Submission deadline is: August 4, 2020 by 5 pm Pacific Time.
Tentative title: Strengthening community and code foundations for brain imaging.
Proposed projects should be 12 months in duration with a projected start date no earlier than January 1, 2021. Actual start date may vary.
Link to grant on online CZI portal.
- DICOM to NIfTI etc support, with integration of dcmstack. This may well involve integration with xarray (see below) (Brendan Moloney, MB)
- Labelled axes for images, probably via xarray (Chris M, MB)
- Integrated API for surface formats - (Chris M) - see Nibabel issue
- Integrate nitransforms for input / output and combination spatial transforms, including standardization of file formats (see HDF5 below) (Oscar E)
- HDF5 serialization for all image formats (Chris M)
- New tractography file standard - see nibabel issue (Oscar E)
- Improved memory performance for image iteration (suggestion by Gaël V).
- Performant desktop and web image visualization, investigate vispy and napari.
- Ability to read images directly from URLs, read / write images to / from AWS, GCS, Azure data storage addresses - c.f. Pandas I/O
MB is floating the idea of integrating an EdX course into the proposal, with the basis from these two Berkeley courses:
This would:
- Serve as documentation and demonstration for the library.
- Attract young developers / practitioners to the library, tools and workflow
- Make it more likely that the practitioner will become a developer.
We propose to make our project more diverse by offering scholarships for the source, which would provide free completion certificates / grading support, in return for 2 pull requests to some library of relevance for imaging, at the end of the course.
MB is trying to get feedback from the CZI for this part of the proposal.