Towards Bi-directional Skip Connections in Encoder-Decoder Architectures and Beyond

  title={Towards Bi-directional Skip Connections in Encoder-Decoder Architectures and Beyond},
  author={Tiange Xiang and Chaoyi Zhang and Xinyi Wang and Yang Song and Dongnan Liu and Heng Huang and Weidong (Tom) Cai},
  journal={Medical image analysis},
U-Net, as an encoder-decoder architecture with forward skip connections, has achieved promising results in various medical image analysis tasks. Many recent approaches have also extended U-Net with more complex building blocks, which typically increase the number of network parameters considerably. Such complexity makes the inference stage highly inefficient for clinical applications. Towards an effective yet economic segmentation network design, in this work, we propose backward skip… 

Unsupervised Domain Adaptive Fundus Image Segmentation with Few Labeled Source Data

This method has outperformed several state-of-the-art UDA segmentation methods under the UDA fundus segmentation with few labeled source data and a cross-style self-supervised learning stage is further designed to improve the segmentation performance on the target images.



MS-NAS: Multi-Scale Neural Architecture Search for Medical Image Segmentation

A Multi-Scale NAS framework that is featured with multi- scale search space from network backbone to cell operation, and multi-scale fusion capability to fuse features with different sizes, and a partial channel connection scheme and a two-step decoding method are utilized to reduce computational overhead while maintaining optimization quality.

Recurrent U-Net for Resource-Constrained Segmentation

This paper introduces a novel recurrent U-Net architecture that preserves the compactness of the original U- net, while substantially increasing its performance to the point where it outperforms the state of the art on several benchmarks.

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

This paper presents UNet++, a new, more powerful architecture for medical image segmentation where the encoder and decoder sub-networks are connected through a series of nested, dense skip pathways, and argues that the optimizer would deal with an easier learning task when the feature maps from the decoder and encoder networks are semantically similar.

U-Net: Convolutional Networks for Biomedical Image Segmentation

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.

A large annotated medical image dataset for the development and evaluation of segmentation algorithms

A large, curated dataset representative of several highly variable segmentation tasks that was used in a crowd-sourced challenge - the Medical Segmentation Decathlon held during the 2018 Medical Image Computing and Computer Aided Interventions Conference in Granada, Spain.

International Conference on Learning Representations (ICLR)

Adam: A Method for Stochastic Optimization

This work introduces Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments, and provides a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework.

nnunet: Self-adapting framework for u-net-based medical image segmentation

  • 2018

Categorical reparameterization with gumbelsoftmax

  • in: International Conference on Learning Representations (ICLR)
  • 2017