Fastcat
A rapid, highly-realistic, scanner specific CBCT simulator.
Features
- X-ray spectra calculation using the models from xpecgen
- Detector response based on MC optical simulation of Scintillators
- High resolution Catphan phantoms
- Bowtie filters, flattening filters, ant-scatter grid models
- Python library and Graphical User Interface
Installation
Fastcat requires a few dependencies to run. Most importantly you need a cuda capable GPU.
Dependencies
- CUDA
- TIGRE
- Scientific python installation (anaconda ect.)
Fastcat requires TIGRE which can be installed using directions here:
https://github.com/CERN/TIGRE/blob/master/Frontispiece/python_installation.md
After installing TIGRE I recommend cloning the repository and installing it in developer mode so that you can access the files
git clone https://github.com/jerichooconnell/fastcat
cd fastcat && python setup.py develop
Fastcat is a work in progress and there are quite a few bugs to work out. So feel free to file issues and I'll take a look.
Citation
Please cite the two publications on the development of fastcat so that I can climb rungs in the academic rat race. Also have a read for more info, there are preprints on arxiv as well:
-
[1] O'Connell, J. and Bazalova-Carter, M. (2021), fastCAT: Fast cone beam CT (CBCT) simulation. Med. Phys., 48: 4448-4458. doi.org/10.1002/mp.15007.
-
[2] O’Connell, J, Lindsay, C, Bazalova-Carter, M. Experimental validation of Fastcat kV and MV cone beam CT (CBCT) simulator. Med. Phys. 2021; 48: 6869– 6880. doi.org/10.1002/mp.15243.