nifti2_data_format > RE: NIFTI-2 proposal
Mar 15, 2011  05:03 PM | Mark Jenkinson
RE: NIFTI-2 proposal
Response to the postings on the public forum:
The NIfTI committee thank all those who responded to the public announcement of the NIfTI-2 proposal.  The outcomes of discussions of these postings and other responses outside of the forum are summarised below.  Please see the associated posting for full details of the revised, finalised NIfTI-2 format.
Changes to the proposed NIfTI-2 format:
 - Removal of unused fields (data_type, db_name, extents, session_error, regular, glmax, glmin).
 - Changing the order of several fields to ensure 8-byte alignment of all doubles (an issue raised outside of the forum, which would be a major problem on some platforms if it was not ensured).
 - Shifting the magic string to the front of the header, immediately following sizeof_hdr.
 - A change in the total header size (including 4-byte extension info) to become 544 bytes.
 - A range for customised codes for INTENT, XFORM, UNITS and SLICE will be specified for temporary customisation, although users will continue to be encouraged to register for official codes and have them publicly documented to avoid any hidden or private information polluting the open standard.
Additional points:
 - There was considerable discussion about the hdr/img format and although several people were in favour of dropping this to distance the format from the older Analyze standard, it was still being actively used by some developers and has the potential for considerable benefit when dealing with very large datasets, where the header information can be read and modified without requiring any reading or writing of the data values. Consequently, it was felt that it was more advantageous to retain the existing hdr/img format and also leave the extension name for the single format (.nii) unchanged.
 - The changes in nifticlib will be considered carefully, although these are separate issues from the definition of the NIfTI-2 standard itself.
 - The standardisation of spatial transformation information is an important issue and currently being looked into in a separate endeavour on developing a standard format for spatial transformation data.
 - There were several requests for changes in how the spatial coordinate information is stored and represented (qform and sform) or even suggestions that the coordinate handedness should be forced/fixed, and multiple transformations be incorporated. The original intention of the proposed NIfTI-2 format was to create an easy to implement upgrade of the existing NIfTI-1 standard that maintains the same information and the same logic. This was a very important point in getting many of the developers to agree to implement this NIfTI-2 standard. Several developers depend on the way that the qform and sform information is currently implemented in NIfTI-1 and changes in meaning or even storage method could cause problems and delays in implementing the NIfTI-2 standard.  The qform and sform issues are therefore much bigger issues than being considered for NIfTI-2 and would alter the logic involved in any implementation compared to NIfTI-1. Therefore, although these issues are extremely important, we feel they are more appropriate for the upcoming discussions on future formats where such major changes to the logic and implementation can be made without restriction. All comments should therefore be forwarded to the discussion forum for future formats.
 - A request to have a pure XML header is also a major change and shares some of the same issues as the above request, and so should be forwarded to the discussion forum for future formats.  Note that there has never been an official nifti-1 format that allowed a purely XML header - there is some code in nifticlib to allow such XML to be generated, but this was never part of the official nifti-1 specification.

Threaded View

TitleAuthorDate
Mark Jenkinson Feb 28, 2011
RE: NIFTI-2 proposal
Mark Jenkinson Mar 15, 2011
Cinly Ooi Mar 15, 2011
Cinly Ooi Mar 2, 2011
Ged Ridgway Mar 7, 2011
Jon Clayden Mar 5, 2011
Cinly Ooi Mar 1, 2011
Andrew Janke Mar 1, 2011
Cinly Ooi Mar 2, 2011
Satrajit Ghosh Mar 5, 2011
Cinly Ooi Mar 1, 2011
Cinly Ooi Mar 1, 2011
Cinly Ooi Mar 1, 2011
Cinly Ooi Mar 1, 2011
Denis Rivière Feb 28, 2011
Cinly Ooi Mar 1, 2011
Brandon Whitcher Mar 1, 2011
Satrajit Ghosh Feb 28, 2011
Jonas Larsson Mar 1, 2011
Mark Horsfield Mar 1, 2011
Andrew Janke Mar 1, 2011
Jochen Weber Feb 28, 2011
Randall Frank Mar 1, 2011
Michael Martinez Feb 28, 2011
Cinly Ooi Feb 28, 2011
Chris Rorden Feb 28, 2011