《计算机应用》唯一官方网站 ›› 2026, Vol. 46 ›› Issue (3): 924-932.DOI: 10.11772/j.issn.1001-9081.2025030322

• 多媒体计算与计算机仿真 • 上一篇    下一篇

基于频谱分解的高频保持医学图像弹性配准模型

姜勇维1,2, 陈晓清1,2(), 付麟杰1,2   

  1. 1.中国科学院 成都计算机应用研究所,成都 610213
    2.中国科学院大学 计算机科学与技术学院 北京 100049
  • 收稿日期:2025-03-28 修回日期:2025-05-12 接受日期:2025-05-13 发布日期:2025-05-27 出版日期:2026-03-10
  • 通讯作者: 陈晓清
  • 作者简介:姜勇维(2000—),男,黑龙江哈尔滨人,硕士研究生,主要研究方向:深度学习、医学图像处理
    付麟杰(1995—),重庆人,博士研究生,主要研究方向:图像配准。
  • 基金资助:
    成都市科技局项目(20240321174430664)

Elastic medical image registration model with high-frequency preservation based on spectrum decomposition

Yongwei JIANG1,2, Xiaoqing CHEN1,2(), Linjie FU1,2   

  1. 1.Chengdu Institution of Computer Application,Chinese Academy of Sciences,Chengdu Sichuan 610213,China
    2.School of Computer Science and Technology,University of Chinese Academy of Sciences,Beijing 100049,China
  • Received:2025-03-28 Revised:2025-05-12 Accepted:2025-05-13 Online:2025-05-27 Published:2026-03-10
  • Contact: Xiaoqing CHEN
  • About author:JIANG Yongwei, born in 2000, M. S. candidate. His research interests include deep learning, medical image processing.
    FU Linjie, born in 1995, Ph. D. candidate. His research interests include image registration.
  • Supported by:
    Chengdu Municipal Science and Technology Bureau Project(20240321174430664)

摘要:

弹性配准是医学图像处理中的关键任务之一,它的效果会直接影响到后续的分割、分类和预测等任务的准确性;然而,由于神经网络的高频不敏感特性,现有的方法难以捕捉图像的高频信息,这影响了配准场的拟合精度。为了解决这个问题,提出一种基于频谱分解的高频保持医学图像配准模型——DFRes (Decomposition in Frequency domain model for Registration)。该模型引入频谱分解策略,并采用双支结构处理原始图像中的高频信息和低频信息;同时,设计具有高频保持特性的可逆神经网络(INN)结构和具有高频低频融合能力的桥式特征融合模块,并通过交替的空间-频谱信息提取模块进一步加强模型对频域和空域信息的提取和融合能力。在IXI、OSSAI和华西直肠癌数据集上,DFRes与现有的先进模型对比的实验结果表明,DFRes在多个指标上取得了显著的提升。在IXI数据集上,相较于TransMorph模型,Dice相似系数(DSC)提高了2.5个百分点,平均表面距离(ASD)降低了0.012,而结构化相似性(SSIM)提高了1.6个百分点。同时,通过消融实验验证了模块设计的有效性。

关键词: 图像配准, 医学图像处理, 频谱分析, 可逆神经网络, 深度学习

Abstract:

Elastic registration is regarded as a key task in medical image processing, whose performance directly affects the accuracy of subsequent tasks such as segmentation, classification, and prediction. However, due to the insensitivity of neural networks to high-frequency components, the existing methods have difficulty in capturing high-frequency information in images, which affects the fitting accuracy of registration field. To address this issue, a high-frequency-preserving medical image registration model based on frequency spectrum decomposition — DFRes (Decomposition in Frequency domain model for Registration) was proposed. In the model, a frequency decomposition strategy was introduced, and a dual-branch structure was adopted to process high- and low-frequency information from the original image. Meanwhile, an Invertible Neural Network (INN) structure with high-frequency preservation characteristics and a bridge-style feature fusion module with ability to fuse high- and low-frequency information were designed, and an alternating spatial-frequency information extraction module was used to further enhance the model’s ability to extract and fuse frequency- and spatial-domain information. Experimental results of comparing DFRes and the existing advanced models on the IXI, OSSAI, and Huaxi rectal cancer datasets show that DFRes achieves significant improvements on multiple metrics. On IXI dataset, compared to the TransMorph model, DFRes has the Dice Similarity Coefficient (DSC) increased by 2.5 percentage points, the Average Surface Distance (ASD) reduced by 0.012, and the Structural SIMilarity (SSIM) increased by 1.6 percentage points. At the same time, the effectiveness of the module design is verified through ablation experiments.

Key words: image registration, medical image processing, spectrum analysis, Inverse Neural Network (INN), deep learning

中图分类号: