MRI: Add NumberOfVolumes?

[yarikoptic]

Don't shoot the messenger, I am just channeling @adelavega here:
https://twitter.com/elneurozorro/status/1412885524671602689

Pros:

• allows to avoid looking into .nii.gz to figure it out

Cons:

• metadata duplication

ATM we have only NumberOfVolumesDiscardedByScanner NumberOfVolumesDiscardedByUser neither of which duplicates the metadata.

WDYT?

[adelavega]

To give more context, this is an issue because downloading only the header from a compressed nifti is difficult. Therefore, it would be nice to use an already existing place to hold meta-data that is easy to fetch for such important information.

Possibly there could be a rule to give precedence to the header information if available.

[tsalo]

This sounds related to bids-standard/bids-2-devel#21 (in which I planned to open an issue here for this same field, but never got around to it).

I think it could be useful, as long as it's an optional field. A required one will have to wait until 2.0.

Possibly there could be a rule to give precedence to the header information if available.

Hopefully #761 can help there.

[ddwagner]

Hi all, just came to issues looking to see if there was any mention of a field for nvols. Thought I'd chime in to say I would also find it very useful to have a NumberofVolumes field. I understand not wanting to duplicate metadata. However, in addition to the issue of wanting to avoid downloading nifti.gz files just for the header information, for those of us who still play in Matlab/Octave reading header metadata from a nii.gz is slow. Also, it seems somewhat inelegant to have developers use one set of commands to get 99% of the metadata and then a different set to get that final 1% from the volume headers. Although I could see that argument going either way ;)