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

@article{Zhou2018UNetAN,
  title={UNet++: A Nested U-Net Architecture for Medical Image Segmentation},
  author={Zongwei Zhou and Md Mahfuzur Rahman Siddiquee and Nima Tajbakhsh and Jianming Liang},
  journal={Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support : 4th International Workshop, DLMIA 2018, and 8th International Workshop, ML-CDS 2018, held in conjunction with MICCAI 2018, Granada, Spain, S...},
  year={2018},
  volume={11045},
  pages={
          3-11
        }
}
  • Zongwei Zhou, M. R. Siddiquee, Jianming Liang
  • Published 18 July 2018
  • Computer Science
  • Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support : 4th International Workshop, DLMIA 2018, and 8th International Workshop, ML-CDS 2018, held in conjunction with MICCAI 2018, Granada, Spain, S...
In this paper, we present UNet++, a new, more powerful architecture for medical image segmentation. [] Key Result Our experiments demonstrate that UNet++ with deep supervision achieves an average IoU gain of 3.9 and 3.4 points over U-Net and wide U-Net, respectively.

Half-UNet: A Simplified U-Net Architecture for Medical Image Segmentation

TLDR
Experiments demonstrate that Half-UNet has similar segmentation accuracy compared U-Net and its variants, while the parameters and floating-point operations are reduced by 98.6 and 81.8%, respectively, compared with U- net.

MDA-Unet: A Multi-Scale Dilated Attention U-Net for Medical Image Segmentation

TLDR
A model named MDA-Unet is proposed, a novel multi-scale deep learning segmentation model that improves upon U-Net and enhances its performance in segmenting medical images with variability in the shape and size of the region of interest.

UNet 3+: A Full-Scale Connected UNet for Medical Image Segmentation

  • Huimin HuangLanfen Lin Jian Wu
  • Computer Science
    ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2020
TLDR
A novel UNet 3+ is proposed, which takes advantage of full-scale skip connections and deep supervisions, and can reduce the network parameters to improve the computation efficiency.

UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation

TLDR
UNet++ is proposed, a new neural architecture for semantic and instance segmentation by alleviating the unknown network depth with an efficient ensemble of U-Nets of varying depths and redesigning skip connections to aggregate features of varying semantic scales at the decoder sub-networks, leading to a highly flexible feature fusion scheme.

Clinical target segmentation using a novel deep neural network: double attention Res-U-Net

TLDR
This work introduced Double Attention Res-U-Net architecture to address medical image segmentation problem in different medical imaging system and outperformed the state-of-the-art U-Net based model on the external CVC-ClinicDB, and multi-site MRI datasets.

Enhancing U-Net with Spatial-Channel Attention Gate for Abnormal Tissue Segmentation in Medical Imaging

TLDR
A novel spatial-channel attention gate is proposed that is easily integrated into an encoder-decoder network to effectively improve the performance of the image segmentation task and outperforms the conventional deep networks in terms of Dice score.

An Improved Breast Cancer Nuclei Segmentation Method Based on UNet++

TLDR
This paper presents an improved UNet++ architecture, a more powerful architecture for nuclei segmentation, which has an increase in ability and is superior to U-Net and the original U- net++.

MDU-Net: Multi-scale Densely Connected U-Net for biomedical image segmentation

TLDR
This paper proposes MDUnet, a multi-scale densely connected U-net for biomedical image segmentation, and adopts the optimal model based on the experiment and proposes a novel Multi-scale Dense U-Net (MDU- net) architecture with quantization, which reduce overfitting in MDU-Net for better accuracy.

Attention Unet++: A Nested Attention-Aware U-Net for Liver CT Image Segmentation

TLDR
A nested attention-aware segmentation network, named Attention UNet++, which has a deep supervised encoder-decoder architecture and a redesigned dense skip connection and achieved very competitive performance on MICCAI 2017 Liver Tumor Segmentation Challenge Dataset.
...

References

SHOWING 1-10 OF 15 REFERENCES

H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation From CT Volumes

TLDR
This work proposes a novel hybrid densely connected UNet (H-DenseUNet), which consists of a 2-D Dense UNet for efficiently extracting intra-slice features and a 3-D counterpart for hierarchically aggregating volumetric contexts under the spirit of the auto-context algorithm for liver and tumor segmentation.

U-Net: Convolutional Networks for Biomedical Image Segmentation

TLDR
It is shown that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks.

H-DenseUNet: Hybrid Densely Connected UNet for Liver and Liver Tumor Segmentation from CT Volumes

TLDR
A novel hybrid densely connected UNet (H-Dense UNet), which consists of a 2D DenseUNet for efficiently extracting intra-slice features and a 3D counterpart for hierarchically aggregating volumetric contexts under the spirit of the auto-context algorithm for liver and tumor segmentation.

Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning?

TLDR
This paper considered four distinct medical imaging applications in three specialties involving classification, detection, and segmentation from three different imaging modalities, and investigated how the performance of deep CNNs trained from scratch compared with the pre-trained CNNs fine-tuned in a layer-wise manner.

Fully convolutional networks for semantic segmentation

TLDR
The key insight is to build “fully convolutional” networks that take input of arbitrary size and produce correspondingly-sized output with efficient inference and learning.

Residual Conv-Deconv Grid Network for Semantic Segmentation

TLDR
The GridNet is a new Convolutional Neural Network architecture for semantic image segmentation that generalizes many well known networks such as conv-deconv, residual or U-Net networks and achieves competitive results on the Cityscapes dataset.

Densely Connected Convolutional Networks

TLDR
The Dense Convolutional Network (DenseNet), which connects each layer to every other layer in a feed-forward fashion, and has several compelling advantages: they alleviate the vanishing-gradient problem, strengthen feature propagation, encourage feature reuse, and substantially reduce the number of parameters.

The Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI): a completed reference database of lung nodules on CT scans.

TLDR
The goal of this process was to identify as completely as possible all lung nodules in each CT scan without requiring forced consensus and is expected to provide an essential medical imaging research resource to spur CAD development, validation, and dissemination in clinical practice.

Et al

TLDR
A large population-based survey of veterans and nondeployed controls found evidence of a deployment-related Gulf War syndrome by factor analysis in Air Force veterans and controls.

I and J