Just to echo some of what has been said already, I'd suggest looking into the BIDS scheme developed by the Stanford Reproducible Neuroscience folks. Most of the previous guidance on this subject was put forth by the same people who worked on BIDS, so if that isn't what you're looking for exactly, I'd dig into their work a bit more.
Here is the initial BIDS paper: https://www.nature.com/articles/sdata201644
Here is some broader guidance from the same team: http://www.nature.com/neuro/journal/v20/n3/abs/nn.4500.html
To provide a little more context, I was a neuroimaging researcher before I started working in libraries and got interested in data management-related problems. The field has historically not been quick to adopt standards related to data management or storage. Even BIDS, which is probably the closest to what you are looking for, is mostly concerned with data sharing and not preservation and archiving. Also, though a couple of journals recommend BIDS, it is not in very widespread use.
I'm actually working on a project that looks at the field of neuroimaging from a data management prospective. Myself and a collaborator are presently surveying fMRI researchers about how they handle their data over time. Our goal is to inform better practice or, at the very least, more forward thinking about managing fMRI data.
Happy to help if you (or anyone) has more specific fMRI/data management-related questions.
-John
--
John Borghi, PhD
CLIR Postdoctoral Fellow
University of California Curation Center (UC3)
California Digital Library
|