FusionU-Net: U-Net with Enhanced Skip Connection for Pathology Image Segmentation

Zongyi Li, Hongbing Lyu, Jun Wang
Proceedings of the 15th Asian Conference on Machine Learning, PMLR 222:694-706, 2024.

Abstract

In recent years, U-Net and its variants have been widely used in pathology image segmentation tasks. One of the key designs of U-Net is the use of skip connections between the encoder and decoder, which helps to recover detailed information after upsampling. While most variations of U-Net adopt the original skip connection design, there is semantic gap between the encoder and decoder that can negatively impact model performance. Therefore, it is important to reduce this semantic gap before conducting skip connection. To address this issue, we propose a new segmentation network called FusionU-Net, which is based on U-Net structure and incorporates a fusion module to exchange information between different skip connections to reduce semantic gaps. Unlike the other fusion modules in existing networks, ours is based on a two-round fusion design that fully considers the local relevance between adjacent encoder layer outputs and the need for bi-directional information exchange across multiple layers. We conducted extensive experiments on multiple pathology image datasets to evaluate our model and found that FusionU-Net achieves better performance compared to other competing methods. We argue our fusion module is more effective than the designs of existing networks, and it could be easily embedded into other networks to further enhance the model performance. Our code is available at: https://github.com/Zongyi-Lee/FusionU-Net

Cite this Paper

BibTeX
@InProceedings{pmlr-v222-li24a, title = {{FusionU-Net}: {U-Net} with Enhanced Skip Connection for Pathology Image Segmentation}, author = {Li, Zongyi and Lyu, Hongbing and Wang, Jun}, booktitle = {Proceedings of the 15th Asian Conference on Machine Learning}, pages = {694--706}, year = {2024}, editor = {Yanıkoğlu, Berrin and Buntine, Wray}, volume = {222}, series = {Proceedings of Machine Learning Research}, month = {11--14 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v222/li24a/li24a.pdf}, url = {https://proceedings.mlr.press/v222/li24a.html}, abstract = { In recent years, U-Net and its variants have been widely used in pathology image segmentation tasks. One of the key designs of U-Net is the use of skip connections between the encoder and decoder, which helps to recover detailed information after upsampling. While most variations of U-Net adopt the original skip connection design, there is semantic gap between the encoder and decoder that can negatively impact model performance. Therefore, it is important to reduce this semantic gap before conducting skip connection. To address this issue, we propose a new segmentation network called FusionU-Net, which is based on U-Net structure and incorporates a fusion module to exchange information between different skip connections to reduce semantic gaps. Unlike the other fusion modules in existing networks, ours is based on a two-round fusion design that fully considers the local relevance between adjacent encoder layer outputs and the need for bi-directional information exchange across multiple layers. We conducted extensive experiments on multiple pathology image datasets to evaluate our model and found that FusionU-Net achieves better performance compared to other competing methods. We argue our fusion module is more effective than the designs of existing networks, and it could be easily embedded into other networks to further enhance the model performance. Our code is available at: https://github.com/Zongyi-Lee/FusionU-Net} }
Endnote
%0 Conference Paper %T FusionU-Net: U-Net with Enhanced Skip Connection for Pathology Image Segmentation %A Zongyi Li %A Hongbing Lyu %A Jun Wang %B Proceedings of the 15th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Berrin Yanıkoğlu %E Wray Buntine %F pmlr-v222-li24a %I PMLR %P 694--706 %U https://proceedings.mlr.press/v222/li24a.html %V 222 %X In recent years, U-Net and its variants have been widely used in pathology image segmentation tasks. One of the key designs of U-Net is the use of skip connections between the encoder and decoder, which helps to recover detailed information after upsampling. While most variations of U-Net adopt the original skip connection design, there is semantic gap between the encoder and decoder that can negatively impact model performance. Therefore, it is important to reduce this semantic gap before conducting skip connection. To address this issue, we propose a new segmentation network called FusionU-Net, which is based on U-Net structure and incorporates a fusion module to exchange information between different skip connections to reduce semantic gaps. Unlike the other fusion modules in existing networks, ours is based on a two-round fusion design that fully considers the local relevance between adjacent encoder layer outputs and the need for bi-directional information exchange across multiple layers. We conducted extensive experiments on multiple pathology image datasets to evaluate our model and found that FusionU-Net achieves better performance compared to other competing methods. We argue our fusion module is more effective than the designs of existing networks, and it could be easily embedded into other networks to further enhance the model performance. Our code is available at: https://github.com/Zongyi-Lee/FusionU-Net
APA
Li, Z., Lyu, H. & Wang, J.. (2024). FusionU-Net: U-Net with Enhanced Skip Connection for Pathology Image Segmentation. Proceedings of the 15th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 222:694-706 Available from https://proceedings.mlr.press/v222/li24a.html.

Related Material