The hackathon will take place in Abingdon for March 16 and 17 and at Rutherford Appleton Laboratory on the 18th.

Proposed CCP PETMR targets for the next “synergistic” hackathon:

• Optimisation with CIL of complex image data (at a minimum by converting between pair-of-real and complex images, and optimising in the former, ideally also by checking some “native” complex optimisation algos)
• Prior for complex image data (as above)
• With all this, do MR reconstruction of undersampled data with CIL
• joint PET-MR reconstruction from simulations (everything in same space)
• joint PET-MR reconstruction from simulations (including resampling to accommodate for different spaces and/or voxel sizes)
• joint PET-MR reconstruction from measured data…

Proposed CCPi CIL targets:

• we are interested in getting CIL to work on PET recon. It would be great to get a proper simulated/real data demo, comparing different recon methods and demonstrating benefit thereof. I think this would involve some of the new functionality like composition and may also require adding more functionals like Huber TV, depending on what you are interested in.
• getting joint reconstruction, where different data sets have a different resolution and different model. Maybe PET and CT. Or on our side, we are interested in combining a conventional X-ray CT high-resolution scan with a hyperspectral low-resolution scan of the same sample.
• Stochastic versions of algorithms is ongoing, and would be great to expand and test more.
• Continue to work at the motion compensation. It does work on a toy model but it'd be nice to test it on a real/simulation data.
• address MR and the issues of complex spaces

We encourage participation to the hackathon, but this one will require some experience with SIRF and/or CIL and is not intended as a training session on the software. Travel and accommodation expenses will be refunded.

Please fill the participation form