search

LLNL / LEAP

comprehensive library of 3D transmission Computed Tomography (CT) algorithms with Python and C++ APIs, a PyQt GUI, and fully integrated with PyTorch
https://leapct.readthedocs.io
MIT License
104 stars 10 forks source link

Support AAPM low-dose CT helical reconstruction #29

Closed stefenmax closed 4 months ago

stefenmax commented 5 months ago

Although Low Dose CT Grand Challenge is 2016 dataset. Nowadays still many paper use it. But it seems not a clear code to guide how to reconstruct to CT image from the raw projection data. If LEAP can do that, I can be use to the down stream task(sparse-view & limited angle).

kylechampley commented 5 months ago

It's been a long time since I've looked at the data in this challenge. I think there is some flying focal spot (FFS) rebinning, but don't remember what else is necessary. Can you remind me? LEAP does not have FFS rebinning, but if that is all that is missing, shouldn't be too hard to get an implementation of this completed.

stefenmax commented 5 months ago

I'm not familar with that, All I can found is this two project(odl and cuda-based) focus to slove this reconstruction problem but they both too old to be run.

kylechampley commented 5 months ago

Thanks for the links, but unfortunately, they don't help me determine what preprocessing is needed without doing a deep dive into the code. It would be more helpful to have a verbal explanation of what is needed.

I would like to provide support for this dataset in LEAP, but I don't have much time to work on this right now. If you can provide a list (or link to a list) of the necessary algorithms for me, then I can try to get this working.