Hi Johanna, thanks for the details.
It sounds like what you have done is partly work that we have planned (but not for the 1.0 release):
Issue 862 will be essential for a lot of things, including DRL comparisons, more useful charts, generating data for national data collection etc. It sounds like you have approached the same issue via a different method!
I've
created a new issue for extracting BodyPartExamined from DX files, though it isn't a field that I have seen in RDSR files for DX exposures. Which table did you put the field in?
For the DRL comparison, if I have understood correctly you are comparing individual dose metrics to the population level average/reference level? I am not comfortable with this as a general approach - I prefer the approach of issue 598 where the DRL generated from your dataset can be compared to the national one. Alternatively, in our institution we generate charts of the normal range of dose metric values against patient weight or breast thickness, so that individual exposures can be usefully compared to spot outliers that need investigation. I haven't yet devised a way of working this into the interface though!
For you to take your changes through to a version 1 release, you will have to reimplement a lot of it as the code has been migrated from Python 2 to Python 3, and there have been quite a few changes as well. For the changes that might be useful to the community it would be really nice if you would try and make the changes against the current code in a fork of the OpenREM repository and submit a pull request so we can all benefit - you will benefit too as you won't face this issue at upgrade time!
For changes that you want to make just in your install, this will be possible for an advanced user such as yourself. I will be writing how in the documentation. Essentially it will involve building your own docker image.
Kind regards
Ed