Dr. Olivier Jaubert, Dr. Javier Montalt-Tordera, Dr. Daniel Knight, Pr. Simon Arridge, Dr. Jennifer Steeden, Pr. Vivek Muthurangu
A modified FastDVDnet [1] network is trained for interactive and low latency cardiac MRI imaging. Provided code provides optimized trajectory, model training and offline reconstruction as implemented for the paper.
The ethics does not allow sharing medical image data. The code uses natural images of roses with additional simulation of the coils and simple motion.
video_orientation_change.mp4
All required packages can be installed using conda in a virtual environment:
$ conda env create --name hyperslice --file environment.yml
Note that only Linux is supported.
Once the environment created, activate using:
$ conda activate hyperslice
You can then run:
- the python file example.py file from project directory.
$ python example.py
- or the ipython notebook example.ipynb (to see intermediate results)
Results are saved in ./Training_folder (as in the already trained exemple model ./Training_folder/Exemple_Trained_FastDVDnet)
Logs can be visualised using tensorboard:
$ tensorboard --logdir path_to_directory
Network architecture inspired from original FastDVDnet [1]. Code relies heavily on TensorFlow [2] and TensorFlow_MRI [3].
[1] Tassano, M., Delon, J., & Veit, T. (2020). FastDVDnet: Towards Real-Time Deep Video Denoising Without Flow Estimation. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 1351–1360.
[2] Abadi M, Barham P, Chen J, et al. TensorFlow: A System for Large-Scale Machine Learning TensorFlow: A system for large-scale machine learning. Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation 2016:265--283.
[3] Montalt Tordera J. TensorFlow MRI. 2022 doi:10.5281/ZENODO.7120930. https://pypi.org/project/tensorflow-mri/