Trained Models of Semi-supervised COVID-19 Infection Segmentation

Trained models in the following paper.

@article{Ma20SemiCOVIDSeg,
    author={Jun Ma and Ziwei Nie and Congcong Wang and Guoqiang Dong and Qiongjie Zhu and Jian He and Luying Gui and Xiaoping Yang},
    title={Active contour regularized semi-supervised learning for COVID-19 CT infection segmentation with limited annotations},
    journal={Physics in Medicine & Biology},
    url={http://iopscience.iop.org/article/10.1088/1361-6560/abc04e},
    year={2020}
}

These trained models were built on nnUNet (https://github.com/MIC-DKFZ/nnUNet).

We also use them to infer this public COVID-19 CT dataset:https://wiki.cancerimagingarchive.net/display/Public/CT+Images+in+COVID-19.

These pseudo labels are also provided.