This is an implementation of "UNet++: A Nested U-Net Architecture for Medical Image Segmentation" in Python and powered by the Keras deep learning framework (Tensorflow as backend). For the first time, a new architecture, called UNet++ (nested U-Net architecture), is proposed for a more precise segmentation. We introduced the intermediate layers to U-Nets, which naturally form multiple new up-sampling expanding paths of different depths, resulting in an ensemble of U-Nets with a partially shared contracting path.

Detectron is released under the MIT.

If you use UNet++ in your research, please consider the following BibTeX entry.

@incollection{zhou2018unet++,
  title={UNet++: A Nested U-Net Architecture for Medical Image Segmentation},
  author={Zhou, Zongwei and Siddiquee, Md Mahfuzur Rahman and Tajbakhsh, Nima and Liang, Jianming},
  booktitle={Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support},
  pages={3--11},
  year={2018},
  publisher={Springer}
}

• Zongwei Zhou, homepage: zongweiz.com
• Md Mahfuzur Rahman Siddiquee, github: mahfuzmohammad