Open raamana opened 2 years ago
what specifically do you want from it? couldn't it just be on phantoms? then may be http://datasets.datalad.org/?dir=/dbic/QA ?
thanks Yarik! this was an old issue I opened as a reminder, to include them for analyses in the paper. We (@sinhaharsh ) did find some open DICOM datasets, although outside neuro of other organs e.g., from the https://www.cancerimagingarchive.net/
yes, phantoms or not does not matter! Now that I think about it, we didn't give it too much thought about the impact of lack of protocol compliance for Phantom/QA data collection. Off the top of my head, I think most of the guidelines for human data apply although with few parameters being not so important and can have larger variation within a project without much impact on the final result. this is just my intuition, and it needs to be properly studied and tested
preferably a neuroimaging dataset