Conversation
| ## Reconstructed cortical surfaces | ||
|
|
||
| Reconstructed cortical surfaces should be stored as GIFTI files, and each | ||
| hemisphere should be stored separately. |
There was a problem hiding this comment.
I think we might want to generalize a little bit, as there isn't much leap from cortical surfaces to boundary-element-method (BEM) surfaces (see mri_make_bem_surfaces) like skull and skin that are not hemisphere-specific. The following text doesn't add those, but might pave the way.
| ## Reconstructed cortical surfaces | |
| Reconstructed cortical surfaces should be stored as GIFTI files, and each | |
| hemisphere should be stored separately. | |
| ## Reconstructed surfaces | |
| A surface is defined by three-dimensional vertex coordinates, and a set of triangles, | |
| or triplets of indices into the coordinate array. | |
| Surfaces MUST be stored as [GIFTI] surface files, with one surface per file. | |
| For structures with hemispheres, the `hemi-` entity SHOULD be used to distinguish the | |
| surface files corresponding to each hemisphere. | |
| For example, reconstructions of the cortical sheet are typically split into two disjoing | |
| surfaces. |
| <pipeline_name>/ | ||
| sub-<participant_label>/ | ||
| anat/ | ||
| <source_keywords>_hemi-{L|R}[_space-<surfspace>][_volspace-<volspace>][_desc-<label>]_<surftype>.surf.gii |
There was a problem hiding this comment.
Do we need space and volspace? The surface file itself defines a surface space, so I suspect the interesting space here is "what volume is it embedded in". Can we think of use cases for either of space or volspace?
I think we should also add the den-<label> entity. As a (probably wrong) proposal, this might obviate the need for surface space.
And looking ahead to BEM surfaces, the hemi- might be considered RECOMMENDED not REQUIRED.
| <source_keywords>_hemi-{L|R}[_space-<surfspace>][_volspace-<volspace>][_desc-<label>]_<surftype>.surf.gii | |
| <source_entities>[_hemi-{L|R}][_den-<label>][_desc-<label>]_<suffix>.surf.gii |
Another approach could be to separate cortical surfaces from BEM surfaces, and keep hemi REQUIRED here.
There was a problem hiding this comment.
Keep space, drop volspace, add den, explain space is mesh.
Ignore the comment about hemi, will be resolved by making file rules. hemi is required for cortical surfaces, will not be permitted for BEM surfaces.
| | flat | The flattened surface (used for visualization) | | ||
|
|
||
| `space` filename keyword is restricted to | ||
| [Surface Coordinate Spaces](../99-appendices/08-coordinate-systems.md#Surface) |
There was a problem hiding this comment.
Currently there is no header for #Surface or #Volume to land on at ../99-appendices/08-coordinate-systems.md
|
maintenance note: added a link to the rendered BEP in the top message of this PR. |
As a follow-up to #265 (Common Derivatives), this PR adds structural derivatives. This was taken directly from the draft PR #207.
I have not made any modifications to bring it into line with the final common derivatives yet, but I wanted to get this started.
I will go through and comment up issues I see so we can discuss a bit before I push through with them.
Link to the HTML rendered BEP: https://bids-specification--518.org.readthedocs.build/en/518/05-derivatives/04-structural-derivatives.html
cc @edickie @vsiless