AdwU-Net: Adaptive Depth and Width U-Net for Medical Image Segmentation by Differentiable Neural Architecture Search

Ziyan Huang, Zehua Wang, Zhikai Yang, Lixu Gu
Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, PMLR 172:576-589, 2022.

Abstract

The U-Net and its variants are proved as the most successful architectures in the medical image segmentation domain. However, the optimal configuration of the hyperparameters in U-Net structure such as depth and width remain challenging to adjust manually due to the diversity of medical image segmentation tasks. In this paper, we propose AdwU-Net, which is an efficient neural architecture search framework to search the optimal task-specific depth and width in the U-Net backbone. Specifically, an adaptive depth and width block is designed and applied hierarchically in U-Net. In each block, the optimal number of convolutional layers and channels in each layer are directly learned from data. To reduce the computational costs and alleviate the memory pressure, we conduct an efficient architecture search and reuse the network weights of different depth and width options in a differentiable manner. Extensive experiments on the Medical Segmentation Decathlon (MSD) dataset show that our method outperforms not only the manually scaled U-Net but also other state-of-the-art architectures. Our code is publicly available at \href{https://github.com/Ziyan-Huang/AdwU-Net}{https://github.com/Ziyan-Huang/AdwU-Net}.

Cite this Paper

BibTeX
@InProceedings{pmlr-v172-huang22b, title = {AdwU-Net: Adaptive Depth and Width U-Net for Medical Image Segmentation by Differentiable Neural Architecture Search}, author = {Huang, Ziyan and Wang, Zehua and Yang, Zhikai and Gu, Lixu}, booktitle = {Proceedings of The 5th International Conference on Medical Imaging with Deep Learning}, pages = {576--589}, year = {2022}, editor = {Konukoglu, Ender and Menze, Bjoern and Venkataraman, Archana and Baumgartner, Christian and Dou, Qi and Albarqouni, Shadi}, volume = {172}, series = {Proceedings of Machine Learning Research}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v172/huang22b/huang22b.pdf}, url = {https://proceedings.mlr.press/v172/huang22b.html}, abstract = {The U-Net and its variants are proved as the most successful architectures in the medical image segmentation domain. However, the optimal configuration of the hyperparameters in U-Net structure such as depth and width remain challenging to adjust manually due to the diversity of medical image segmentation tasks. In this paper, we propose AdwU-Net, which is an efficient neural architecture search framework to search the optimal task-specific depth and width in the U-Net backbone. Specifically, an adaptive depth and width block is designed and applied hierarchically in U-Net. In each block, the optimal number of convolutional layers and channels in each layer are directly learned from data. To reduce the computational costs and alleviate the memory pressure, we conduct an efficient architecture search and reuse the network weights of different depth and width options in a differentiable manner. Extensive experiments on the Medical Segmentation Decathlon (MSD) dataset show that our method outperforms not only the manually scaled U-Net but also other state-of-the-art architectures. Our code is publicly available at \href{https://github.com/Ziyan-Huang/AdwU-Net}{https://github.com/Ziyan-Huang/AdwU-Net}.} }
Endnote
%0 Conference Paper %T AdwU-Net: Adaptive Depth and Width U-Net for Medical Image Segmentation by Differentiable Neural Architecture Search %A Ziyan Huang %A Zehua Wang %A Zhikai Yang %A Lixu Gu %B Proceedings of The 5th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2022 %E Ender Konukoglu %E Bjoern Menze %E Archana Venkataraman %E Christian Baumgartner %E Qi Dou %E Shadi Albarqouni %F pmlr-v172-huang22b %I PMLR %P 576--589 %U https://proceedings.mlr.press/v172/huang22b.html %V 172 %X The U-Net and its variants are proved as the most successful architectures in the medical image segmentation domain. However, the optimal configuration of the hyperparameters in U-Net structure such as depth and width remain challenging to adjust manually due to the diversity of medical image segmentation tasks. In this paper, we propose AdwU-Net, which is an efficient neural architecture search framework to search the optimal task-specific depth and width in the U-Net backbone. Specifically, an adaptive depth and width block is designed and applied hierarchically in U-Net. In each block, the optimal number of convolutional layers and channels in each layer are directly learned from data. To reduce the computational costs and alleviate the memory pressure, we conduct an efficient architecture search and reuse the network weights of different depth and width options in a differentiable manner. Extensive experiments on the Medical Segmentation Decathlon (MSD) dataset show that our method outperforms not only the manually scaled U-Net but also other state-of-the-art architectures. Our code is publicly available at \href{https://github.com/Ziyan-Huang/AdwU-Net}{https://github.com/Ziyan-Huang/AdwU-Net}.
APA
Huang, Z., Wang, Z., Yang, Z. & Gu, L.. (2022). AdwU-Net: Adaptive Depth and Width U-Net for Medical Image Segmentation by Differentiable Neural Architecture Search. Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 172:576-589 Available from https://proceedings.mlr.press/v172/huang22b.html.

Related Material