This is an implementation of "UNet++: A Nested U-Net Architecture for Medical Image Segmentation" in Keras deep learning framework (Tensorflow as backend). UNet++ (nested U-Net architecture) is proposed for a more precise segmentation. We introduce intermediate layers to skip connections of U-Net, which naturally form multiple new up-sampling paths from different depths, ensembling U-Nets of various receptive fields.


UNet++: A Nested U-Net Architecture for Medical Image Segmentation

Zhou Zongwei, Md Mahfuzur Rahman Siddiquee, Nima Tajbakhsh, and Jianming Liang

Biomedical Informatics, Arizona State University

Deep Learning in Medical Image Analysis (DLMIA) 2018. (Oral)

  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},


Python 3.x, Keras 2.2.2, Tensorflow 1.4.1 and other common packages listed in requirements.txt.

Available architectures

Available backbones

Backbone model Name Weights
VGG16 vgg16 imagenet
VGG19 vgg19 imagenet
ResNet18 resnet18 imagenet
ResNet34 resnet34 imagenet
ResNet50 resnet50 imagenet
ResNet101 resnet101 imagenet
ResNet152 resnet152 imagenet
ResNeXt50 resnext50 imagenet
ResNeXt101 resnext101 imagenet
DenseNet121 densenet121 imagenet
DenseNet169 densenet169 imagenet
DenseNet201 densenet201 imagenet
Inception V3 inceptionv3 imagenet
Inception ResNet V2 inceptionresnetv2 imagenet


git clone https://github.com/MrGiovanni/Nested-UNet.git
cd Nested-UNet
pip install -r requirements.txt
git submodule update --init --recursive

Running the scripts

Application 1: Data Science Bowl 2018

CUDA_VISIBLE_DEVICES=0 python DSB2018_application.py --run 1 \
                                                     --arch Xnet \
                                                     --backbone vgg16 \
                                                     --init random \
                                                     --decoder transpose \
                                                     --input_rows 96 \
                                                     --input_cols 96 \
                                                     --input_deps 3 \
                                                     --nb_class 1 \
                                                     --batch_size 2048 \
                                                     --weights None \
                                                     --verbose 1

Application 2: Liver Tumor Segmentation Challenge (LiTS)

Application 3: Polyp Segmentation (ASU-Mayo)

Application 4: Lung Image Database Consortium image collection (LIDC-IDRI)

Application 5: Multiparametric Brain Tumor Segmentation (BRATS 2013)

CUDA_VISIBLE_DEVICES=0 python BRATS2013_application.py --run 1 \
                                                     --arch Xnet \
                                                     --backbone vgg16 \
                                                     --init random \
                                                     --decoder transpose \
                                                     --input_rows 256 \
                                                     --input_cols 256 \
                                                     --input_deps 3 \
                                                     --nb_class 1 \
                                                     --batch_size 2048 \
                                                     --weights None \
                                                     --verbose 1

Code examples for your own data

Train a UNet++ structure (Xnet in the code):

from segmentation_models import Unet, Nestnet, Xnet

# prepare data
x, y = ... # range in [0,1]

# prepare model
model = Xnet(backbone_name='resnet50', encoder_weights='imagenet') # build UNet++
# model = Unet(backbone_name='resnet50', encoder_weights='imagenet') # build U-Net
# model = NestNet(backbone_name='resnet50', encoder_weights='imagenet') # build DLA

model.compile('Adam', 'binary_crossentropy', ['binary_accuracy'])

# train model
model.fit(x, y)

To do

  • [x] Add VGG backbone for UNet++
  • [x] Add ResNet backbone for UNet++
  • [x] Add ResNeXt backbone for UNet++
  • [ ] Add DenseNet backbone for UNet++
  • [ ] Add Inception backbone for UNet++
  • [ ] Add Tiramisu and Tiramisu++
  • [ ] Add FPN++
  • [ ] Add Linknet++
  • [ ] Add PSPNet++