including Siemens metadata in the NIFTI header

[jaystock]

It's inconvenient to work with .json sidecar files to access NIFTI metadata related to the MRI acquisition. Even the most basic scan metadata such as the delta_TE for B0 map phase difference images is lost in the standard NIFTI headers.

I have always thought that it would be nice to have the Siemens metadata included in the NIFT header itself. This would only increase the file size modestly. Would it be worthwhile to change dcm2niix to store the metadata in this way? Or at least add an option to let users add this metadata? That way, you don't have to deal with the .json file anymore.

[neurolabusc]

@jaystock This request is out of scope for dcm2niix. The purpose of dcm2niix is to convert data from DICOM files (which vary across manufacturers and software releases) into formats that comply with the BIDS standard. Decisions about how metadata should be stored are governed by BIDS itself, which has an established process for proposing and discussing changes through the BIDS Enhancement Proposal (BEP) framework.

dcm2niix aims to support the BIDS specification rather than redefine it. The software is released under a permissive open-source license, so you are welcome to fork it and implement alternative metadata storage if that better suits your needs.

For full disclosure, I was a vocal critic of several design decisions in the original BIDS specification. It was not the format I advocated for. That said, I respect the creators’ willingness to consider diverse perspectives while remaining committed to their vision of making data human-readable and comprehensible. Its widespread adoption suggests that it filled an important niche and has helped unify the field.