[1]杨丽丽,王 玉,王明泉,等.基于交叉注意的脑部 MR 无监督配准算法研究[J].机械与电子,2024,42(11):11-16.
　YANG Lili,WANG Yu,WANG Mingquan,et al.Study on Brain MR Unsupervised Registration Algorithm Based on Cross-attention[J].Machinery & Electronics,2024,42(11):11-16.
点击复制

基于交叉注意的脑部 MR 无监督配准算法研究()
分享到：

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:
42
期数:
2024年11期
页码:
11-16
栏目:
研究与设计
出版日期:
2024-11-27

文章信息/Info

Title:
Study on Brain MR Unsupervised Registration Algorithm Based on Cross-attention
文章编号:
1001-2257 ( 2024 ) 11-0011-06
作者:
杨丽丽; 王 玉; 王明泉; 商奥雪; 崔瑞杰; 商 然
中北大学信息与通信工程学院,山西 太原 030051
Author(s):
YANG Lili ;  WANG Yu ;  WANG Mingquan ;  SHANG Aoxue ;  CUI Ruijie ;  SHANG Ran
( School of Information and Communication Engineering , North University of China , Taiyuan 030051 , China )
关键词:
深度学习; 图像配准;  UNet 网络; 交叉注意力
Keywords:
deep learning ;  image registration ;  UNet network ;  cross attention
分类号:
TP391.4
文献标志码:
A
摘要:
针对三维图像配准存在易丢失空间信息、拓扑结构保持困难及耗时长等问题,提出一种混合的 Transformer-ConvNet 模型。在经典 VoxelMorph 模型基础上,引入了交叉注意力模块,进行有效的远程建模和高维数据处理;针对运算量大的问题,在卷积层引入了 inception 模块,采用并行连接的方式将多个不同尺寸和不同类型的卷积层相互融合,学习细粒度特征,以提高配准精度。在脑部 MR 数据集上,进行了消融实验,结果表明,引入交叉注意力机制和 inception 模块网络较 VoxelMorph , Dice 系数提高了 11.5% 。将所提算法与 4 种经典算法进行了对比,结果表明,所提模型配准性能有了明显提升,较 VoxelMorph 、 CycleMorph 、 ViT-V-Net 、 TransMorph 分别提高 9.6% 、 8.7% 、 9.1% 、 6.3% ,且参数量较少。
Abstract:
For 3D image registration , there are some problems such as easy loss of spatial information , being difficult to effectively maintain topological structure and time-consuming.Therefore , a mixed Transformer ConvNet model is proposed.Based on the classical VoxelMorph model , cross attention module is introduced to carry out effective remote modeling and high-dimensional data processing ; to solve the problem of large amount of computation , inception module is introduced into the convolutional layer , which uses parallel connection to integrate multiple convolution layers of different sizes and types to learn fine-grained features to improve registration accuracy.Ablation experiments are performed on the brain MR Dataset , and the results show that the Dice score is improved by 11.5% compared with the VoxelMorph model by introducing the cross-attention mechanism and inception module network.The proposed algorithms are compared with four classical algorithms.The result show that the registration performance of the proposed model is significantly improved , which is 9.6% , 8.7% , 9.1% and 6.3% higher than that of theVoxelMorph , the CycleMorph , the ViT-V-Net and the TransMorph , respectively , and the number of parameters is small.

参考文献/References:

[ 1 ] SOTIRAS A , DAVATZIKOS C , PARAGIOS N.Deformable medical image registration : A survey [ J ] . IEEE Transactions on medical imaging , 2013 , 32 ( 7 ):1153-1190.

[ 2 ] DE VOS B D , BERENDSEN F F , VIERGEVER M A , et al.End-to-end unsupervised deformable image regi stration with a convolutional neural network [ C ] ∥Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support , 2017 : 204-212.
[ 3 ] FAN J F , CAO X H , YAP P T , et al.BIRNet : brain image registration using dual-supervised fully convolutional networks [ J ] .Medical image analysis , 2019 , 54 : 193-206.
[ 4 ] MIAO S , WANG Z J , LIAO R.A CNN regression approach for real-time 2D / 3D registration [ J ] .IEEE Transactions on medical imaging , 2016 , 35 ( 5 ): 1352-1363.
[ 5 ] BALAKRISHNAN G , ZHAO A , SABUNCU M R , et al.VoxelMorph : a learning framework for deformable medical image registration [ J ] .IEEE Transactions on medical imaging , 2019 , 38 ( 8 ): 1788-1800.
[ 6 ] KIM H , SONG W J.LAS : locality-aware scheduling for GEMM-accelerated convolutions in GPUs [ J ] . IEEE Transactions on parallel and distributed systems , 2023 , 34 ( 5 ): 1479-1494.
[ 7 ] GHAHREMANI M , KHATERI M , JIAN B L , et al.H ViT : a hierarchical vision transformer for deformable image registration [ C ] ∥Proceedings of the IEEE / CVF Conference on Computer Vision and Pattern Recognition , 2024 : 11513-11523.
[ 8 ] CHEN J Y , FREY E C , HE Y F , et al.TransMorph : transformer for unsupervised medical image registration [ J ] .Medical image analysis , 2022 , 82 : 1-34.
[ 9 ] LIU Z , LIN Y T , CAO Y , et al.Swin transformer : hierarchical vision transformer using shifted windows [ C ] ∥ IEEE / CVF International Conference on Computer Vision ( ICCV ), 2021 : 9992-10002.
[ 10 ] LIU L H , HUANG Z N , LI? P , et al.You only look at patches : a patch-wise framework for 3D unsupervised medical image registration [ C ] ∥International Workshop on Biomedical Image Registration , 2022 : 190-193.
[ 11 ] SHI J C , HE Y T , KONG Y Y , et al.Xmorpher : full transformer for deformable medical image registration via cross attention [ C ] ∥ International Conference on Medical Image Computing and Computer-Assisted Intervention ( MICCAI ), 2022 : 217-226.
[ 12 ] KIM H H , YU S Z , YUAN S , et al.Cross-attention transformer for video interpolation [ C ] ∥ Asian Conference on Computer Vision , 2022 : 320-337.
[ 13 ] 邹茂扬,杨昊,潘光晖,等 . 深度学习在医学图像配准上的研究进展与挑战 [ J ] . 生物医学工程学杂志,2019, 36 ( 4 ): 677-683.
[ 14 ] 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 [ J ] .IEEE Transactions on medical imaging ,2022 , 42 ( 3 ): 697-712.
[ 15 ] GUO M H , LIU Z N , MU T J , et al.Beyond self-attention : external attention using two linear layers for visual tasks [ J ] .IEEE Transactions on pattern analysis and machine intelligence , 2022 , 45 ( 5 ): 5436-5447.
[ 16 ] CHEN J Y , LIU Y H , HE Y F , et al.Deformable cross-attention transformer for medical image registration [ C ] ∥International Workshop on Machine Learning in Medical Imaging , 2023 : 115-125.
[ 17 ] KUMTHEKAR A , REDDY G R.An integrated deep learning framework of U-Net and inception module for cloud detection of remote sensing images [ J ] .Arabian journal of geosciences , 2021 , 14 ( 18 ): 1900.
[ 18 ] CHEN Y P , DAI X Y , LIU M C , et al.Dynamic convolution : attention over convolution kernels [ C ] ∥2020 IEEE / CVF Conference on Computer Vision and Pattern Recognition , 2020 : 11030-11039.
[ 19 ] WAN Z H , YANG H , XU J P , et al.BACNN : multiscale feature fusion based bilinear attention convolutional neural network for wood NIR classification [ J ] . Journal of forestry research , 2024 , 35 : 4.
[ 20 ] KIM B , KIM D H , PARK S H , et al.CycleMorph : cycle consistent unsupervised deformable image registration [ J ] .Medical image analysis , 2021 , 71 : 102036.

相似文献/References:

[1]王骁贤,张保华,陆思良.基于连续小波变换和卷积神经网络的无刷直流电机故障诊断[J].机械与电子,2018,(06):29.
　WANG Xiaoxian,ZHANG Baohua,LU Siliang.Fault Diagnosis of Brushless Direct Current Motor Based on Continuous Wavelet Transform and Convolutional Neural Network[J].Machinery & Electronics,2018,(11):29.
[2]刘志宇,黄亦翔.基于深度学习和迁移学习的液压泵健康评估方法[J].机械与电子,2018,(09):67.
　LIU Zhiyu,HUANG Yixiang.Health Assessment for Hydraulic Pump Based on Deep Learning and Transfer Learning[J].Machinery & Electronics,2018,(11):67.
[3]肖倩宏,康 鹏,杜 江,等.深度学习在电网智能调控系统中应用研究[J].机械与电子,2021,(01):38.
　XIAO Qianhong,KANG Peng,DU Jiang,et al.Research on the Application of Deep Learning Theory in Power Grid Intelligent Dispatching[J].Machinery & Electronics,2021,(11):38.
[4]许 哲,张少帅,郭 璐,等.无人机深度学习去雾算法[J].机械与电子,2021,(04):13.
　XU Zhe,ZHANG Shaoshuai,GUO Lu,et al. Deep Learning Defogging Algorithm for UAV[J].Machinery & Electronics,2021,(11):13.
[5]齐爱玲１,李 琳１,朱亦轩２,等.基于融合特征的双通道CNN滚动轴承故障识别[J].机械与电子,2021,(05):15.
　QI Ailing,LI Lin,ZHU Yixuan,et al.Dual Channel CNN Bearing Fault Identification Based on Fusion Feature[J].Machinery & Electronics,2021,(11):15.
[6]徐先峰,郑少杰,赵 依,等.基于数据分解与重构的光伏发电功率超短期预测[J].机械与电子,2022,(04):20.
　XU Xianfeng,ZHENG Shaojie,ZHAO Yi,et al.Ultra-short-term Prediction of Photovoltaic Power Generation Based on Data Decomposition and Deconstruction[J].Machinery & Electronics,2022,(11):20.
[7]江 励,熊达明,汤健华,等.自然光线环境中的空间物体快速识别和定位算法研究[J].机械与电子,2022,(06):8.
　JIANG Li,XIONG Daming,TANG Jianhua,et al.Recognition and Positioning Algorithm of Space Objects in Natural Light Environment[J].Machinery & Electronics,2022,(11):8.
[8]王西志,管声启,张理博,等.基于视觉引导的工业棒材上料系统研究[J].机械与电子,2023,41(05):19.
　WANG Xizhi,GUAN Shengqi,ZHANG Libo,et al.Research on Industrial Bar Feeding System Based on Visual Guidance[J].Machinery & Electronics,2023,41(11):19.
[9]王 青,吕绪山,党 帅,等.基于深度学习的纱管识别方法研究[J].机械与电子,2023,41(12):20.
　WANG Qing,LYU Xushan,DANG Shuai,et al.Research on Yarn Bobbin Detection Method Based on Deep Learning[J].Machinery & Electronics,2023,41(11):20.
[10]姜越夫,王 青,吕绪山.改进 YOLOv5s 的纱管目标检测方法[J].机械与电子,2024,42(02):29.
　JIANG Yuefu,WANG Qing,LYU Xushan.Improved YOLOv5s Method for Yarn Tube Object Detection[J].Machinery & Electronics,2024,42(11):29.
[11]刘 源,王 玉,杨丽丽,等.脑部 MRI 图像配准中 CNN 与 Transformer 并行架构的算法研究[J].机械与电子,2025,(10):11.
　LIU Yuan,WANG Yu,YANG Lili,et al.Research on a CNN-Transformer Parallel Architecture for Brain MRI Image Registration[J].Machinery & Electronics,2025,(11):11.

备注/Memo

备注/Memo:
收稿日期: 2024-06-12
基金项目:山西省重点研发计划资助项目( 201803D121069 );山西省应用基础研究项目面上自然基金项目( 201801D121162 )
作者简介:杨丽丽 ( 1998- ),女,山西吕梁人,硕士研究生,研究方向为图像处理、图像配准;王 玉 ( 1979- ),女,山西太原人,博士,副教授,研究方向为图像处理、图像配准,通信作者。
更新日期/Last Update: 2024-12-09