Posts by gordon

  • Lua crashes when doing dicommove on missing/faulty studies

    Hi Marcel,


    after some more digging I found that I hadn't enabled JPEG(2000) support - once I had done this it seems to work perfectly. Silly me, but perhaps this can be the basis for a more informative transfer syntax error? How about: "No valid presentation contexts/transfer syntax found in 1 candidates (make sure that you have enabled JPEG 2000 support)".


    Thanks again for your time and this awesome software.


    Best
    Max


    PS. I found the ARCNEMA.MAP documentation a little hard to grasp. Perhaps adding a few examples would help, for instance can one have more than one compression alternative set? DS

  • Lua crashes when doing dicommove on missing/faulty studies

    Yes, an IP-timeout would make sense. There seem to be 3 error messages before it continues:



    It would be nice to shorten the pain to 1 failed call and then return to the script to try next item, or as I previously mentioned - try to design a dicomquery that avoids these particular images.


    Best
    Max

  • Lua crashes when doing dicommove on missing/faulty studies

    Unfortunately I need to run the software on a Windows server. I'm not that familiar with all the inner workings of setting up a Visual Studio project, but it would be nice if there was a Github-repository that one could clone and try some exploring.


    I've talked to the folks here at the IT PACS department and the (J) marker that I've found for the images causing the trouble stands (as expected) for JPEG-compression. It is a little odd that the PACS doesn't send unencrypted data as specified in the ACRNEMA.MAP. I think it could be possible to avoid these images by having a look at the TransferSyntaxUID (0002,0010), but I can't find it among the newdicomobject() alternatives. I see that it's among the dgate.dic keywords (https://github.com/marcelvanhe…03e2e8f293b/dgate.dic#L71) but I haven't quite grasped how the options for the newdicomobject() are initiated (https://github.com/marcelvanhe…b/master/luaint.cpp#L1029). Is there a way to directly specify what header element to retrieve?


    I'm also beginning to suspect that the dicommove actually doesn't crash, but takes forever to complete. Is there a way to set a time-out after 20 minutes?


    Best
    Max

  • Lua crashes when doing dicommove on missing/faulty studies

    There should not be any movies, there may be CT/MRI-scans but the images that have failed are generally plan x-ray images. When I've tried to locate them in the regular viewer they've appeared missing or with some odd feature attached to the (like the J-sign I wrote about).


    When I do the print(Global.StatusString) I get "dicommove: 0 of 1 images transferred". After the transfer syntax error I now also get "dicommove: remote DICOM error". There is currently also some kind of additional problem on the IntelliSpace side that they're trying to fix, that appeared though yesterday afternoon and I'm not sure that its related.


    By the way, it would be nice to know how to compile my own version of ConQuest in case I wanted to tinker around a little. The provided source code seems incomplete and your GitHub (https://github.com/marcelvanherk/dgate) hasn't been updated for 2 years. Where can I find the complete code with all the necessary packages? From what I understand you are using Visual Studio and since the 2013 Community is freely available it would be nice to be able to open the code and have a closer look at it there.


    Thanks by the way for an awesome program!

  • Lua crashes when doing dicommove on missing/faulty studies

    I'm trying to move a study from a IntelliSpace PACS server to Conquest. Unfortunately I run into a somewhat uninformative transfer syntax error:


    Quote


    2015-05-05 10:43:02 [CONQUESTSRV1] ***No valid presentation contexts/transfer syntax found in 1 candidates
    2015-05-05 10:43:02 [CONQUESTSRV1] ***In 1 presentation contexts
    2015-05-05 10:43:02 [CONQUESTSRV1] ***#Possible transfer syntaxes: 1
    2015-05-05 10:43:02 [CONQUESTSRV1] *** multiplex: connection terminated


    The Lua-code that causes is this rather straight forward:


    Code
    local cmd = newdicomobject()cmd.QueryRetrieveLevel = 'STUDY'cmd.StudyInstanceUID= sUIDdicommove(srcAet, destAet, cmd)


    Not sure if this has to do with the compression format, it is currently set to uncompressed according to the ACRNEMA.MAP (IP-addresses changed):


    Code
    CONQUESTSRV1                  1.1.1.1         5678		unCONQUESTSRV2                  1.1.1.1         5679		unINTELLISPACE_QRP              1.1.1.2         200	 	unINTELLISPACE_SU               1.1.1.2         200	 	unV*                            *               1234      unW*                            *               666       unS*                            *               5678      un


    I believe that there may be some issue with this particular study and if possible I would like to have an assert:

    Code
    assert(dicommove(srcAet, destAet, cmd), "Failed to move " .. sUID)

    Where I allow the script to skip this particular study with a mere warning. Unfortunately Lua crashes despite the assert when this happens and I need to stop the script and restart.


    I've looked into what studies are causing the move to crash and so far I've found that they seem to be correlated with odd (faulty?) or missing studies. In one study it crashed on an image that appears with a (J) sign when viewing it within the IntelliSpace viewer while another one wasn't available from the source. I've tried to look at the dicomquery() to get some kind of parameter that would hint that the study should not be retrieved but so far no luck...


    Any ideas how to tackle this issue? Where can I find the dicommove in the source code, how do I build it (Windows), and is there a way to change its behavior so that it returns a 0 on failure?