{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T15:28:54Z","timestamp":1788449334243,"version":"build-2803163510"},"reference-count":51,"publisher":"Wiley","issue":"8","license":[{"start":{"date-parts":[[2025,7,25]],"date-time":"2025-07-25T00:00:00Z","timestamp":1753401600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62406211"],"award-info":[{"award-number":["62406211"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100018542","name":"Natural Science Foundation of Sichuan Province","doi-asserted-by":"publisher","award":["2024NSFSC0654"],"award-info":[{"award-number":["2024NSFSC0654"]}],"id":[{"id":"10.13039\/501100018542","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100018542","name":"Natural Science Foundation of Sichuan Province","doi-asserted-by":"publisher","award":["2023NSFSC0636"],"award-info":[{"award-number":["2023NSFSC0636"]}],"id":[{"id":"10.13039\/501100018542","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["J Applied Clin Med Phys"],"published-print":{"date-parts":[[2025,8]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Background<\/jats:title><jats:p>Thyroid cancer is one of the most common cancers in clinical practice, and accurate classification of thyroid nodule ultrasound images is crucial for computer\u2010aided diagnosis. Models based on a convolutional neural network (CNN) or a transformer struggle to integrate local and global features, which impacts the recognition accuracy.<\/jats:p><\/jats:sec><jats:sec><jats:title>Purpose<\/jats:title><jats:p>Our method is designed to capture both the key local fine\u2010grained features and the global spatial features essential for thyroid nodule diagnosis simultaneously. It adapts to the irregular morphology of thyroid nodules, dynamically focuses on the key pixel\u2010level regions of thyroid nodules, and thereby improves the model's recognition accuracy and generalization ability.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>The proposed multi\u2010scale fusion model, the local and global feature fusion network (LGF\u2010Net), inspired by the dual\u2010path mechanism of human visual diagnosis, consists of two branches: a CNN branch and a Transformer branch. The CNN branch integrates the wavelet transform and deformable convolution module (WTDCM) to enhance the model's ability to capture discriminative local features and recognize fine\u2010grained textures. By introducing the aggregated attention (AA) mechanism, which mimics biological vision, into the Transformer branch, spatial features are effectively captured. The adaptive feature fusion module (FFM) is then utilized to integrate the multi\u2010scale features of thyroid nodules, further improving classification performance.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>We evaluated our model on the public thyroid nodule classification dataset (TNCD) and a private clinical dataset using accuracy, recall, precision, and F1\u2010score. On TNCD, the model achieved 81.50%, 79.51%, 79.92%, and 79.70%, respectively. On the private dataset, it reached 91.24%, 88.90%, 90.73%, and 89.73%, respectively. These results outperformed state\u2010of\u2010the\u2010art methods. We also conducted ablation studies and visualization analysis to validate the model's components and interpretability.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>The experiments demonstrate that our method improves the accuracy of thyroid nodule recognition, shows its strong generalization ability and potential for clinical application, and provides interpretability for clinicians' diagnoses.<\/jats:p><\/jats:sec>","DOI":"10.1002\/acm2.70149","type":"journal-article","created":{"date-parts":[[2025,7,26]],"date-time":"2025-07-26T03:45:14Z","timestamp":1753501514000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["LGF\u2010Net: A multi\u2010scale feature fusion network for thyroid nodule ultrasound image classification"],"prefix":"10.1002","volume":"26","author":[{"given":"Yao","family":"Xiao","sequence":"first","affiliation":[{"name":"College of Biomedical Engineering Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yan","family":"Zhuang","sequence":"additional","affiliation":[{"name":"College of Biomedical Engineering Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenwu","family":"Ling","sequence":"additional","affiliation":[{"name":"Department of Ultrasound West China Hospital Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shouyu","family":"Jiang","sequence":"additional","affiliation":[{"name":"College of Biomedical Engineering Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ke","family":"Chen","sequence":"additional","affiliation":[{"name":"College of Biomedical Engineering Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guoliang","family":"Liao","sequence":"additional","affiliation":[{"name":"College of Biomedical Engineering Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuhua","family":"Xie","sequence":"additional","affiliation":[{"name":"College of Biomedical Engineering Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yao","family":"Hou","sequence":"additional","affiliation":[{"name":"College of Biomedical Engineering Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lin","family":"Han","sequence":"additional","affiliation":[{"name":"College of Biomedical Engineering Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhan","family":"Hua","sequence":"additional","affiliation":[{"name":"Department of General Surgery China\u2010Japan Friendship Hospital Beijing China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yan","family":"Luo","sequence":"additional","affiliation":[{"name":"Department of Ultrasound West China Hospital Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2030-9681","authenticated-orcid":false,"given":"Jiangli","family":"Lin","sequence":"additional","affiliation":[{"name":"College of Biomedical Engineering Sichuan University  Chengdu China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2025,7,25]]},"reference":[{"key":"e_1_2_13_2_1","doi-asserted-by":"publisher","DOI":"10.3322\/caac.21820"},{"key":"e_1_2_13_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jncc.2024.01.006"},{"key":"e_1_2_13_4_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jacr.2017.01.046"},{"key":"e_1_2_13_5_1","doi-asserted-by":"publisher","DOI":"10.1038\/nature21056"},{"key":"e_1_2_13_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2017.07.005"},{"key":"e_1_2_13_7_1","doi-asserted-by":"publisher","DOI":"10.1002\/ima.22363"},{"key":"e_1_2_13_8_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2024.108087"},{"key":"e_1_2_13_9_1","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2022.878718"},{"key":"e_1_2_13_10_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ada5a6"},{"key":"e_1_2_13_11_1","unstructured":"VaswaniA ShazeerN ParmarN et\u00a0al.Attention is all you need.Proceedings of the 31st International Conference on Neural Information Processing Systems.arXiv;2017."},{"key":"e_1_2_13_12_1","volume-title":"An Image Is Worth 16\u00d716 Words: Transformers For Image Recognition At Scale","author":"Dosovitskiy A","year":"2020"},{"key":"e_1_2_13_13_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-16852-9_2"},{"key":"e_1_2_13_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2022.106444"},{"key":"e_1_2_13_15_1","first-page":"1","article-title":"Dense swin transformer for classification of thyroid nodules","volume":"2023","author":"Baima N","year":"2023","journal-title":"Annu Int Conf IEEE Eng Med Biol Soc"},{"key":"e_1_2_13_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2024.103255"},{"key":"e_1_2_13_17_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00371-023-02984-x"},{"key":"e_1_2_13_18_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2023.107921"},{"key":"e_1_2_13_19_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2020.2980290"},{"key":"e_1_2_13_20_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2020.101665"},{"key":"e_1_2_13_21_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2022.106172"},{"key":"e_1_2_13_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2022.3153559"},{"key":"e_1_2_13_23_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2023.102905"},{"key":"e_1_2_13_24_1","doi-asserted-by":"crossref","unstructured":"FukuiH HirakawaT YamashitaT FujiyoshiH.Attention branch network: learning of attention mechanism for visual explanation.2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 15\u201320 June 2019.IEEE;2019.","DOI":"10.1109\/CVPR.2019.01096"},{"key":"e_1_2_13_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.artmed.2023.102699"},{"key":"e_1_2_13_26_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2022.3140797"},{"key":"e_1_2_13_27_1","doi-asserted-by":"publisher","DOI":"10.1145\/3505244"},{"key":"e_1_2_13_28_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2022.3152247"},{"key":"e_1_2_13_29_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2023.120969"},{"key":"e_1_2_13_30_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2023.106791"},{"key":"e_1_2_13_31_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2023.105534"},{"key":"e_1_2_13_32_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.medengphy.2024.104101"},{"key":"e_1_2_13_33_1","doi-asserted-by":"publisher","DOI":"10.1186\/s12859-023-05446-2"},{"key":"e_1_2_13_34_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2023.107999"},{"key":"e_1_2_13_35_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2024.128749"},{"key":"e_1_2_13_36_1","doi-asserted-by":"crossref","unstructured":"DaiJ QiH XiongY et\u00a0al.Deformable convolutional networks.2017 IEEE International Conference on Computer Vision (ICCV) 22\u201329 Oct. 2017.arXiv;2017.","DOI":"10.1109\/ICCV.2017.89"},{"key":"e_1_2_13_37_1","doi-asserted-by":"crossref","unstructured":"ShiD.TransNeXt: robust foveal visual perception for vision transformers.2024 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 16\u201322 June 2024.IEEE;2024.","DOI":"10.1109\/CVPR52733.2024.01683"},{"key":"e_1_2_13_38_1","doi-asserted-by":"crossref","unstructured":"HuJ ShenL SunG Squeeze\u2010and\u2010excitation networks.2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition 18\u201323 June 2018.arXiv;2018.","DOI":"10.1109\/CVPR.2018.00745"},{"key":"e_1_2_13_39_1","unstructured":"JaderbergM SimonyanK ZissermanA KavukcuogluK.Spatial transformer networks.Proceedings of the 29th International Conference on Neural Information Processing Systems.arXiv;2015."},{"key":"e_1_2_13_40_1","doi-asserted-by":"crossref","unstructured":"GongH ChengH XieY et\u00a0al.Less is more: adaptive curriculum learning for thyroid nodule diagnosis.Medical Image Computing and Computer Assisted Intervention \u2013 MICCAI 2022 2022.arXiv;2022.","DOI":"10.1007\/978-3-031-16440-8_24"},{"key":"e_1_2_13_41_1","unstructured":"KingmaD BaJ.Adam: a method for stochastic optimization.2015 Internationtional Conference on Learning Representations. arXiv;2015."},{"key":"e_1_2_13_42_1","doi-asserted-by":"crossref","unstructured":"HeK ZhangX RenS SunJ.Deep residual learning for image recognition.2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 27\u201330 June 2016.arXiv;2016.","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_2_13_43_1","doi-asserted-by":"crossref","unstructured":"HowardA SandlerM ChenB et\u00a0al.Searching for MobileNetV3.2019 IEEE\/CVF International Conference on Computer Vision (ICCV) 27 Oct\u20102 Nov 2019.arXiv;2019.","DOI":"10.1109\/ICCV.2019.00140"},{"key":"e_1_2_13_44_1","doi-asserted-by":"crossref","unstructured":"LiuZ MaoH WuCY FeichtenhoferC DarrellT XieS.A ConvNet for the 2020s.2022 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 18\u201324 June 2022.arXiv;2022.","DOI":"10.1109\/CVPR52688.2022.01167"},{"key":"e_1_2_13_45_1","doi-asserted-by":"crossref","unstructured":"LiuZ HuH LinY et\u00a0al.Swin transformer v2: scaling up capacity and resolution.2022 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR); 18\u201324 June 2022.arXiv;2022.","DOI":"10.1109\/CVPR52688.2022.01170"},{"key":"e_1_2_13_46_1","volume-title":"MobileViT: Light\u2010weight, General\u2010Purpose, and Mobile\u2010friendly Vision Transformer","author":"Mehta S","year":"2021"},{"key":"e_1_2_13_47_1","doi-asserted-by":"crossref","unstructured":"ChenY DaiX ChenD et\u00a0al.Mobile\u2010former: bridging mobilenet and transformer.2022 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 18\u201324 June 2022.aXiv;2022.","DOI":"10.1109\/CVPR52688.2022.00520"},{"key":"e_1_2_13_48_1","volume-title":"Uniformer: Unified Transformer for Efficient Spatiotemporal Representation Learning","author":"Li K","year":"2022"},{"key":"e_1_2_13_49_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2023.3243048"},{"key":"e_1_2_13_50_1","doi-asserted-by":"crossref","unstructured":"CorcinoA Calder\u00f3nM.Better together than separate: prediagnosis of thyroid nodules through ultrasound imaging using CNN VIT and hybrid models.2024 43rd International Conference of the Chilean Computer Science Society (SCCC) 28\u201330 Oct. 2024.IEEE;2024.","DOI":"10.1109\/SCCC63879.2024.10767636"},{"key":"e_1_2_13_51_1","doi-asserted-by":"publisher","DOI":"10.1088\/2057-1976\/ad9f68"},{"key":"e_1_2_13_52_1","doi-asserted-by":"crossref","unstructured":"SelvarajuRR CogswellM DasA VedantamR ParikhD BatraD.Grad\u2010CAM: visual explanations from deep networks via gradient\u2010based localization.2017 IEEE International Conference on Computer Vision (ICCV) 22\u201329 Oct. 2017.arXiv;2017.","DOI":"10.1109\/ICCV.2017.74"}],"container-title":["Journal of Applied Clinical Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/acm2.70149","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,16]],"date-time":"2025-08-16T03:56:01Z","timestamp":1755316561000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/acm2.70149"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,25]]},"references-count":51,"journal-issue":{"issue":"8","published-print":{"date-parts":[[2025,8]]}},"alternative-id":["10.1002\/acm2.70149"],"URL":"https:\/\/doi.org\/10.1002\/acm2.70149","archive":["Portico"],"relation":{},"ISSN":["1526-9914","1526-9914"],"issn-type":[{"value":"1526-9914","type":"print"},{"value":"1526-9914","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,25]]},"assertion":[{"value":"2025-03-06","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-05-07","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-07-25","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70149"}}