Abstract
Medical image segmentation plays a critical role in assisting diagnosis and prognosis. Since the transformer was first introduced to the field, the neural network structure has experienced a transition from ConvNet into Transformer. However, some redesigned ConvNets in recent works show astonishing effects, which outperforms classic elaborate ConvNets, and even complicated transformers. Inspired by these works, we introduced large-kernel convolutions to improve the ConvNets in capturing the long-range dependency. Cooperated with a novel multi-scale feature fusion method, we proposed a U-shaped convolutional structure, dubbed UMixer, which effectively integrates shallow spatial information with deep semantic information and high-resolution detailed information with low-resolution global information. Without any attention mechanism and pre-training on large datasets, UMixer achieves more accurate segmentation results than traditional ConvNets and Transformers on the Synapse dataset. Experiments demonstrate the effectiveness of this multi-scale feature fusion structure and its capability in modeling long-range dependency.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3431–3440 (2015)
Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-24574-4_28
Zhou, Z., Rahman Siddiquee, M.M., Tajbakhsh, N., Liang, J.: UNet++: a nested U-Net architecture for medical image segmentation. In: Stoyanov, D., et al. (eds.) DLMIA/ML-CDS -2018. LNCS, vol. 11045, pp. 3–11. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-00889-5_1
Dosovitskiy, A., et al.: An image is worth 16 \(\times \) 16 words: transformers for image recognition at scale (2020)
Chen, J., et al.: TransUNet: transformers make strong encoders for medical image segmentation (2021)
Valanarasu, J.M.J., Oza, P., Hacihaliloglu, I., Patel, V.M.: Medical transformer: gated axial-attention for medical image segmentation (2021)
Xie, E., Wang, W., Yu, Z., Anandkumar, A., Alvarez, J.M., Luo, P.: SegFormer: simple and efficient design for semantic segmentation with transformers (2021a)
Cao, H., et al.: Swin-Unet: Unet-like pure trans-former for medical image segmentation (2021)
Huang, X., Deng, Z., Li, D., Yuan, X.: MISSFormer: an effective medical image segmentation transformer (2021)
Liu, Z., Mao, H., Wu, C.-Y., Feichtenhofer, C., Darrell, T., Xie, S.: A ConvNet for the 2020s (2022)
Trockman, A., Zico Kolter, J.: Patchecs Are All You Need? (2022)
Wang, H., Cao, P., Wang, J., Zaiane, O. R.: UCTransNet: rethinking the skip connections in U-Net from a channel-wise perspective with transformer (2021)
Ding, X., Zhang, X., Zhou, Y., Han, J., Ding, G., Sun, J.: Scaling up your kernels to 31 \(\times \) 31: revisiting large kernel design in CNNs (2022)
Oktay, O.: Attention U-Net: learning where to look for the pancreas (2018)
Islam, M.A., Jia, S., Bruce, N.D.: How much position information do convolutional neural networks encode? (2020)
Chu, X., et al.: Conditional positional encodings for vision transformers (2021b)
Li, Y., Zhang, K., Cao, J., Timofte, R., Van Gool, L.: LocalViT: bringing locality to vision transformers (2021)
Wang, Z., Cun, X., Bao, J., Liu, J.: Uformer: a general U-shaped transformer for image restoration (2021d)
Acknowledgments
This work is supported by the Beijing municipal education committee scientific and technological planning Project (KM201811232024), and Beijing Information Science and Technology University Research Fund (2021XJJ30, 2021XJJ34).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Su, Y., Huang, H., Song, Z., Lin, L., Liu, J. (2022). UMixer: A Novel U-shaped Convolutional Mixer for Multi-scale Feature Fusion in Medical Image Segmentation. In: Deng, W., et al. Biometric Recognition. CCBR 2022. Lecture Notes in Computer Science, vol 13628. Springer, Cham. https://doi.org/10.1007/978-3-031-20233-9_70
Download citation
DOI: https://doi.org/10.1007/978-3-031-20233-9_70
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-20232-2
Online ISBN: 978-3-031-20233-9
eBook Packages: Computer ScienceComputer Science (R0)