English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

The hybrid architecture of convolutional neural networks (CNNs) andTransformer are very popular for medical image segmentation. However, itsuffers from two challenges. First, although a CNNs branch can capture thelocal image features using vanilla convolution, it cannot achieve adaptivefeature learning. Second, although a Transformer branch can capture the globalfeatures, it ignores the channel and cross-dimensional self-attention,resulting in a low segmentation accuracy on complex-content images. To addressthese challenges, we propose a novel hybrid architecture of convolutionalneural networks hand in hand with vision Transformers (CiT-Net) for medicalimage segmentation. Our network has two advantages. First, we design a dynamicdeformable convolution and apply it to the CNNs branch, which overcomes theweak feature extraction ability due to fixed-size convolution kernels and thestiff design of sharing kernel parameters among different inputs. Second, wedesign a shifted-window adaptive complementary attention module and a compactconvolutional projection. We apply them to the Transformer branch to learn thecross-dimensional long-term dependency for medical images. Experimental resultsshow that our CiT-Net provides better medical image segmentation results thanpopular SOTA methods. Besides, our CiT-Net requires lower parameters and lesscomputational costs and does not rely on pre-training. The code is publiclyavailable at https://github.com/SR0920/CiT-Net.

CiT-Net: Convolutional Neural Networks Hand in Hand with Vision Transformers for Medical Image Segmentation