Hi @kjamison,
I'd like to build a NeMo container for Neurodesk (neurodesk.org), the open neuroimaging environment used by thousands of researchers, so the tool is available alongside the other lesion/connectome packages it already ships (BCBtoolkit, DeepDisco, LSTAI, etc.). The container would be public, citing your work per your repo's citation guidance.
LQT's repo now points users to NeMo as the recommended successor, so I expect demand from groups currently using LQT-based workflows.
The blocker on a self-contained container is the tractography database generated by databaseprep/. I can't tell from the README whether the precomputed sparse matrices (warped streamlines, endpoints, chunklists for the 420 HCP reference subjects) are downloadable anywhere, or whether the AWS web interface is currently the only path for users without HCP DUA access and the compute to regenerate them.
If there's a hosted download (S3 path, Zenodo deposit, etc.), I'd mirror it onto Neurodesk's storage and bundle it. If not, I'd love to know whether the lab would consider distributing one, happy to help with that.
Thanks for your work on this tool.
Cheers,
Michèle
Hi @kjamison,
I'd like to build a NeMo container for Neurodesk (neurodesk.org), the open neuroimaging environment used by thousands of researchers, so the tool is available alongside the other lesion/connectome packages it already ships (BCBtoolkit, DeepDisco, LSTAI, etc.). The container would be public, citing your work per your repo's citation guidance.
LQT's repo now points users to NeMo as the recommended successor, so I expect demand from groups currently using LQT-based workflows.
The blocker on a self-contained container is the tractography database generated by databaseprep/. I can't tell from the README whether the precomputed sparse matrices (warped streamlines, endpoints, chunklists for the 420 HCP reference subjects) are downloadable anywhere, or whether the AWS web interface is currently the only path for users without HCP DUA access and the compute to regenerate them.
If there's a hosted download (S3 path, Zenodo deposit, etc.), I'd mirror it onto Neurodesk's storage and bundle it. If not, I'd love to know whether the lab would consider distributing one, happy to help with that.
Thanks for your work on this tool.
Cheers,
Michèle