MTS should follow updated BIDS specs

[bpinsard]

https://bids-specification.readthedocs.io/en/stable/04-modality-specific-files/01-magnetic-resonance-imaging-data.html#anatomy-imaging-data

[bpinsard]

I am bit confused on how to match our acquisitions to the BIDS spec for MT now that it has been merged. For brain the GRE sequences we acquire are suffixed MTw/PDw/T1w and for c-spine MT0/MT1/T1w/ME. In BIDS it needs to have this naming

 sub-<label>[_ses-<label>][_acq-<label>][_ce-<label>][_rec-<label>][_run-<index>][_echo-<index>]_flip-<index>_mt-<on|off>[_part-<mag|phase|real|imag>]_MPM.json
sub-<label>[_ses-<label>][_acq-<label>][_ce-<label>][_rec-<label>][_run-<index>][_echo-<index>]_flip-<index>_mt-<on|off>[_part-<mag|phase|real|imag>]_MPM.nii[.gz]
sub-<label>[_ses-<label>][_acq-<label>][_ce-<label>][_rec-<label>][_run-<index>][_echo-<index>]_flip-<index>_mt-<on|off>[_part-<mag|phase|real|imag>]_MTS.json
sub-<label>[_ses-<label>][_acq-<label>][_ce-<label>][_rec-<label>][_run-<index>][_echo-<index>]_flip-<index>_mt-<on|off>[_part-<mag|phase|real|imag>]_MTS.nii[.gz]

@agahkarakuzu @jcohenadad could you please help my qMRI ignorance:

• I haven't seen anywhere in the specs what MPM stands for. Do we acquire MPM and which of our sequence is that?
• If I am correct, we don't have multiple FlipAngle in the MT scans, as such the flip- index should be 1 for all scans?

Thanks!

[bpinsard]

I looked into the spine-generic dataset and they don't seems to follow the latest BIDS spec for MTS though. https://github.com/spine-generic/data-single-subject Would fixing our BIDS will prevent our dataset to go through your pipelines?

[jcohenadad]

For brain the GRE sequences we acquire are suffixed MTw/PDw/T1w and for c-spine MT0/MT1/T1w/ME.

i find this example helpful

I haven't seen anywhere in the specs what MPM stands for. Do we acquire MPM and which of our sequence is that?

that would be a multi-echo MTS sequence. Here is an example

If I am correct, we don't have multiple FlipAngle in the MT scans, as such the flip- index should be 1 for all scans?

Or we could probably remove the flip index altogether (lighter for the eyes)

I looked into the spine-generic dataset and they don't seems to follow the latest BIDS spec for MTS though. https://github.com/spine-generic/data-single-subject

indeed they don't --> https://github.com/spine-generic/data-single-subject/issues/25

[bpinsard]

Thanks for the blazing fast response! I will look into the examples you linked and try to match our acquisitions.

Or we could probably remove the flip index altogether (lighter for the eyes)

The flip- is non-optional in the spec though.

[bpinsard]

I think our B1Plusmap should instead be named TB1TFL.

[agahkarakuzu]

If you refer to the scaled flip angle map produced by Siemens, then yes it should be (with acq-famp):

If you are referring to a B1+ map that is converted to BIDS units (relative %B1+), which should be a derivative, then it is called TB1map (stands for transmit (T) B1map).