smc_linescan missing from directroy in rti files

[JerePhys]

Hello,

I am trying to directly implement a new Treatment Machine by following the step-by-step guide provided on this GitHub. However, I am encountering an issue when I try to run a test with the command line tool (https://github.com/topasmc/dicom-interface/wiki/test-with-cli).

When trying to run the CMakeLists file in rti/test/session/, I receive an error stating there is no source file named smc_linescan.cpp.
Realizing there is indeed no such file, I therefor commented out the code in the CMakeLists file referencing smc_linescan.

Then, when I try to execute the CMakeLists file, I receive an error message to do with missing libraries ("lgdcmexpat", "lgdcmzlib", "lgdcmopenjpeg", "lgdcmcharls", "lgdcmuuid"). Indeed, these libraries don't seem to be among those that I have downloaded (such as "libgdcmCommon.so" and so on).

If I try to directly compile the treatment_session.cpp file by using the command "g++ -o treatment_session treatment_session.cpp",
I get a similar error.

I downloaded the RTI files directly from this GitHub a few days ago and my current version of GDCM is 3.0.24 and TOPAS 3.9.

I am feeling quite stumped, not knowing whether my problem is the fact that I am missing the referenced smc_linescan file or that I might be missing certain GDCM libraries. Or is it to do with how the files are installed on my PC?

Any input or advice would be welcome.

Thanks in advance for your time and effort,

Jeremy

[jungwookshin]

Hi Jeremy,

smc_linescan should be commented out from CMakeLists.txt as it was for other sites. Speaking about gdcm, I recommend to use gdcm.2.6.8 which TOPAS was linked to and came with TOPAS package. so as long as you are building this extension with pre-built topas package, you should use 2.6.8 version not from your own.

[JerePhys]

Hello Jungwook. Thank you for your insight.

I have changed to gdcm.2.6.8 libraries and am able to run the CMakeList to compile the treatment_session.cpp file. However, I can only accomplish the "test direct class" step (https://github.com/topasmc/dicom-interface/wiki/test-with-cli#test-direct-class) on the condition that I slightly alter a section of code in rti_dataset.hpp.

In said .hpp code, I was getting an error to do with the Value Multiplicity having a length of zero for my own RT Ion plan, but also for rti_demo.dcm. As I was troubleshooting, I simply changed "assert(ndim>=1)" to "assert(ndim>=0)" and it permitted me to run the "test direct class" step without errors.

I do, however, end up with a few unrecognized values in the output of that test :
"
...
SpecificCharacterSet: (0008,0005), VR: ??, length: 10
InstanceCreationDate: (0008,0012), VR: ??, length: 8
InstanceCreationTime: (0008,0013), VR: ??, length: 6
SOPClassUID: (0008,0016), VR: ??, length: 30
SOPInstanceUID: (0008,0018), VR: ??, length: 58
...
"

This issue does not prevent me from running simulations from TOPAS, but I am worried it might signify problems down the road. For example, I am worried that my GDCM reader might not be correctly reading the inputted RT Ion plans which may be leading to these complications.

I'd be thankful to hear your intuition on the underlying issue in this situation.

Thank you again for your time,

Jeremy

[jungwookshin]

Hi Jeremy,

Oh, I remember the VM error. I think I did something similar when I encountered the same thing. It worked one dataset but didn't for other dataset, etc. I haven't come with the best solution to cover both cases. As long as you see histories are generated, I'm sure you are fine. Besides, we are not utilizing the other unrecognized DICOM tags anyways in our simulations. sorry for your inconvenience. I wish I had more time to work on fixing the details but haven't have any time yet... but feel free to post here or sending me an email once you have any problems. thanks and good luck!

[JerePhys]

Hello Jungwook,

I am very thankful for your help. My RT Ion setup is now up and running. If ever the need arises, I will gladly reach out again.

Until then, best regards,

Jeremy