3_16_0_ym19_1_1487327948470_28020"><br></div><div dir="ltr" id="yui_3_16_0_ym19_1_1487327948470_28021">I use GATE in GPU for PET imaging. So I want to ask you, if we reconstruct the image (root file) that is extracted, without changing the code, will we produce the real source or just the box seen in root?</div><div dir="ltr" id="yui_3_16_0_ym19_1_1487327948470_28304"><br></div><div dir="ltr" id="yui_3_16_0_ym19_1_1487327948470_28311">Thank you in advance,</div><div dir="ltr" id="yui_3_16_0_ym19_1_1487327948470_28303">Konstantinos<br></div><div id="yui_3_16_0_ym19_1_1487327948470_27928"><span></span></div><div id="yui_3_16_0_ym19_1_1487327948470_28027"> </div><div class="signature" id="yui_3_16_0_ym19_1_1487327948470_28026"><div id="yui_3_16_0_ym19_1_1487327948470_28025"><span style="font-style:italic;" id="yui_3_16_0_ym19_1_1487327948470_28545">-----------------------------------</span></div><span style="font-style:italic;" id="yui_3_16_0_ym19_1_1487327948470_28307">Konstantinos Chatzipapas<br>Medical Physicist, </span><span style="font-style:italic;" id="yui_3_16_0_ym19_1_1487327948470_30422">MSc<br></span><div id="yui_3_16_0_ym19_1_1487327948470_28306"><div id="yui_3_16_0_ym19_1_1487327948470_28305"><span style="font-style:italic;" id="yui_3_16_0_ym19_1_1487327948470_28544">University of Patras</span></div><div id="yui_3_16_0_ym19_1_1487327948470_28531"><div id="yui_3_16_0_ym19_1_1487327948470_28530"><span style="font-style:italic;"><br></span></div><div id="yui_3_16_0_ym19_1_1487327948470_28543"><span style="font-style:italic;" id="yui_3_16_0_ym19_1_1487327948470_28542">email: con@physicsmag.com</span></div><div id="yui_3_16_0_ym19_1_1487327948470_28541"><span style="font-style:italic;" id="yui_3_16_0_ym19_1_1487327948470_28540">web page: www.physicsmag.com</span></div></div></div></div> <div class="qtdSeparateBR"><br><br></div><div class="yahoo_quoted" style="display: block;"> <div style="font-family: Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif; font-size: 16px;"> <div style="font-family: HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif; font-size: 16px;"> <div dir="ltr"><font size="2" face="Arial"> Στις 6:16 μ.μ. Τετάρτη, 11 Ιανουαρίου 2017, ο/η Liliana Caldeira <llcaldeira@gmail.com> έγραψε:<br></font></div>  <br><br> <div class="y_msg_container"><div id="yiv5453314868"><div><div dir="ltr"><div><div><div><div>Hi Mirco,<br clear="none"><br clear="none"></div>The GATE GPU version does not save the source position, but only the position when it leaves the voxelized phantom, thus why you only see the faces of the phantom. <br clear="none"></div>If you are really interested in the source positions, then you can change the GPU code to output that information to a file (although it might be very large txt file). <br clear="none">In the file GateEmisTomo.cc:<br clear="none"><br clear="none">kernel_voxelized_source_b2b<<<grid, threads>>>(gamma1_d, gamma2_d, activities_d, E,<br clear="none">                                                   phantom_d.size_in_vox, phantom_d.voxel_size);<br clear="none">    cudaThreadSynchronize();<br clear="none">    printf(" ::   Generation ok\n");<br clear="none"></div>PRINT INFO HERE: gamma1_d.eventID gamma1_d.px gamma1_d.py gamma1_d.pz<br clear="none"></div><div><br clear="none">Cheers,<br clear="none"></div><div>Liliana<br clear="none"></div><div><div><div>Forschungszentrum Jülich GmbH<br clear="none"></div><div><br clear="none"></div></div></div></div><div class="yiv5453314868gmail_extra"><br clear="none"><div class="yiv5453314868gmail_quote">On 10 January 2017 at 14:11, Mirco Heß <span dir="ltr"><<a rel="nofollow" shape="rect" ymailto="mailto:mirco.hess@uni-muenster.de" target="_blank" href="mailto:mirco.hess@uni-muenster.de">mirco.hess@uni-muenster.de</a>></span> wrote:<br clear="none"><blockquote class="yiv5453314868gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"><div class="yiv5453314868yqt1019961395" id="yiv5453314868yqt31311"><div dir="ltr">Hi all,<div><br clear="none"></div><div>I am trying to run the GPU PET example but am facing some problems. When looking at the output, it seems like the XYZ positions of the source are accumulated on the faces of the provided phantom/source volume, i.e. the bounding box of the 3D image. This is only true for the GPU-based simulation. When running the CPU-based simulation everythig seems to be fine. I have attached two screenshots to visualize the problem and the difference between the two cases. The problem is present in both the ASCII and ROOT output.</div><div><br clear="none"></div><div>I already had this problem about two years ago with Gate 7.0 and was not able to resolve it.</div><div>Now I am running Gate 7.2 with GPU support activated, on Ubuntu 14.04 with updated NVIDIA and CUDA drivers. </div><div><br clear="none"></div><div><br clear="none"></div><div>Has anybody been able to successfully run this example?  (gate_v7.2/examples/ gpumacros/pet/main-gpu.mac)</div><div><br clear="none"></div><div>Any advice and pointers to resolve this issue would be highly appreciated. Thanks a lot in advance!</div><div><br clear="none"></div><div><br clear="none"></div><div>Cheers, </div><div>Mirco</div><br clear="all"><div><br clear="none"></div>-- <br clear="none"><div class="yiv5453314868m_-4465431999979118705gmail_signature"><div dir="ltr"><font size="-1"><b>Mirco Heß</b><br clear="none">
        PhD Candidate<br clear="none">
        European Institute for Molecular Imaging (EIMI)<br clear="none">
        
        </font><div><img src="cid:1DX5HEEBKlUIksuMbSu4" alt="_European Institute for Molecular Imaging - EIMI_" height="70"></div><div><font size="-1">European Institute for Molecular Imaging - EIMI<br clear="none">
        Waldeyerstr. 15<br clear="none">
        48149 Münster<br clear="none">
        <a rel="nofollow" shape="rect" target="_blank" href="http://www.uni-muenster.de/eimi">http://www.uni-muenster.de/ eimi</a><br clear="none">
        <br clear="none">
        </font><font size="-1"><span><a rel="nofollow" shape="rect" href="">+49 (0)251 83 - 49318</a></span><br clear="none">
        <a rel="nofollow" shape="rect" href="">+49 (0)251 83 - 49313</a> (Fax)<br clear="none">
        eMail: <a rel="nofollow" shape="rect" ymailto="mailto:mirco.hess@uni-muenster.de" target="_blank" href="mailto:mirco.hess@uni-muenster.de">mirco.hess@uni-muenster.de</a></font></div></div></div></div></div>
<br clear="none">______________________________ _________________<br clear="none">
Gate-users mailing list<br clear="none">
<a rel="nofollow" shape="rect" ymailto="mailto:Gate-users@lists.opengatecollaboration.org" target="_blank" href="mailto:Gate-users@lists.opengatecollaboration.org">Gate-users@lists. opengatecollaboration.org</a><br clear="none">
<a rel="nofollow" shape="rect" target="_blank" href="http://lists.opengatecollaboration.org/mailman/listinfo/gate-users">http://lists. opengatecollaboration.org/ mailman/listinfo/gate-users</a><br clear="none"></blockquote></div><br clear="none"></div></div></div><br><div class="yqt1019961395" id="yqt71055">_______________________________________________<br clear="none">Gate-users mailing list<br clear="none"><a shape="rect" ymailto="mailto:Gate-users@lists.opengatecollaboration.org" href="mailto:Gate-users@lists.opengatecollaboration.org">Gate-users@lists.opengatecollaboration.org</a><br clear="none"><a shape="rect" href="http://lists.opengatecollaboration.org/mailman/listinfo/gate-users" target="_blank">http://lists.opengatecollaboration.org/mailman/listinfo/gate-users</a></div><br><br></div>  </div> </div>  </div></div></body></html>