---
license: mit
language:
- en
pipeline_tag: image-segmentation
tags:
- CT scans
- U-Net
---
# Model Card for Models of "Automatic semantic segmentation of the osseous structures of the paranasal sinuses"

This is a model repository. Please check all the information in our [paper](https://doi.org/10.1101/2024.06.21.599833) and [GitHub](https://github.com/BYO-UPM/Craneal_CT) page.

## Model Sources

- **Repository:** [GitHub page](https://github.com/BYO-UPM/Craneal_CT)
- **Paper:** Automatic semantic segmentation of the osseous structures of the paranasal sinuses ([link](https://doi.org/10.1101/2024.06.21.599833))

## Get Start

- 3D models are associated with TensorFlow and Keras.
- 2D and final models are associated with Pytorch.

You can use the 2D and 3D models for reproduction.

Check this [code](https://github.com/BYO-UPM/Craneal_CT/blob/main/Codes/2D_unets/transfer_learning_ddp.py) for transfer learning using our final model.

## Ethical Issue

Please use our models and datasets appropriately and comply with relevant medical information protection regulations. We have wiped out all data involving private personal information.

## Citation

If you use our model or dataset in your research, please cite our paper:

```
@article{yichun2024automatic,
  title={Automatic Semantic Segmentation of the Osseous Structures of the Paranasal Sinuses},
  author={Yichun Sun, Alejandro Guerrero-López, Julián D. Arias-Londoño, Juan I. Godino-Llorente},
  journal={bioRxiv 2024.06.21.599833},
  year={2024},
}
```