|
|
|
| Multi-window Multi-layer Perceptron Feature Pyramid Network for Unsupervised MRI Image Registration |
| YU Han1, 2, SUN Zheng1,2, ZHANG Shengnan1, GAO Zhangshuo1, DING Gang'ao1 |
1. Department of Electronic and Communication Engineering, Nor-th China Electric Power University, Baoding, 071003;
2. Hebei Key Laboratory of Power Internet of Things Technology, North China Electric Power University, Baoding 071003 |
|
|
|
|
Abstract Current unsupervised magnetic resonance imaging(MRI) registration methods are typically based on convolutional neural networks(CNNs) or Transformer architectures. However, significant limitations exist in both of them. CNNs have difficulty in modeling long-range dependencies because they are constrained by the local receptive fields. Transformers often struggle to achieve fine-grained registration at full image resolution due to the high computational complexity of the self-attention mechanism. To address these issues, a multi-window multi-layer perceptron feature pyramid network(PyraMLP-Net) is proposed. It is designed for efficient and accurate full-resolution brain MRI registration. First, a weight-sharing feature extraction module extracts multi-scale features from a pair of images through parallel dual-path convolutional encoding. Then, with the correlation-aware multi-window multi-layer perceptron being its core, a feature pyramid decoding module gradually fuses feature information of different scales through a bottom-up path to achieve coarse-to-fine optimization of the deformation field. Finally, a spatial transformation network module applies the deformation field as parameters to perform differentiable resampling on the image to be registered and generate the final registration result. Experiments on three public datasets demonstrate that PyraMLP-Net outperforms mainstream models in terms of registration accuracy, stability and efficiency.
|
|
Received: 22 October 2025
|
|
|
| Fund:National Natural Science Foundation of China(No.62571188,62071181) |
|
Corresponding Authors:
SUN Zheng, Ph.D., professor. Her research interests include photoelectric instrument and image processing.
|
About author:: YU Han, Master student. Her research interests include image processing and deep learning.
ZHANG Shengnan, Master student. Her research interests include image processing and deep learning.
GAO Zhangshuo, Master student. Her research interests include medical image processing.
DING Gang'ao, Master student. His research interests include biomedical imaging. |
|
|
|
[1] 黄一超,傅锐芝,王昕辰,等.医学影像图像与组织学图像配准的研究进展.中国医学计算机成像杂志, 2024, 30(5): 646-652.
(HUANG Y C, FU R Z, WANG X C, et al. Research Progress on Registration of Medical Image and Histological Image. Chinese Computed Medical Imaging, 2024, 30(5): 646-652)
[2] 冯筠,邓佳慧,周末,等.多模态医学图像配准算法综述.华中科技大学学报(自然科学版), 2024, 52(5): 29-49, 157.
(FENG J, DENG J H, ZHOU M, et al. Review of Multimodal Medical Image Registration Algorithm. Journal of Huazhong University of Science and Technology(Natural Science Edition), 2024, 52(5): 29-49, 157)
[3] CHEN J Y, LIU Y H, WEI S W, et al. A Survey on Deep Learning in Medical Image Registration: New Technologies, Uncertainty, Evaluation Metrics, and Beyond. Medical Image Analysis, 2025, 100. DOI: 10.1016/j.media.2024.103385.
[4] LIU S, YANG B, WANG Y, et al. 2D/3D Multimode Medical Ima-ge Registration Based on Normalized Cross-Correlation. Applied Sciences, 2022, 12(6). DOI: 10.3390/app12062828.
[5] 彭昆,张桂梅,王杰,等.基于可变形卷积和多尺度特征聚焦的X 线图像非刚性配准.生物医学工程学杂志, 2023, 40(3): 492-498, 507.
(PENG K, ZHANG G M, WANG J, et al. Non-rigid Registration for Medical Images Based on Deformable Convolution and Multiscale Feature Focusing Modules. Journal of Biomedical Engineering, 2023, 40(3): 492-498, 507)
[6] EGOZI A, KELLER Y, GUTERMAN H.A Probabilistic Approach to Spectral Graph Matching. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(1): 18-27.
[7] DUAN T S, CHEN W K, RUAN M L, et al. Unsupervised Deep Learning-Based Medical Image Registration: A Survey. Physics in Medicine and Biology, 2025, 70(2). DOI: 10.1088/1361-6560/ad9e69.
[8] 应时辉,杨菀,杜少毅,等.基于深度学习的医学影像配准综述.模式识别与人工智能, 2021, 34(4): 287-299.
(YING S H, YANG W, DU S Y, et al. Deep Learning Based Medi-cal Image Registration: A Review. Pattern Recognition and Artificial Intelligence, 2021, 34(4): 287-299)
[9] ABBASI S, TAVAKOLI M, BOVEIRI H R, et al. Medical Image Registration Using Unsupervised Deep Neural Network: A Scoping Literature Review. Biomedical Signal Processing and Control, 2022, 73. DOI: 10.1016/j.bspc.2021.103444.
[10] JADERBERG M, SIMONYAN K, ZISSERMAN A, et al. Spatial Transformer Network[C/OL].[2025-10-09]. https://arxiv.org/pdf/1506.02025.
[11] BALAKRISHNAN G, ZHAO A, SABUNCU M R, et al. VoxelMorph: A Learning Framework for Deformable Medical Image Regi-stration. IEEE Transactions on Medical Imaging, 2019, 38(8): 1788-1800.
[12] MA M R, XU Y B, SONG L, et al. Symmetric Transformer-Based Network for Unsupervised Image Registration. Knowledge-Based Systems, 2022, 257. DOI: 10.1016/j.knosys.2022.109959.
[13] CHEN J Y, FREY E C, HE Y F,et al. TransMorph: Transformer for Unsupervised Medical Image Registration. Medical Image Ana-lysis, 2022, 82. DOI: 10.1016/j.media.2022.102615.
[14] SHI J C, HE Y T, KONG Y Y, et al. XMorpher: Full Transfor-mer for Deformable Medical Image Registration via Cross Attention // Proc of the 25th International Conference on Medical Image Computing and Computer-Assisted Intervention. Berlin, Germany: Springer, 2022: 217-226.
[15] CHEN Z Y, ZHENG Y J, GEE J C.TransMatch: A Transformer-Based Multilevel Dual-Stream Feature Matching Network for Unsupervised Deformable Image Registration. IEEE Transactions on Medical Imaging, 2024, 43(1): 15-27.
[16] MENG M Y, BI L, FULHAM M, et al. Non-iterative Coarse-to-Fine Transformer Networks for Joint Affine and Deformable Image Registration // Proc of the 26th International Conference on Medical Image Computing and Computer Assisted Intervention. Berlin, Germany: Springer, 2023: 750-760.
[17] ZHOU H Y, GUO J S, ZHANG Y H, et al. nnFormer: Volumetric Medical Image Segmentation via a 3D Transformer. IEEE Transactions on Image Processing, 2023, 32: 4036-4045.
[18] MENG M Y, FENG D G, BI L, et al. Correlation-Aware Coarse-to-Fine MLPs for Deformable Medical Image Registration // Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2024: 9645-9654.
[19] LIU H X, DAI Z H, SO D R, et al.Pay Attention to MLPs // Proc of the 35th International Conference on Neural Information Processing Systems. Cambridge, USA: MIT Press, 2021: 9204-9215.
[20] HU J, SHEN L, ALBANIE S, et al. Squeeze-and-Excitation Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(8): 2011-2023.
[21] HERING A, HANSEN L, MOK T C W, et al. Learn2Reg: Comprehensive Multi-task Medical Image Registration Challenge, Dataset and Evaluation in the Era of Deep Learning. IEEE Transactions on Medical Imaging, 2022, 42(3): 697-712.
[22] MARCUS D S, WANG T H, PARKER J, et al. Open Access Series of Imaging Studies(OASIS): Cross-Sectional MRI Data in Young, Middle Aged, Nondemented, and Demented Older Adults. Journal of Cognitive Neuroscience, 2007, 19(9): 1498-1507.
[23] SHATTUCK D W, MIRZA M, ADISETIYO V, et al. Construction of a 3D Probabilistic Atlas of Human Cortical Structures. Neuro-Image, 2008, 39(3): 1064-1080.
[24] FISCHL B.FreeSurfer. NeuroImage, 2012, 62(2): 774-781.
[25] YU T, LI X, CAI Y F, et al. S2-MLP: Spatial-Shift MLP Architecture for Vision // Proc of the IEEE/CVF Winter Conference on Applications of Computer Vision. Washington, USA: IEEE, 2022: 3615-3624.
[26] TANG C X, ZHAO Y C, WANG G T, et al. Sparse MLP for Image Recognition: Is Self-Attention Really Necessary. Proc of the AAAI Conference on Artificial Intelligence, 2022, 36(2): 2344-2351. |
|
|
|
2026, Vol. 39 
