Repository of computational imaging (roci) is a collection of clean, self-contained implementations of algorithms used in computational imaging. Currently focused on computational MRI applications (reconstruction, synthesis, quantification) and techniques based on representation learning and generative modeling. For educational and research purposes only.

Organization

The directory structure is simple: algorithm = python_file + demo_notebook + readme.

algorithms
    |
    |- algo_1
    |   |- algo_1.py
    |   |- Demo.ipynb
    |   |- README.md
    |
    |- algo_2
    |   |- algo_2.py
    |   |- Demo.ipynb
    |   |- README.md
    ...

Available Algorithms

MRI reconstruction:

• SENSE parallel-imaging reconstruction
• CG-SENSE parallel-imaging reconstruction
• Compressed sensing reconstruction

MRI physics simulation:

• Bloch signal simulation
• Extended phase graphs method

Image representations:

• Implicit neural representations for differentiable uncalibrated imaging

Coming Soon

• Deep image prior
• Gaussian representations
• Equivariant imaging
• Double Blind Imaging with Generative Modeling
• Compressed-sensing with generative modeling (CSGM)
• noise2noise denoising
• Plug-and-play denoiser-based reconstruction
• Diffusion model-based inversion
• Flow matching-based inversion
• ESPIRiT parallel-imaging
• Magnetic resonance spin tomography in time-domain (MR-STAT)