Weak-Neural-Variational-Inference_harmonic_elastography

Weak-Neural-Variational-Inference framework application for a synthetic problem in harmonic elastography.

The paper can be found open-access here: The full open-access paper can be found at: https://www.sciencedirect.com/science/article/pii/S0045782524007473

If you are looking for the base Weak-Neural-Variational-Inference framework with application in static elastography, you can find the GitHub page here: https://github.com/pkmtum/Weak-Neural-Variational-Inference

Dependencies

• Python
• Fenics (install first and all following using pip)
• Fenics Adjoint
• torch with cuda
• scipy
• matplotlib
• tqdm

Installation

Install Python and all dependencies mentioned above. To clone this repo:

git clone https://github.com/pkmtum/Weak-Neural-Variational-Inference_harmonic_elastography.git

How to run

We provide the code, input files, and example data to run the code.

• Input.py can change code details like which data to load, where to calculate (GPU /CPU), posterior, etc.
• Generate_Ground_Truth_Fenics.py can generate ground truth displacement data that can be loaded instead of the data provided
• main.py executes the code

In the current form, 24 GB of GPU memory (we used a Nvidia RTX 4090) was used.

Citation

If this code is relevant to your research, we would be grateful if you cite our original work on the framework:

@article{scholz2024weak,
  title={Weak neural variational inference for solving Bayesian inverse problems without forward models: applications in elastography},
  author={Scholz, Vincent C and Zang, Yaohua and Koutsourelakis, Phaedon-Stelios},
  journal={arXiv preprint arXiv:2407.20697},
  year={2024}
}

Contact

If you have questions or problems regarding the code or paper, please feel invited to reach out to us using the following E-Mails

Dipl.-Ing. Vincent C. Scholz: [email protected]

Dr. Yaohua Zang: [email protected]