WaveTrans: A Wavelet Pyramid Network for Medical Image Registration with Swin Transformer

作者：Chuanbao Zhou, Wei Teng, Zhanli Hu, Na Zhang, Wenjian Qin, Yaoqin Xie · 年份：2026 · DOI：10.1109/nnice68970.2026.11465517 · 研究领域：Image and Signal Denoising Methods、Brain Tumor Detection and Classification、Medical Image Segmentation Techniques

Deformable image registration plays an important role in medical image analysis. Recent deep learning-based methods typically adopt either U-Net or pyramid architectures. However, U-Net-based approaches often struggle with large deformations, while pyramid methods—despite better handling of large deformations—suffer from irreversible detail loss and accumulated interpolation errors due to multi-scale deformation field composition and repeated sampling operations. To address these limitations, we propose WaveTrans, a wavelet pyramid registration framework that decomposes images into wavelet coefficients, fuses multi-scale frequency features through a Swin Transformer backbone, and reconstructs deformation fields via inverse wavelet transform. Experiments on two diverse registration tasks demonstrate that WaveTrans improves registration accuracy over spatial-domain baselines and markedly outperforms prior wavelet-based approaches on large-deformation cases.