Julien Cohen-Adad
Julien Cohen-Adad
I'm not sure what is the best strategy here, because ultimately the sidecar json files are generated by dcm2niix, so what we should be fetching is the BIDS version that...
> @jcohenadad seems to me that BIDS version is very ambiguous since people can output almost everything they want from dcm2bids. If you agree, I would remove completely the version...
hum... based on @agahkarakuzu comment I would keep the field then, at least to make sure we can be cross-compatible with previous BIDS versions. Eg in 2y from now it...
For now, let's use this issue for documentation. Here's the Zenodo DOI: https://zenodo.org/record/6596007/export/hx#.Yqo88ezMK3U In the future, we might want to update/curate this citation @arnaudbore (eg: it says 'Nick' as one...
This is not a proper demonstration that your tool produces the same result as fsl. The phantom has not structure and B0 is flat there
> Is there a way to connect qMRI capabilities with spinal cord toolbox? Would this need to be done with the python implementation of sc toolbox or with the older...
Never close an issue without referencing to the commit that fixes the issue, or if no fix is provided, explain why the issue can be closed.
i am also questioning why our recent pytorch models are "big", while our previous tensorflow models, which already worked really well, are ~5MB each.
> What models should be considered in this issue? > I'm aware of the ones contained in the `models` directory in the OSF project (which are linked in `spinalcordtoolbox/deepseg/models.py`), but...
Assuming most models (>90%) will be