MF-Net: multi-scale feature extraction-integration network for unsupervised deformable registration

作者全名：Li, Andi; Ying, Yuhan; Gao, Tian; Zhang, Lei; Zhao, Xingang; Zhao, Yiwen; Song, Guoli; Zhang, He

作者地址：[Li, Andi; Ying, Yuhan; Gao, Tian; Zhao, Xingang; Zhao, Yiwen; Song, Guoli] Chinese Acad Sci, Shenyang Inst Automat, State Key Lab Robot, Shenyang, Peoples R China; [Li, Andi; Ying, Yuhan; Gao, Tian; Zhao, Xingang; Zhao, Yiwen; Song, Guoli] Chinese Acad Sci, Inst Robot & Intelligent Mfg, Shenyang, Peoples R China; [Li, Andi; Ying, Yuhan] Univ Chinese Acad Sci, Beijing, Peoples R China; [Gao, Tian] Shenyang Ligong Univ, Sch Automat & Elect Engn, Shenyang, Peoples R China; [Zhang, Lei] China Med Univ, Shengjing Hosp, Spine Surg Unit, Shenyang, Peoples R China; [Zhang, He] Chongqing Med Univ, Affiliated Hosp 2, Orthoped Dept, Chongqing, Peoples R China

通信作者：Zhang, H (通讯作者)，Chongqing Med Univ, Affiliated Hosp 2, Orthoped Dept, Chongqing, Peoples R China.

来源：FRONTIERS IN NEUROSCIENCE

ESI学科分类：NEUROSCIENCE & BEHAVIOR

WOS号：WOS:001208085000001

JCR分区：Q2

影响因子：3.2

年份：2024

卷号：18

期号：　

开始页：　

结束页：　

文献类型：Article

关键词：deformable image registration; unsupervised learning; convolutional neural network; multi-scale; gating mechanism

摘要：Deformable registration plays a fundamental and crucial role in scenarios such as surgical navigation and image-assisted analysis. While deformable registration methods based on unsupervised learning have shown remarkable success in predicting displacement fields with high accuracy, many existing registration networks are limited by the lack of multi-scale analysis, restricting comprehensive utilization of global and local features in the images. To address this limitation, we propose a novel registration network called multi-scale feature extraction-integration network (MF-Net). First, we propose a multiscale analysis strategy that enables the model to capture global and local semantic information in the image, thus facilitating accurate texture and detail registration. Additionally, we introduce grouped gated inception block (GI-Block) as the basic unit of the feature extractor, enabling the feature extractor to selectively extract quantitative features from images at various resolutions. Comparative experiments demonstrate the superior accuracy of our approach over existing methods.

基金机构：National Key R&D Program of China

基金资助正文：No Statement Available