Skip to content

This repository contains code to generate DRRs using deepDRR from CT scans in the IHUdeLiver10 dataset at https://doi.org/10.57745/EUBXGH .

Notifications You must be signed in to change notification settings

coolteemf/IHUdeLiver10-2D_3D-deformable-registration

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

IHUdeLiver10 2D-3D registration

Description

This repository contains code to automatically configure the projective geometry to generate Digitally Rendered Radiographs (DRRs) from a CT scan.

This code is released with the IHUdeLiver10 dataset (to be released) to render DRRs from CT scans in an standardized way.

The library used for DRR rendering is DeepDRR, which depends on CUDA.

Table of Contents

Installation

The package dependencies are listed in requirements.txt.

Usage

The code is split between a main and a utils file.

The main file takes command line arguments to generate the projection parameter files and optionally render DRRs to better understand the effect of projection parameters.

Example usage from the terminal:

python ./ihudeliver10_2d3d/main.py --volume_path "path/to/volume" --cuda_path "/usr/local/cuda/bin" --max_disp 150 --center x y z --roi_size x_size y_size z_size

Contributing

Suggestions are welcome.
If you encounter an issue or a bug, please feel free to raise an issue.

Acknowledgements

This work was funded by the ANR (ANR-20-CE19-0015).
The IHUdeliver10 was acquired by Juan Verde, MD., at the Strasbourg IHU, who provided invaluable support during the development process.
We thank the authors of the DeepDRR framework, which was instrumental to the development of this code.

About

This repository contains code to generate DRRs using deepDRR from CT scans in the IHUdeLiver10 dataset at https://doi.org/10.57745/EUBXGH .

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages