[MINC-users] [Fwd: OHBM Announcement 3/9/04]

[Alex ZIJDENBOS]

This is also what I have picked up through the grapevine - that MINC
2.x has a reasonably high likelihood of becoming NIfTI-2. But I agree
that there probably is more work to be done to make that happen. I
would hope the DFWG will come out with some requirements/specs for
NIfTI-2 such that we (well some of you :) can focus (y)our
efforts. ANALYZE is after all still just ANALYZE.

Is there a timeline for NIfTI-1 and NIfTI-2 (their web site is
currently down)? Not that it's any of my concern, but I have to admit
I don't really see the point of first unifying ANALYZE for NIfTI-1,
and then possibly (probably) switching to something entirely new for
NIfTI-2. Seems like make-work to me, and also - if ANALYZE is actually
improved - it will be harder to get people away from it down the
road. Why not bite the NIfTI-2 bullet right away, me wonders.

-- Alex

On Fri, Mar 12, 2004 at 10:19:19AM +1000, Andrew Janke wrote:
> On Thu, 10 Mar 2004, Yasser Ad-Dab'bagh wrote:
> 
> > Should we be concerned?
> 
> "perhaps".
> 
> I have been in contact with a few of the DFWG members in the last few days with
> respect to this.  What seems to have happened so far is this:
> 
>   * NIfTI-1 is the patch-up job for the various forms of analyze.
>      - this is actually a big improvement over previous versions.
> 
>   * NIfTI-2 is to be the replacement for ANALYZE that will be all-singing
>        and all-dancing.
> 
> However no decision or specs are available so far for NIfTI-2.  Apparently a
> number of the members of the comittee are all for just using MINC 1.2/2.0.
> however there are some who are opposed to even HDF5 or NetCDF.  I have asked for
> feedback on the problems these people on the comittee have against MINC 2.0 in
> order that we might be able to alleviate their concerns with MINC 2.0.
> 
> I haven't recieved anything yet though.  Apaprently the group will be meeting
> again soon to discuss NIfTI-2 in more detail.
> 
> In the meantime if you want your voice to be heard, send an email to the group
> voicing your opinion, the more people who indicate that they would like to see
> MINC 2.0 replace NIfTI-2 the better. However remember that the NIfTI group is
> only concerned with Data formats for fMRI data.  Whilst MINC can (and does)  do
> fMRI data, there is still a bit of work needed on MATLAB interfaces to MINC.
> 
> 
>    http://nifti.nimh.nih.gov/
> 
> 
> --
> Andrew Janke   ( rotor@cmr.uq.edu.au || www.cmr.uq.edu.au/~rotor )
> Australia->University of Queensland->Centre for Magnetic Resonance
> W: +61 7 3365 4100  ||  H: +61 7 3800 4042  ||  M: +61 4 2138 8581
> 
> 
> 
> > -----Forwarded Message-----
> > From: Organization for Human Brain Mapping <info@humanbrainmapping.org>
> > To: yasser.ad-dabbagh@MAIL.MCGILL.CA
> >
> > We are writing on behalf of the Data Format Working Group (DFWG), a
> > committee set up under the Neuroimaging Informatics Technology
> > Initiative (NIfTI, http://nifti.nimh.nih.gov), which is jointly
> > sponsored by NIMH and NINDS. We have been charged with arriving "at a
> > technical solution to the problem of multiple data formats in fMRI." We
> > have met several times and decided that one of our first priorities is
> > to describe and attempt to merge the variants of the ANALYZE(TM) 7.5
> > image header that are being used by the developers of the major fMRI
> > data analysis packages.
> >
> > We have defined a compromise data format, dubbed NIfTI-1, that is based
> > on these variants and is "mostly compatible" with the ANALYZE(TM) 7.5
> > header, while using many of the "unused and lesser-used" fields for
> > fMRI-specific purposes. Developers of five of the major fMRI data
> > analysis packages (FSL, SPM, AFNI, BrainVoyager, and FreeSurfer) have
> > agreed to use this compromise format in the future, and will support
> > both input and output of the NIfTI-1 format. A considerable effort was
> > required to reach agreement with all of these developers. Given the
> > extremely limited space in the 348 byte header, it is unlikely that any
> > substantial changes will be possible in the NIfTI-1 format if we are to
> > retain the support of all five packages.
> >
> > We envisage NIfTI-1 replacing the current range of ANALYZE 7.5 variants
> > both for file interchange and data storage.  NIfTI-1, while still
> > limited, contains a considerably richer set of metadata than exists in
> > the ANALYZE-based formats.
> >
> > As of this message, the NIfTI-1 format specification, demonstration
> > programs, and related documents are being released to the general
> > neuroimaging community on the DFWG website
> > http://nifti.nimh.nih.gov/dfwg/ along with an interactive forum for
> > comments and discussion. The DFWG committee will next meet in early May
> > 2004, and any comments you care to make and would like to have
> > considered at the next committee meeting should be received by April 15,
> > 2004.
> >
> 
> _______________________________________________
> MINC-users@bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users