Sparse-view CT reconstruction

[Guzaiwang]

Thanks for your excellent work on 3D MedDiffusion.

I am writing to inquire whether you can release the codes and docs showing how to fine-tune for downstreaming tasks, such as sparse-view CT reconstrcution.

[schadol]

Great work and thank you very much for sharing your code.
I would also be very interested in the conditioning process using the ControlNet in the case of MR reconstruction.
Did you use the single coil complex data or a single magnitude image as conditioning?
Also did you observe generative variations in the reconstruction outputs?
I would also greatly appreciate the codes for the downstream tasks. Thank you!

Best regards,
Oliver

[Mors20]

Sorry for the late response. I am continuing to explore this work and plan to release a more refined version in the future, so I am not actively maintaining this code repository at the moment. But for your request, I will provide the ControlNet code and write a short instruction using sparse-view CT reconstruction as an example when I have time. Thank you.