Synthetic CT
Collection
Resources for synthetic CT generation β’ 7 items β’ Updated
How to use VBoussot/Synthrad2025 with KonfAI:
pip install "konfai[imaging]" konfai-apps
# List the apps this repository holds konfai-apps list VBoussot/Synthrad2025 # Run one of them on an image konfai-apps infer VBoussot/Synthrad2025:<app> -i image.nii.gz -o Output
This repository hosts the pretrained models developed for the SynthRAD 2025 challenge on synthetic CT generation from MRI (Task 1) and CBCT (Task 2).
All models are trained with 5-fold cross-validation and released for reproducible research.
There are two types of models:
Each set comes with ready-to-use prediction configurations for KonfAI.
Task_1/AB-TH/CV_0.pt β¦ CV_4.pt + Prediction.yml Task_1/HN/CV_0.pt β¦ CV_4.pt + Prediction.ymlπ Full usage instructions: GitHub β Synthrad2025 Task 1
Task_2/AB-TH/CV_0.pt β¦ CV_4.pt + Prediction.yml Task_2/HN/CV_0.pt β¦ CV_4.pt + Prediction.ymlπ Full usage instructions: GitHub β Synthrad2025 Task 2