{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,8]],"date-time":"2025-12-08T10:34:35Z","timestamp":1765190075224,"version":"3.46.0"},"reference-count":69,"publisher":"Wiley","issue":"7","license":[{"start":{"date-parts":[[2025,4,25]],"date-time":"2025-04-25T00:00:00Z","timestamp":1745539200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62272135","62372135","62202092"],"award-info":[{"award-number":["62272135","62372135","62202092"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2025,7]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Accurate and efficient deformable medical image registration is crucial in medical image analysis. While recent deep learning\u2010based registration methods have achieved state\u2010of\u2010the\u2010art accuracy, they often suffer from extensive network parameters and slow inference times, leading to inefficiency. Efforts to reduce model size and input resolution can improve computational efficiency but frequently result in suboptimal accuracy.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>To address the trade\u2010off between high accuracy and efficiency, we propose a Lightweight Cross\u2010Resolution Coarse\u2010to\u2010Fine registration framework, termed\u00a0LightCRCF.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Our method is built on an ultra\u2010lightweight U\u2010Net architecture with only 0.1 million parameters, offering remarkable efficiency. To mitigate accuracy degradation resulting from fewer parameters while preserving the lightweight nature of the networks, LightCRCF introduces three key innovations as follows: (1) selecting an efficient cross\u2010resolution coarse\u2010to\u2010fine (C2F) registration strategy and integrating it into the lightweight network to progressively decompose the deformation fields into multiresolution subfields to capture fine\u2010grained deformations; (2) a Texture\u2010aware Reparameterization (TaRep) module that integrates Sobel and Laplacian operators to extract rich textural information; (3) a Group\u2010flow Reparameterization (GfRep) module that captures diverse deformation modes by decomposing the deformation field into multiple groups. Furthermore, we introduce a structural reparameterization technique that enhances training accuracy through multibranch structures of the TaRep and GfRep modules, while maintaining efficient inference by equivalently transforming these multibranch structures into single\u2010branch standard\u00a0convolutions.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>We evaluate LightCRCF against various methods on the three public MRI datasets (LPBA, OASIS, and ACDC) and one CT dataset (abdomen CT). Following the previous data division methods, the LPBA dataset comprises 30 training image pairs and nine testing image pairs. For the OASIS dataset, the training, validation, and testing data consist of 1275, 110, and 660 image pairs, respectively. Similarly, for the ACDC dataset, the training, validation, and testing data include 180, 20, and 100 image pairs, respectively. For intersubject registration of the abdomen CT dataset, there are 380 training pairs, six validation pairs, and 42 testing pairs. Compared to state\u2010of\u2010the\u2010art C2F methods, LightCRCF achieves comparable accuracy scores (DSC, HD95, and MSE), while demonstrating significantly superior performance across all efficiency metrics (Params, VRAM, FLOPs, and inference time). Relative to efficiency\u2010first approaches, LightCRCF significantly outperforms these methods in accuracy metrics.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>\n                      Our LightCRCF method offers a favorable trade\u2010off between accuracy and efficiency, maintaining high accuracy while achieving superior efficiency, thereby highlighting its potential for clinical applications. The code will be available at\n                      <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" xlink:href=\"https:\/\/github.com\/PerceptionComputingLab\/LightCRCF\">https:\/\/github.com\/PerceptionComputingLab\/LightCRCF<\/jats:ext-link>\n                      .\n                    <\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.17827","type":"journal-article","created":{"date-parts":[[2025,4,25]],"date-time":"2025-04-25T22:25:47Z","timestamp":1745619947000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Lightweight cross\u2010resolution coarse\u2010to\u2010fine network for efficient deformable medical image registration"],"prefix":"10.1002","volume":"52","author":[{"given":"Jun","family":"Liu","sequence":"first","affiliation":[{"name":"School of Computer Science and Technology Harbin Institute of Technology Harbin Heilongjiang China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Nuo","family":"Shen","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology Harbin Institute of Technology Harbin Heilongjiang China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wenyi","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology Harbin Institute of Technology Harbin Heilongjiang China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiangyu","family":"Li","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology Harbin Institute of Technology Harbin Heilongjiang China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wei","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology Harbin Institute of Technology Shenzhen Shenzhen Guangdong China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yongfeng","family":"Yuan","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology Harbin Institute of Technology Harbin Heilongjiang China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ye","family":"Tian","sequence":"additional","affiliation":[{"name":"Department of Cardiology The First Affiliated Hospital Cardiovascular Institute Harbin Medical University Harbin Heilongjiang China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gongning","family":"Luo","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology Harbin Institute of Technology Harbin Heilongjiang China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kuanquan","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology Harbin Institute of Technology Harbin Heilongjiang China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","published-online":{"date-parts":[[2025,4,25]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12256"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab843e"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.16367"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.17526"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.102292"},{"key":"e_1_2_8_7_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compmedimag.2024.102365"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1097\/RLI.0000000000001035"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2022.3141013"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2019.2897538"},{"key":"e_1_2_8_11_1","doi-asserted-by":"publisher","DOI":"10.1023\/B:VISI.0000043755.93987.aa"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2007.06.004"},{"key":"e_1_2_8_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2008.10.040"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2009.2035616"},{"key":"e_1_2_8_15_1","doi-asserted-by":"crossref","unstructured":"Roh\u00e9MM DatarM HeimannT SermesantM PennecX.SVF\u2010Net: learning deformable image registration using shape matching. In:Proceedings of the International Conference on Medical Image Computing and Computer\u2010Assisted Intervention (MICCAI).Springer;2017:266\u2010274.","DOI":"10.1007\/978-3-319-66182-7_31"},{"key":"e_1_2_8_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2017.07.008"},{"key":"e_1_2_8_17_1","doi-asserted-by":"crossref","unstructured":"MokTC ChungAC.Large deformation diffeomorphic image registration with Laplacian pyramid networks. In:Proceedings of the International Conference on Medical Image Computing and Computer\u2010Assisted Intervention (MICCAI).Springer;2020:211\u2010221.","DOI":"10.1007\/978-3-030-59716-0_21"},{"key":"e_1_2_8_18_1","doi-asserted-by":"crossref","unstructured":"JiaX BartlettJ ZhangT LuW QiuZ DuanJ.U\u2010Net vs transformer: Is U\u2010Net outdated in medical image registration?In:Proceedings of the International Workshop on Deep Learning Medical Image Analysis.Springer;2022:151\u2010160.","DOI":"10.1007\/978-3-031-21014-3_16"},{"key":"e_1_2_8_19_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2023.3288136"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2023.102740"},{"key":"e_1_2_8_21_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2024.3362968"},{"key":"e_1_2_8_22_1","doi-asserted-by":"crossref","unstructured":"LiuJ WangW ShenN et\u00a0al.Mutualreg: mutual learning for unsupervised medical image registration. In:IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP).2024:1561\u20101565.","DOI":"10.1109\/ICASSP48485.2024.10445904"},{"key":"e_1_2_8_23_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102615"},{"key":"e_1_2_8_24_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.17129"},{"key":"e_1_2_8_25_1","doi-asserted-by":"crossref","unstructured":"MengM BiL FulhamM FengD KimJ.Non\u2010iterative coarse\u2010to\u2010fine transformer networks for joint affine and deformable image registration. In:Proceedings of the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI).Springer;2023:750\u2010760.","DOI":"10.1007\/978-3-031-43999-5_71"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102612"},{"key":"e_1_2_8_27_1","doi-asserted-by":"crossref","unstructured":"ShiJ HeY KongY et\u00a0al.XMorpher: full transformer for deformable medical image registration via cross attention. In:Proceedings of the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI).2022:217\u2010226.","DOI":"10.1007\/978-3-031-16446-0_21"},{"key":"e_1_2_8_28_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2024.107990"},{"key":"e_1_2_8_29_1","doi-asserted-by":"crossref","unstructured":"GhahremaniM KhateriM JianB WiestlerB AdeliE WachingerC.H\u2010ViT: A Hierarchical Vision Transformer for Deformable Image Registration. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR).2024:11513\u201011523.","DOI":"10.1109\/CVPR52733.2024.01094"},{"key":"e_1_2_8_30_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2018.11.010"},{"key":"e_1_2_8_31_1","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2019.2951024"},{"key":"e_1_2_8_32_1","doi-asserted-by":"crossref","unstructured":"ZhaoS DongY ChangEI XuY.Recursive cascaded networks for unsupervised medical image registration. In:Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV).2019:10600\u201010610.","DOI":"10.1109\/ICCV.2019.01070"},{"key":"e_1_2_8_33_1","doi-asserted-by":"crossref","unstructured":"MengM BiL FengD KimJ.Non\u2010iterative coarse\u2010to\u2010fine registration based on single\u2010pass deep cumulative learning.In:.2022:88\u201097.","DOI":"10.1007\/978-3-031-16446-0_9"},{"key":"e_1_2_8_34_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2023.102978"},{"key":"e_1_2_8_35_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compmedimag.2023.102322"},{"key":"e_1_2_8_36_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2022.3178178"},{"key":"e_1_2_8_37_1","doi-asserted-by":"crossref","unstructured":"MaT DaiX ZhangS WenY.PIViT: large deformation image registration with pyramid\u2010iterative vision transformer. In:Proceedings of the International Conference on Medical Image Computing and Computer\u2010Assisted Intervention (MICCAI).2023:602\u2010612.","DOI":"10.1007\/978-3-031-43999-5_57"},{"key":"e_1_2_8_38_1","unstructured":"QiuH QinC SchuhA HammernikK RueckertD.Learning diffeomorphic and modality\u2010invariant registration using b\u2010splines. In:Medical Imaging with Deep Learning.2021:1\u201020."},{"key":"e_1_2_8_39_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-018-1099-x"},{"key":"e_1_2_8_40_1","doi-asserted-by":"crossref","unstructured":"JiaX BartlettJ ChenW et\u00a0al.Fourier\u2010Net: fast image registration with band\u2010limited deformation. In:Proceedings of the AAAI.37:2023:1015\u20101023.","DOI":"10.1609\/aaai.v37i1.25182"},{"key":"e_1_2_8_41_1","doi-asserted-by":"crossref","unstructured":"JiaX LuW ChengX DuanJ.Decoder\u2010only image registration.IEEE Trans Med Imaging.2025.","DOI":"10.1109\/TMI.2025.3562056"},{"key":"e_1_2_8_42_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2024.106476"},{"key":"e_1_2_8_43_1","doi-asserted-by":"crossref","unstructured":"WangH NiD WangY.ModeT: learning deformable image registration via motion decomposition transformer. In:Proceedings of the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI).2023:740\u2010749.","DOI":"10.1007\/978-3-031-43999-5_70"},{"key":"e_1_2_8_44_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2024.3400603"},{"key":"e_1_2_8_45_1","doi-asserted-by":"crossref","unstructured":"DingX ZhangX MaN HanJ DingG SunJ.RepVGG: making VGG\u2010style ConvNets great again. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR).2021:13733\u201013742.","DOI":"10.1109\/CVPR46437.2021.01352"},{"key":"e_1_2_8_46_1","doi-asserted-by":"crossref","unstructured":"DingX ZhangX HanJ DingG.Diverse branch block: building a convolution as an inception\u2010like unit. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR).2021:10886\u201010895.","DOI":"10.1109\/CVPR46437.2021.01074"},{"key":"e_1_2_8_47_1","doi-asserted-by":"crossref","unstructured":"RonnebergerO FischerP BroxT.U\u2010Net: convolutional networks for biomedical image segmentation. In:Proceedings of the International Conference on Medical Image Computing and Computer\u2010Assisted Intervention (MICCAI).Springer;2015:234\u2010241.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_2_8_48_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.16387"},{"key":"e_1_2_8_49_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102379"},{"key":"e_1_2_8_50_1","doi-asserted-by":"crossref","unstructured":"MengM FengD BiL KimJ.Correlation\u2010aware coarse\u2010to\u2010fine MLPs for deformable medical image registration. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR).2024:9645\u20109654.","DOI":"10.1109\/CVPR52733.2024.00921"},{"key":"e_1_2_8_51_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2024.103212"},{"key":"e_1_2_8_52_1","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2022.3189696"},{"key":"e_1_2_8_53_1","unstructured":"JaderbergM SimonyanK ZissermanA KavukcuogluK.Spatial transformer networks. In:Proceedings of the Advances in Neural Information Processing Systems. Vol28.2015."},{"key":"e_1_2_8_54_1","doi-asserted-by":"crossref","unstructured":"XuZ YanJ LuoJ LiX JagadeesanJ.Unsupervised multimodal image registration with adaptative gradient guidance. In:IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP).2021:1225\u20101229.","DOI":"10.1109\/ICASSP39728.2021.9414320"},{"key":"e_1_2_8_55_1","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2020.620235"},{"key":"e_1_2_8_56_1","doi-asserted-by":"crossref","unstructured":"Sideri\u2010LampretsaV KaissisG RueckertD.Multi\u2010modal unsupervised brain image registration using edge maps. In:Proceedings of the IEEE International Symposium on Biomedical Imaging (ISBI).2022:1\u20105.","DOI":"10.1109\/ISBI52829.2022.9761637"},{"key":"e_1_2_8_57_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2020.07.066"},{"key":"e_1_2_8_58_1","doi-asserted-by":"crossref","unstructured":"ZhangY ShiD WangJ XiangJ ZhangW.CCAC: Contrastive learning with channel attention and contour loss for brain image registration. In:IEEE International Conference on Bioinformatics and Biomedicine (BIBM).IEEE;2023:1704\u20101709.","DOI":"10.1109\/BIBM58861.2023.10385656"},{"key":"e_1_2_8_59_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2019.2895894"},{"key":"e_1_2_8_60_1","doi-asserted-by":"crossref","unstructured":"SunJ XuZ ShumHY.Image super\u2010resolution using gradient profile prior. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR).2008:1\u20138.","DOI":"10.1109\/CVPR.2008.4587659"},{"key":"e_1_2_8_61_1","doi-asserted-by":"crossref","unstructured":"GouW YiZ XiangY et\u00a0al.SYENet: A simple yet effective network for multiple low\u2010level vision tasks with real\u2010time performance on mobile device. In:Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV).2023:12148\u201012161.","DOI":"10.1109\/ICCV51070.2023.01119"},{"key":"e_1_2_8_62_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2007.09.031"},{"key":"e_1_2_8_63_1","doi-asserted-by":"publisher","DOI":"10.1162\/jocn.2007.19.9.1498"},{"key":"e_1_2_8_64_1","doi-asserted-by":"crossref","unstructured":"HoopesA HoffmannM FischlB GuttagJ DalcaAV.Hypermorph: amortized hyperparameter learning for image registration. In:Proceedings of the International Conference on Information Processing in Medical Imaging (IPMI).2021:3\u201017.","DOI":"10.1007\/978-3-030-78191-0_1"},{"key":"e_1_2_8_65_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2022.3213983"},{"key":"e_1_2_8_66_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2018.2837502"},{"key":"e_1_2_8_67_1","doi-asserted-by":"crossref","unstructured":"QinY LiX.FSDiffReg: feature\u2010wise and score\u2010wise diffusion\u2010guided unsupervised deformable image registration for cardiac images. In:Proceedings of the International Conference on Medical Image Computing and Computer\u2010Assisted Intervention (MICCAI).2023:655\u2010665.","DOI":"10.1007\/978-3-031-43999-5_62"},{"key":"e_1_2_8_68_1","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2016.2574816"},{"key":"e_1_2_8_69_1","first-page":"1","article-title":"Spatially covariant image registration with text prompts","author":"Chen X","year":"2024","journal-title":"IEEE Trans Neural Netw Learn Syst"},{"key":"e_1_2_8_70_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2024.3462248"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.17827","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,8]],"date-time":"2025-12-08T10:30:22Z","timestamp":1765189822000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.17827"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,25]]},"references-count":69,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2025,7]]}},"alternative-id":["10.1002\/mp.17827"],"URL":"https:\/\/doi.org\/10.1002\/mp.17827","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"type":"print","value":"0094-2405"},{"type":"electronic","value":"2473-4209"}],"subject":[],"published":{"date-parts":[[2025,4,25]]},"assertion":[{"value":"2024-12-12","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-03-25","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-04-25","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e17827"}}