[Neuroimaging] JSON-LD and DICOM?
Jasper van den Bosch
japsai at gmail.com
Fri Jun 30 12:35:51 EDT 2017
I have to agree with Andrey on the yet another format argument. Also, tools
will have to convert it to other formats anyway, so if you do end up
storing something in the header, as long as you document it, just like you
would other nibabel properties, I'd go for the simplest solution.
I did a little survey among niprov users and it turns out they usually
don't end up using the header to store the metadata, because the "central"
storage can more easily keep track of the relationships between files.
And if the file is overwritten, or moved, the metadata remains.
On 30 June 2017 at 17:18, Andrey Fedorov <andrey.fedorov at gmail.com> wrote:
> It is always puzzling to me why neuro people are so eager to discard
> the original data.
> Yes, non-DICOM is expected by most image analysis software. Fine -
> give that representation to those tools, but always keep the original
> DICOM. Instead, you guys are coming up with "yet another" format...
> I think coming up with a new format, developing supporting tools and
> toolkits, propagating it through the community, debugging etc - that
> is indeed a high price to pay, with benefits to be seen.
> For what it's worth, keeping track of a relationship seems like a lot
> more natural solution to me!
> My 2c!
> On Fri, Jun 30, 2017 at 12:06 PM, Matthew Brett <matthew.brett at gmail.com>
> > Hi,
> > On Fri, Jun 30, 2017 at 4:53 PM, Andrey Fedorov
> > <andrey.fedorov at gmail.com> wrote:
> > > Matthew,
> > >
> > > I believe that image data and derived data belongs to DICOM, and DICOM
> > > is the format that should be used directly.
> > >
> > > We applied these ideas and demonstrated some capabilities here
> > > https://peerj.com/articles/2057/, and we continue the development of
> > > supporting tools in DCMTK and in https://github.com/qiicr/dcmqi.
> > >
> > > Perhaps if you give me some specifics about "why not DICOM", we could
> > > have a bit more focused conversation.
> > >
> > > I do realize and anticipate that you will take this as a complete
> > Well - not heresy. I understand what you mean, but there's a reason
> > that most brain imaging software doesn't use DICOMs as the native
> > format, so, if we don't import the information from the DICOMs, then
> > the user has to keep track of the relationship of the DICOMs and
> > NIfTIs, which seems like a high price to pay.
> > Cheers,
> > Matthew
> Neuroimaging mailing list
> Neuroimaging at python.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Neuroimaging