{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T20:32:07Z","timestamp":1786393927729,"version":"build-2736575974"},"reference-count":53,"publisher":"Wiley","issue":"7","license":[{"start":{"date-parts":[[2026,7,1]],"date-time":"2026-07-01T00:00:00Z","timestamp":1782864000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"},{"start":{"date-parts":[[2026,7,1]],"date-time":"2026-07-01T00:00:00Z","timestamp":1782864000000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["12074258"],"award-info":[{"award-number":["12074258"]}],"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":[[2026,7]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Scoliosis is a spinal disorder characterized by a three\u2010dimensional (3D) deformity of the vertebral column. 3D ultrasound imaging has been used for scoliosis assessment, and automatic vertebra detection is important for objective evaluation of spinal deformity. However, existing automatic assessment methods are mainly based on two\u2010dimensional (2D) projection images and require large amounts of labeled data, leading to incomplete use of volumetric ultrasound information and substantial annotation\u00a0effort.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>To develop a multi\u2010view constrained semi\u2010supervised learning (SSL) method for vertebra detection that reduces annotation requirements and facilitates fully automatic 3D spinal deformity\u00a0assessment.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>The proposed framework integrated maximum intensity projection (MIP) images in the coronal view with ultrasound slices in the transverse view to exploit complementary multi\u2010view information from 3D ultrasound volumes. A semi\u2010supervised training strategy was adopted, in which coronal bounding boxes were incorporated into pseudo\u2010label generation for the transverse lamina detector. The resulting 3D lamina positions were used for spinal curve fitting, curvature estimation, and 3D visualization. The training set comprised 83 593 unlabeled transverse ultrasound slices from 180 subjects and 856 labeled slices from 10 subjects, whereas the testing set included 2163 slices from 163 patients. To evaluate label efficiency, not all labeled slices were used for training; instead, subsets of different sizes were sampled from the labeled training set to analyze the effect of labeled training set size on detection performance. The proposed method was compared with supervised learning and several SSL benchmarks, including Pseudo\u2010Labeling, FixMatch, MeanTeacher, Uncertainty\u2010Aware Mean Teacher (UA\u2010MT), and Vertmatch. Detection performance was evaluated using accuracy, precision, recall, and F1 score, whereas scoliosis angle estimation in 30 ultrasound subjects was assessed against the radiographic Cobb angle and manually labeled ultrasound spine curvature using mean absolute difference (MAD) and standard deviation (STD). Statistical analysis was performed using the Wilcoxon signed\u2010rank test at a significance level of , with Bonferroni correction for multiplicity, Cohen's d effect\u2010size analysis, and 95% confidence intervals estimated from 1000 bootstrap\u00a0resamples.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      On the vertebra detection task, the proposed method achieved an accuracy of 91.19%, a precision of 93.02%, a recall of 97.89%, and an F1 score of 95.39% using a subset of 400 labeled training slices. Compared to all SSL methods, the proposed method improved accuracy by 0.22% to 10.34% and precision by 1.71% \u2013 9.03%. These comparisons were statistically significant (), and effect\u2010size analysis showed large effects for most metrics, except for Recall versus FixMatch (Cohen's\n                      <jats:italic>d<\/jats:italic>\n                      = 0.26). Ablation experiments showed that the multi\u2010view constraint increased accuracy from 83.87% to 91.19%. On the clinical dataset, both manual and automatic ultrasound\u2010based measurements showed good agreement with the radiographic measurements (automatic MAD: , SD: ; manual MAD: , SD: ). The difference was not statistically significant ().\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>The proposed method reduces annotation workload and improves vertebra detection performance, providing a promising non\u2010radiative approach for scoliosis\u00a0assessment.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.70557","type":"journal-article","created":{"date-parts":[[2026,7,1]],"date-time":"2026-07-01T17:31:46Z","timestamp":1782927106000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Multi\u2010view constrained semi\u2010supervised vertebra detection for 3D ultrasound spine volume"],"prefix":"10.1002","volume":"53","author":[{"given":"Yuchong","family":"Gao","sequence":"first","affiliation":[{"name":"School of Information Science and Technology ShanghaiTech University Shanghai China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianhao","family":"Zhao","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology ShanghaiTech University Shanghai China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yi","family":"Mao","sequence":"additional","affiliation":[{"name":"The First Medical Center of Chinese PLA General Hospital Beijing China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaowei","family":"Song","sequence":"additional","affiliation":[{"name":"Maoyu(Qingdao) Medical Technology Co., Ltd. Beijing Branch Beijing China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mingbo","family":"Zhang","sequence":"additional","affiliation":[{"name":"The First Medical Center of Chinese PLA General Hospital Beijing China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rui","family":"Zheng","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology ShanghaiTech University Shanghai China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongye","family":"Zeng","sequence":"additional","affiliation":[{"name":"University of Washington Seattle Washington USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,7,1]]},"reference":[{"key":"e_1_2_9_2_1","doi-asserted-by":"publisher","DOI":"10.1097\/BRS.0000000000001197"},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.2106\/00004623-197557070-00015"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1097\/00007632-198810000-00029"},{"key":"e_1_2_9_5_1","first-page":"261","article-title":"Outline for the study of scoliosis","volume":"5","author":"COBB JR","year":"1948","journal-title":"Instr Course Lect"},{"key":"e_1_2_9_6_1","doi-asserted-by":"publisher","DOI":"10.1007\/s002560050445"},{"key":"e_1_2_9_7_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejrad.2015.12.032"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1097\/00007632-200008150-00009"},{"key":"e_1_2_9_9_1","doi-asserted-by":"publisher","DOI":"10.1158\/1055-9965.EPI-07-2628"},{"key":"e_1_2_9_10_1","doi-asserted-by":"publisher","DOI":"10.1117\/12.2292729"},{"key":"e_1_2_9_11_1","doi-asserted-by":"crossref","unstructured":"ZengH ZhengR LeLH TaD.Measuring spinous process angle on ultrasound spine images using the GVF segmentation method. In:2019 IEEE International Ultrasonics Symposium (IUS) pages.2019:1477\u20131480.","DOI":"10.1109\/ULTSYM.2019.8925710"},{"key":"e_1_2_9_12_1","first-page":"1942","article-title":"Zeng et\u00a0al. Training\u2010free image style alignment for domain shift on handheld ultrasound devices","volume":"44","year":"2024","journal-title":"IEEE Trans Med Imaging"},{"key":"e_1_2_9_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jspd.2014.08.008"},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1109\/TUFFC.2021.3087712"},{"key":"e_1_2_9_15_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00586-015-3855-8"},{"key":"e_1_2_9_16_1","unstructured":"ZengH\u2010Y GeS\u2010H GaoY\u2010C et\u00a0al.Automatic segmentation of vertebral features on ultrasound spine images using Stacked Hourglass Network.arXiv preprint arXiv:2105.038472021."},{"key":"e_1_2_9_17_1","article-title":"Anatomical prior and inter\u2010slice consistency for semi\u2010supervised vertebral structure detection in 3D ultrasound volume","author":"Zeng H","year":"2024","journal-title":"IEEE J Biomed Health Inform"},{"key":"e_1_2_9_18_1","unstructured":"ZhouY LeeTT\u2010Y LaiKK\u2010L et\u00a0al.Automatic ultrasound curve angle measurementviaaffinity clustering for adolescent idiopathic scoliosis evaluation.arXiv preprint arXiv:2405.03141(2024)."},{"key":"e_1_2_9_19_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ultrasmedbio.2023.12.015"},{"key":"e_1_2_9_20_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10439-022-02925-0"},{"key":"e_1_2_9_21_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.16440"},{"key":"e_1_2_9_22_1","doi-asserted-by":"publisher","DOI":"10.1155\/2019\/6357171"},{"key":"e_1_2_9_23_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.medengphy.2022.103848"},{"key":"e_1_2_9_24_1","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics13162658"},{"key":"e_1_2_9_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.102258"},{"key":"e_1_2_9_26_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-022-09354-6"},{"key":"e_1_2_9_27_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00138-022-01360-4"},{"key":"e_1_2_9_28_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.15961"},{"key":"e_1_2_9_29_1","first-page":"91","article-title":"Faster r\u2010cnn: Towards real\u2010time object detection with region proposal networks","volume":"28","author":"Ren S","year":"2015","journal-title":"Adv Neural Inf Process Syst"},{"key":"e_1_2_9_30_1","first-page":"1","volume-title":"2021 IEEE International Ultrasonics Symposium (IUS)","author":"Liu Z","year":"2021"},{"key":"e_1_2_9_31_1","doi-asserted-by":"crossref","unstructured":"CarionN MassaF SynnaeveG UsunierN KirillovA ZagoruykoS.End\u2010to\u2010end object detection with transformers. In:Computer Vision\u2013ECCV 2020: 16th European Conference Glasgow UK August 23\u201328 2020 Proceedings Part I 16.Springer;2020:213\u2013229.","DOI":"10.1007\/978-3-030-58452-8_13"},{"key":"e_1_2_9_32_1","doi-asserted-by":"publisher","DOI":"10.1109\/IUS54386.2022.9958621"},{"key":"e_1_2_9_33_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ultras.2023.107012"},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2025.126410"},{"key":"e_1_2_9_35_1","doi-asserted-by":"crossref","unstructured":"XieQ LuongM\u2010T HovyE LeQV.Self\u2010training with noisy student improves imagenet classification. In:Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition.2020:10687\u201310698.","DOI":"10.1109\/CVPR42600.2020.01070"},{"key":"e_1_2_9_36_1","first-page":"596","article-title":"Fixmatch: Simplifying semi\u2010supervised learning with consistency and confidence","volume":"33","author":"Sohn K","year":"2020","journal-title":"Adv Neural Inf Process Syst"},{"key":"e_1_2_9_37_1","unstructured":"LeeD\u2010H.Pseudo\u2010label: The simple and efficient semi\u2010supervised learning method for deep neural networks. In:Workshop on challenges in representation learning ICML. Atlanta GA;2013;1\u20136."},{"key":"e_1_2_9_38_1","first-page":"1195","article-title":"Mean teachers are better role models: Weight\u2010averaged consistency targets improve semi\u2010supervised deep learning results","volume":"30","author":"Tarvainen A","year":"2017","journal-title":"Adv Neural Inf Process Syst"},{"key":"e_1_2_9_39_1","doi-asserted-by":"crossref","unstructured":"YuL WangS LiX FuC\u2010W HengP\u2010A.Uncertainty\u2010aware self\u2010ensembling model for semi\u2010supervised 3D left atrium segmentation. In:Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2019: 22nd International Conference Shenzhen China October 13\u201317 2019 proceedings part II 22.Springer;2019:605\u2013613.","DOI":"10.1007\/978-3-030-32245-8_67"},{"key":"e_1_2_9_40_1","doi-asserted-by":"crossref","unstructured":"WangD ZhangY ZhangK WangL.FocalMix: Semi\u2010supervised learning for 3D medical image detection. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition.2020:3951\u20133960.","DOI":"10.1109\/CVPR42600.2020.00401"},{"key":"e_1_2_9_41_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102460"},{"key":"e_1_2_9_42_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.102315"},{"key":"e_1_2_9_43_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.102117"},{"key":"e_1_2_9_44_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2020.2995518"},{"key":"e_1_2_9_45_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102508"},{"key":"e_1_2_9_46_1","doi-asserted-by":"crossref","unstructured":"NewellA YangK DengJ.Stacked hourglass networks for human pose estimation. In:Computer Vision\u2013ECCV 2016: 14th European Conference Amsterdam The Netherlands October 11\u201014 2016 Proceedings Part VIII 14.Springer;2016:483\u2013499.","DOI":"10.1007\/978-3-319-46484-8_29"},{"key":"e_1_2_9_47_1","doi-asserted-by":"crossref","unstructured":"CubukED ZophB ShlensJ LeQV.Randaugment: Practical automated data augmentation with a reduced search space. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition Workshops.2020:702\u2013703.","DOI":"10.1109\/CVPRW50498.2020.00359"},{"key":"e_1_2_9_48_1","unstructured":"EsterM KriegelH\u2010P SanderJ XuX.A density\u2010based algorithm for discovering clusters in large spatial databases with noise. In:KDD 96 1996:226\u2013231."},{"key":"e_1_2_9_49_1","doi-asserted-by":"crossref","unstructured":"GeS ZengH ZhengR.Automatic measurement of spinous process angles on ultrasound spine images. In:2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC).2020:2101\u20132104.","DOI":"10.1109\/EMBC44109.2020.9176211"},{"key":"e_1_2_9_50_1","doi-asserted-by":"crossref","unstructured":"WoolsonRF.Wilcoxon signed\u2010rank test.Wiley encyclopedia of clinical trials.2007:1\u20133.","DOI":"10.1002\/9780471462422.eoct979"},{"key":"e_1_2_9_51_1","unstructured":"WeissteinEW.Bonferroni correction.https:\/\/mathworld.wolfram.com\/2004."},{"key":"e_1_2_9_52_1","doi-asserted-by":"crossref","unstructured":"DienerMJ.Cohen's d.The Corsini encyclopedia of psychology.2010:1\u20131.","DOI":"10.1002\/9780470479216.corpsy0569"},{"key":"e_1_2_9_53_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jspd.2014.02.006"},{"key":"e_1_2_9_54_1","first-page":"1","volume-title":"2022 IEEE International Ultrasonics Symposium (IUS)","author":"Gao Y","year":"2022"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70557","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.70557","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70557","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T19:55:16Z","timestamp":1786391716000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.70557"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7]]},"references-count":53,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2026,7]]}},"alternative-id":["10.1002\/mp.70557"],"URL":"https:\/\/doi.org\/10.1002\/mp.70557","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7]]},"assertion":[{"value":"2025-08-19","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-06-12","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-07-01","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70557"}}