Suffix to represent White Matter null MRI scans (FGATIR)

[rbakker]

There are MRI suffixes for T1w T2w T2starw and FLAIR but not for FGATIR. I propose to add an FGATIR extension, or perhaps a WMn (white matter null) extension.

Where FLAIR (Fluid attenuated inversion recovery) is a modality that nulls out signal from fluids (like CSF) and is widely used for tumor detection, FGATIR (Fast Gray Matter Acquisition T1 Inversion Recovery) is designed to null out white matter signal, and is widely used in deep brain stimulation because it make it possible to see thalamic nuclei that can be used as a DBS target.

FGATIR is not the only technique used for White Matter null scans, the other one is WMn-MPRAGE. That could be a reason to prefer the WMn suffix over the FGATIR one., since both scan types serve the same purpose. I am not enough of a BIDS expert to know which is to be preferred, should the extension reflect the purpose of the scan or the common name of the protocol? To initiate a discussion, here are some quotes from a Google query:

"Visualization of intra-thalamic nuclei with optimized white-matter-nulled MPRAGE at 7T" "The recently proposed MP2RAGE, which is a modification of the MPRAGE and generates two image sets at different inversion times, was employed to generate the FGATIR-like contrast (FLAWS1) and MPRAGE-like contrast" "Inversion repetition times (TS) of 4.5s and 6s were found to yield the highest SNR efficiency for WMn MP-RAGE at 3T and 7T, respectively"

An this article introduces FGATIR, and contains a table with scanner settings: https://movementdisorders.ufhealth.org/research/basictranslational-research-for-movement-disorders/fgatir-new-scan-for-surgical-targeting/

Parameter | T1-w 3D MP-RAGE | T2-w 3D FLAIR | T1-w 3D FGATIR Repetition time (TR) | 1600ms | 6000ms | 3000ms Echo time (TE) | 4.38ms | 353ms | 4.39ms Inversion time (TI) | 800ms | 2200ms | 409ms Inversion pulse angle | 90° | 180° | 180°

And finally a link to the neurostars question that started this: https://neurostars.org/t/representing-fgatir-mri-scan-in-bids/32542/2

[effigies]

Hi, thanks for starting this discussion!

I think this seems like a reasonable thing to add to BIDS if it's in use. The UF page describes it as "T1-w 3D FGATIR", but clearly it is not interchangeable with other T1w scans.

I would probably lean toward WMn or WMnull as being more descriptive.

Is there anything standardized in DICOM that would allow these different sequences (FGATIR, WMn-MPRAGE, ...) to be detected automatically? Are there specialized metadata that are necessary/helpful to interpret these scans that are not already covered in https://bids-specification.readthedocs.io/en/stable/modality-specific-files/magnetic-resonance-imaging-data.html?

cc @yarikoptic @neurolabusc who will be able to ask better questions to determine how big a lift supporting these sequences would be.

[yarikoptic]

are there sample DICOMs somewhere? For reproin convention/heuristic would be just a matter of naming it anat-WHATEVER since that one doesn't even try to be smart ;-)

[rbakker]

In terms of parameters you would fine in DICOM FGATIR scans, I think the Philips and Siemens settings listed on the ufhealth page are quite informative: https://movementdisorders.ufhealth.org/files/2009/06/PhilipsAchieva-FGATIR1mm.webarchive.txt https://movementdisorders.ufhealth.org/files/2009/06/SiemensAllegra_FGATIR.pdf

I have access to patient dicoms from a Philips Achieva scanner, but cannot share them in their entirety. I can run some code to extract the header and remove patient info, would that be of help, and which code would you suggest (pydicom, save as json?)

[yarikoptic]

any chance to get some small shareable phantom images so they could be added to the corresponding repo at https://github.com/neurolabusc?tab=repositories&q=dcm ?