Datasets are still valuable for the broader community - there are many datasets out there, like the Brain project or the Oasis project which provide important insights into brain illnesses to train state of the art ML models on.
Any practitioner who wishes to work on those datasets is to sign a contract where they pledge not to try to identify any patient.
These kinds of contracts have been around for a while now, and if we could negotiate a situation as tricky as brain illnesses, I’m certain the same can be done for which version you have of a software, which particular config changes one has made or, say, if you use KActivities at all.
Datasets are still valuable for the broader community - there are many datasets out there, like the Brain project or the Oasis project which provide important insights into brain illnesses to train state of the art ML models on.
Any practitioner who wishes to work on those datasets is to sign a contract where they pledge not to try to identify any patient.
These kinds of contracts have been around for a while now, and if we could negotiate a situation as tricky as brain illnesses, I’m certain the same can be done for which version you have of a software, which particular config changes one has made or, say, if you use KActivities at all.