Mixed Transformer U-Net for Medical Image Segmentation
By Hongyi Wang, Shiao Xie, Lanfen Lin, Yutaro Iwamoto, X. Han, Yenwei Chen, Ruofeng Tong. IEEE International Conference on Acoustics, Speech, and Signal Processing, 2021
424 citations recorded by the source corpus.
Though U-Net has achieved tremendous success in medical image segmentation tasks, it lacks the ability to explicitly model long-range dependencies. Therefore, Vision Transformers have emerged as alternative segmentation structures recently, for their innate ability of capturing long-range correlations through Self-Attention (SA). However, Transformers usually rely on large-scale pre-training and have high computational complexity. Furthermore, SA can only model self-affinities within a single sample, ignoring the potential correlations of the overall dataset. To address these problems, we propose a novel Transformer module named Mixed Transformer Module (MTM) for simultaneous inter- and intra- affinities learning. MTM first calculates self-affinities efficiently through our well-designed Local-Global Gaussian-Weighted Self-Attention (LGG-SA). Then, it mines inter-connections between data samples through External Attention (EA). By using MTM, we construct a U-shaped model named Mixed Transformer U-Net (MT-UNet) for accurate medical image segmentation. We test our method on two different public datasets, and the experimental results show that the proposed method achieves better performance over other state-of-the-art methods. The code is available at: https://github.com/Dootmaan/MT-UNet.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/435a43bbb6d6fbbf2beb03ef17868df9d5be87e8
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Ehsan Khodapanah Aghdam
- Dinggang Shen
- Dorit Merhof
- Reza Azad
- Yiwei Jia
- Qian Wang
- Ehsan Adeli
- Islem Rekik
In Computer Vision and Pattern Recognition
- Qian Wang
- Ehsan Khodapanah Aghdam
- Dinggang Shen
- Dorit Merhof
- Reza Azad
- Zihao Yin
- Wen Ji
- Junfeng Zhang
Citations in the corpus
- Medical Image Segmentation Review: The Success of U-Net (2022, methodology): “[43] proposed the MT-UNet network utilized with the Mixed Transformer Module (MTM) to capture long-range dependencies wisely concerning the most neighboring contextual information.”
- Transformers in Medical Image Analysis: A Review (2022, methodology): “Wang et al.[95] proposed the Mixed Transformer Module (MTM), which calculates self-affinities through welldesigned Local-Global Gaussian-Weighted Self-Attention (LGG-SA) and then mines inter-connections between data samples through External Attention (EA).”