AMTA-Net

This is a python (PyTorch) implementation of Asymmetrical Multi-Task Attention Network proposed in our MICCAI2020 paper "Asymmetrical Multi-Task Attention U-Net for the Segmentation of Prostate Bed in CT Image". An exhaustive extension of this work is published on Medical Image Analysis journal paper "Asymmetric Multi-Task Attention Network for Prostate Bed Segmentation in Computed Tomography Images".

Network architecture

Inner structure of attention module

Citation

@inproceedings{Xu2020PBSeg, 
  title = {Asymmetrical multi-task attention u-net for the segmentation of prostate bed in CT image},
  author = {Xuanang Xu and Chunfeng Lian and Shuai Wang and Andrew Wang and Trevor Royce and Ronald Chen and Jun Lian and Dinggang Shen},
  booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
  year = {2020},
  pages = {470--479},
  organization={Springer}
}

@article{Xu2021PBSeg,
  title = {Asymmetric multi-task attention network for prostate bed segmentation in computed tomography images},
  author = {Xuanang Xu and Chunfeng Lian and Shuai Wang and Tong Zhu and Ronald C. Chen and Andrew Z. Wang and Trevor J. Royce and Pew-Thian Yap and Dinggang Shen and Jun Lian},
  year = {2021},
  publisher = {Elsevier},
  pages = {102116},
  journal = {Medical Image Analysis}
}