Elekta MR Linac Unity MeterSetMap comparison

1

Hello,

I’m interested to compare dicom RT plan from Monaco with TRF files for the MR Linac Unity from Elekta.

In the past I have adapted a development version of pymedphys with some trf decoding to make the mudensity tool work.

image
image586×739 106 KB

But when I now try to do this in the current version of pymedphys gui MeterSetMap comparison I got errors.

The first error is that the problem of the expected MLC positions is different for MR Linac.
See second past for screenshot

The second error has to do with the different format of the TRF file. I think i have some code available to fix this problem.
See second post for screenshot.

Best Regards,
Martijn Kusters

2

The first error is that the problem of the expected MLC positions is different for MR Linac.
See second past for screenshot.

pymedphys
pymedphys1463×863 233 KB

3

The second error has to do with the different format of the TRF file. I think i have some code available to fix this problem.

See this post for screenshot.

pymedphys2
pymedphys21840×932 329 KB

4

@sjswerdloff might you in a position to give @Martijn_Kusters a helping hand for him to potentially create a PR that addresses these issues?

5

Sure.
I will get started on it…

6

Dear Stuart,

I made some changes in the code to make this work. Shall I share my changes in code?

reportnew
reportnew993×1251 138 KB

7

Hi Martin,

Short answer: Yes! Working Code is the best.

I’m sorry for the delay.
I believe you have already posted in #1285 in the pymedphys GitHub issues list.

The ideal way to provide code changes is to create a Pull Request.
The way I got started was to do a social fork, create a branch, and then create a pull request in pymedphys that pointed to that branch.
On the web version of GitHub there is a fork button at top right side of the web page for the pymedphys project.
If you have access to creating a branch in the pymedphys project itself, then it’s even easier.

If you haven’t already gotten yourself set up to be a contributor:
https://docs.pymedphys.com/contrib/index.html

Once you have a PR constructed, you can put me and Simon in as reviewers.

You’ve already posted some anonymised data. If you put that in zenodo (the mechanism we have been using for storing data for automated tests) then you can add a test that “demonstrates” your fix/change works (and previous tests will ensure that nothing got broken).

I’m not sure how familiar you are with any of the tools, but get as far as you can with “reasonable” effort and if/when you hit a bump, just let us know and we (I) will try to help you.

Stuart