{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T11:57:56Z","timestamp":1780401476432,"version":"3.54.1"},"reference-count":48,"publisher":"Wiley","issue":"6","license":[{"start":{"date-parts":[[2025,3,6]],"date-time":"2025-03-06T00:00:00Z","timestamp":1741219200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2025,6]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>Major challenges in current semi\u2010supervised segmentation methods: (1) The complementary nature of information in pseudo\u2010label: a key limitation of consistent regularization methods is the tendency of sub\u2010networks to converge to the consensus case early on, leading to the degradation of co\u2010trained models into self\u2010trained models, whereas disagreement between sub\u2010networks is important for joint training. (2) Quantity\u2013quality weighting imbalance in pseudo\u2010label methods: threshold\u2010based pseudo\u2010label is to train the model with pseudo\u2010labels whose predicted confidence is higher than a hard threshold. In contrast, other pseudo\u2010labels are simply ignored. This study aims to propose a semi\u2010supervised model based on pseudo\u2010labeled weight balancing for medical image segmentation tasks for the above\u2010mentioned\u00a0problems.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We adopted a truncated Gaussian function weight balancing method based on the marginal hypothesis distribution to generate high\u2010quality pseudo\u2010labels while maintaining a high utilization rate of pseudo\u2010labels, and based on which we applied a uniform alignment strategy to solve the pseudo\u2010label imbalance problem due to the difference in the learning difficulty of different classes. In addition, to address the problem that self\u2010training algorithms rely too much on the quality of pseudo\u2010labels generated, we inherit the idea of knowledge refinement and integrate the mean teacher model of co\u2010training, thus proposing a novel semi\u2010supervised medical image segmentation framework, SCMT (Semi\u2010supervised Co\u2010training Mean Teacher), which is aimed at improving the existing self\u2010training algorithms or co\u2010training algorithms limitations of a single\u00a0model.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>We validate the effectiveness of the method by performing experimental evaluation on two commonly used benchmark medical datasets, LA, and Pancreas\u2010CT, by using 10%\/20% labeled data and 90%\/80% unlabeled data for training. On the LA dataset, the model obtained Hausdorff distance (HD) of 6.65\u00a0mm\/5.63\u00a0mm, average symmetric surface distance of 1.91\u00a0mm\/0.02\u00a0mm, Dice similarity coeffcient of 90.09%\/91.05%, and Jaccard of 81.08%\/83.64%. On the Pancreas\u2010CT dataset, the model obtained HD of 12.71\u00a0mm\/6.63\u00a0mm, average symmetric surface distance of 2.01\u00a0mm\/1.27\u00a0mm, Dice similarity coeffcient of 74.64%\/81.77% and Jaccard of 60.48%\/69.51%. The results show that our method not only outperforms existing semi\u2010supervised segmentation methods but also significantly improves segmentation performance and reduces the dependence on labeled data to achieve consistent and stable prediction results.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>\n                      We proposed a weight\u2010balanced co\u2010trained cross\u2010consistent semi\u2010supervised model SCMT for semi\u2010supervised segmentation of medical images, which consists of a CMT (Co\u2010training Mean Teacher) structure and quantity\u2010quality\u2010balanced pseudo\u2010label\u2010guided mutual consistency constraints. Compared with other models, we effectively exploit the challenging region and can more accurately capture the contours and finer details of the segmented objects without any shape or boundary constraints, resulting in highly accurate and detail\u2010rich segmentation results. In addition, we conduct comparative experiments with existing semi\u2010supervised models, and the experimental results show that our proposed model is capable of handling complex structures and segmenting details commonly missed by other methods. The segmentation results obtained are relatively stable and consistent and have certain advantages in improving the performance of surface segmentation. Code is available at:\n                      <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" xlink:href=\"https:\/\/github.com\/zhaojiashi\/SCMT\">https:\/\/github.com\/zhaojiashi\/SCMT<\/jats:ext-link>\n                      .\n                    <\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.17712","type":"journal-article","created":{"date-parts":[[2025,3,11]],"date-time":"2025-03-11T21:01:52Z","timestamp":1741726912000},"page":"3854-3876","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":4,"title":["Co\u2010training semi\u2010supervised medical image segmentation based on pseudo\u2010label weight balancing"],"prefix":"10.1002","volume":"52","author":[{"given":"Jiashi","family":"Zhao","sequence":"first","affiliation":[{"name":"School of Computer Science and Technology Changchun University of Science and Technology Changchun China"},{"name":"Zhongshan Institute of Changchun University of Science and Technology Zhongshan China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Li","family":"Yao","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology Changchun University of Science and Technology Changchun China"},{"name":"Zhongshan Institute of Changchun University of Science and Technology Zhongshan China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wang","family":"Cheng","sequence":"additional","affiliation":[{"name":"Graduate School of Informatics Nagoya University Nagoya Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Miao","family":"Yu","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology Changchun University of Science and Technology Changchun China"},{"name":"Zhongshan Institute of Changchun University of Science and Technology Zhongshan China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weili","family":"Shi","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology Changchun University of Science and Technology Changchun China"},{"name":"Zhongshan Institute of Changchun University of Science and Technology Zhongshan China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianhua","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Radiology The Second Hospital of Jilin University Changchun China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhengang","family":"Jiang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology Changchun University of Science and Technology Changchun China"},{"name":"Zhongshan Institute of Changchun University of Science and Technology Zhongshan China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2025,3,6]]},"reference":[{"key":"e_1_2_10_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2020.07.2329"},{"key":"e_1_2_10_3_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.15322"},{"key":"e_1_2_10_4_1","first-page":"1305","article-title":"A survey on medical image segmentation","volume":"6","author":"Kaur G","year":"2017","journal-title":"Int J Sci Res"},{"key":"e_1_2_10_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2020.103629"},{"key":"e_1_2_10_6_1","doi-asserted-by":"crossref","unstructured":"BaiW OktayO SinclairM et\u00a0al.Semi\u2010supervised learning for network\u2010based cardiac MR image segmentation. In:Medical Image Computing and Computer\u2010Assisted Intervention\u2010 MICCAI 2017: 20th International Conference September 11\u201013 2017 Proceedings Part II 20.Springer;2017:253\u2010260.","DOI":"10.1007\/978-3-319-66185-8_29"},{"key":"e_1_2_10_7_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2022.3220219"},{"key":"e_1_2_10_8_1","first-page":"18408","article-title":"Flexmatch: boosting semi\u2010supervised learning with curriculum pseudo labeling","volume":"34","author":"Zhang B","year":"2021","journal-title":"Adv Neural Inf Process Syst"},{"key":"e_1_2_10_9_1","unstructured":"LaineS AilaT.Temporal ensembling for semi\u2010supervised learning. arXiv preprint arXiv:1610.02242;2016."},{"key":"e_1_2_10_10_1","unstructured":"TangM ValipourS ZhangZ CobzasD JagersandM.A deep level set method for image segmentation. In:Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support: Third International Workshop DLMIA 2017 and 7th International Workshop ML\u2010CDS 2017 Held in Conjunction with MICCAI 2017 September 14 Proceedings 3.Springer;2017:126\u2010134."},{"key":"e_1_2_10_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2016.2621185"},{"key":"e_1_2_10_12_1","doi-asserted-by":"crossref","unstructured":"YangL QiL FengL ZhangW ShiY.Revisiting weak\u2010to\u2010strong consistency in semi\u2010supervised semantic segmentation. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition.IEEE;2023:7236\u20107246.","DOI":"10.1109\/CVPR52729.2023.00699"},{"key":"e_1_2_10_13_1","doi-asserted-by":"crossref","unstructured":"ShenZ CaoP YangH LiuX YangJ ZaianeOR.Co\u2010training with high\u2010confidence pseudo labels for semi\u2010supervised medical image segmentation. arXiv preprint arXiv:2301.04465;2023.","DOI":"10.24963\/ijcai.2023\/467"},{"key":"e_1_2_10_14_1","doi-asserted-by":"crossref","unstructured":"ZhuJ BolsterleeB ChowBV SongY MeijeringE.Hybrid dual mean\u2010teacher network with double\u2010uncertainty guidance for semi\u2010supervised segmentation of MRI scans. arXiv preprint arXiv:2303.05126;2023.","DOI":"10.1016\/j.compmedimag.2024.102383"},{"key":"e_1_2_10_15_1","doi-asserted-by":"crossref","unstructured":"ZhouY WangY TangP et\u00a0al.Semi\u2010supervised 3D abdominal multi\u2010organ segmentation via deep multi\u2010planar co\u2010training. In:2019 IEEE Winter Conference on Applications of Computer Vision (WACV).IEEE;2019:121\u2010140.","DOI":"10.1109\/WACV.2019.00020"},{"key":"e_1_2_10_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2020.101766"},{"key":"e_1_2_10_17_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2020.107269"},{"key":"e_1_2_10_18_1","doi-asserted-by":"crossref","unstructured":"KervadecH DolzJ Granger\u00c9 Ben AyedI.Curriculum semi\u2010supervised segmentation. In:Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2019: 22nd International Conference October 13\u201317 2019 Proceedings Part II 22.Springer;2019:568\u2010576.","DOI":"10.1007\/978-3-030-32245-8_63"},{"key":"e_1_2_10_19_1","doi-asserted-by":"crossref","unstructured":"OualiY HudelotC TamiM.Semi\u2010supervised semantic segmentation with cross\u2010consistency training. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition.IEEE;2020:12674\u201012684.","DOI":"10.1109\/CVPR42600.2020.01269"},{"key":"e_1_2_10_20_1","doi-asserted-by":"crossref","unstructured":"ChenX YuanY ZengG WangJ.Semi\u2010supervised semantic segmentation with cross pseudo supervision. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition.2021:2613\u20102622.","DOI":"10.1109\/CVPR46437.2021.00264"},{"key":"e_1_2_10_21_1","doi-asserted-by":"crossref","unstructured":"FrenchG AilaT LaineS MackiewiczM FinlaysonG.Semi\u2010supervised semantic segmentation needs strong high\u2010dimensional perturbations.2019.","DOI":"10.5244\/C.34.154"},{"key":"e_1_2_10_22_1","doi-asserted-by":"crossref","unstructured":"LiuY TianY ChenY LiuF BelagiannisV CarneiroG.Perturbed and strict mean teachers for semi\u2010supervised semantic segmentation. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition.IEEE;2022:4258\u20104267.","DOI":"10.1109\/CVPR52688.2022.00423"},{"key":"e_1_2_10_23_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_10_24_1","doi-asserted-by":"crossref","unstructured":"WangW ZhouT YuF DaiJ KonukogluE Van GoolL.Exploring cross\u2010image pixel contrast for semantic segmentation. In:Proceedings of the IEEE\/CVF International Conference on Computer Vision.IEEE;2021:7303\u20107313.","DOI":"10.1109\/ICCV48922.2021.00721"},{"key":"e_1_2_10_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2023.120289"},{"key":"e_1_2_10_26_1","doi-asserted-by":"crossref","unstructured":"ChenL\u2010C ZhuY PapandreouG SchroffF AdamH.Encoder\u2010decoder with Atrous separable convolution for semantic image segmentation. In:Proceedings of the European conference on computer vision (ECCV).Springer;2018:801\u2010818.","DOI":"10.1007\/978-3-030-01234-2_49"},{"key":"e_1_2_10_27_1","doi-asserted-by":"crossref","unstructured":"BaiY ChenD LiQ ShenW WangY.Bidirectional copy\u2010paste for semi\u2010supervised medical image segmentation. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition.IEEE;2023:11514\u201011524.","DOI":"10.1109\/CVPR52729.2023.01108"},{"key":"e_1_2_10_28_1","article-title":"RCPS: Rectified contrastive pseudo supervision for semi\u2010supervised medical image segmentation","author":"Zhao X","year":"2023","journal-title":"IEEE J Biomed Health Inform"},{"key":"e_1_2_10_29_1","doi-asserted-by":"crossref","unstructured":"KeZ QiuD LiK YanQ LauRW.Guided collaborative training for pixel\u2010wise semi\u2010supervised learning. In:Computer Vision\u2013ECCV 2020: 16th European Conference August 23\u201328 2020 Proceedings Part XIII 16.Springer;2020:429\u2010445.","DOI":"10.1007\/978-3-030-58601-0_26"},{"key":"e_1_2_10_30_1","unstructured":"HungW\u2010C TsaiY\u2010H LiouY\u2010T LinY\u2010Y YangM\u2010H.Adversarial learning for semi\u2010supervised semantic segmentation.2018. arXiv preprint arXiv:1802.07934."},{"key":"e_1_2_10_31_1","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_10_32_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 October 13\u201317 2019 Proceedings part II 22.Springer;2019:605\u2010613.","DOI":"10.1007\/978-3-030-32245-8_67"},{"key":"e_1_2_10_33_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.artmed.2022.102476"},{"key":"e_1_2_10_34_1","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2020.2995319"},{"key":"e_1_2_10_35_1","unstructured":"ZouY ZhangZ ZhangH et\u00a0al.Pseudoseg: Designing pseudo labels for semantic segmentation. arXiv preprint arXiv:2010.09713;2020."},{"key":"e_1_2_10_36_1","doi-asserted-by":"crossref","unstructured":"MilletariF NavabN AhmadiS\u2010A.V\u2010net: Fully convolutional neural networks for volumetric medical image segmentation. In:2016 Fourth International Conference on 3D Vision (3DV).IEEE;2016:565\u2010571.","DOI":"10.1109\/3DV.2016.79"},{"key":"e_1_2_10_37_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102530"},{"key":"e_1_2_10_38_1","doi-asserted-by":"crossref","unstructured":"ZhangY XiangT HospedalesTM LuH.Deep mutual learning. In:Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.IEEE;2018:4320\u20104328.","DOI":"10.1109\/CVPR.2018.00454"},{"key":"e_1_2_10_39_1","unstructured":"ParkG MoonW LeeS KimT\u2010Y HeoJ\u2010P.Mitigating Background Shift in Class\u2010Incremental Semantic Segmentation. arXiv preprint arXiv:2407.11859;2024."},{"key":"e_1_2_10_40_1","doi-asserted-by":"crossref","unstructured":"LiuX LiW YuanY.DiffRect: Latent Diffusion Label Rectification for Semi\u2010supervised Medical Image Segmentation. arXiv preprint arXiv:2407.09918;2024.","DOI":"10.1007\/978-3-031-72390-2_6"},{"key":"e_1_2_10_41_1","article-title":"Realistic evaluation of deep semi\u2010supervised learning algorithms","volume":"31","author":"Oliver A","year":"2018","journal-title":"Adv Neural Inf Process Syst"},{"key":"e_1_2_10_42_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2020.101832"},{"key":"e_1_2_10_43_1","doi-asserted-by":"crossref","unstructured":"LuoX ChenJ SongT WangG.Semi\u2010supervised medical image segmentation through dual\u2010task consistency. In:Proceedings of the AAAI conference on artificial intelligence. Vol.35.2021:8801\u20108809.","DOI":"10.1609\/aaai.v35i10.17066"},{"key":"e_1_2_10_44_1","doi-asserted-by":"crossref","unstructured":"LiS ZhangC HeX.Shape\u2010aware semi\u2010supervised 3D semantic segmentation for medical images. In:Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2020: 23rd International Conference October 4\u20138 2020 Proceedings Part I 23.Springer;2020:552\u2010561.","DOI":"10.1007\/978-3-030-59710-8_54"},{"key":"e_1_2_10_45_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.15923"},{"key":"e_1_2_10_46_1","unstructured":"WangY ChenH HengQ et\u00a0al.Freematch: Self\u2010adaptive thresholding for semi\u2010supervised learning. arXiv preprint arXiv:2205.07246;2022."},{"key":"e_1_2_10_47_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-024-44824-z"},{"key":"e_1_2_10_48_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2018.2837502"},{"key":"e_1_2_10_49_1","doi-asserted-by":"crossref","unstructured":"MiaoJ ChenC ZhangK ChuaiJ LiQ HengP\u2010A.Cross Prompting Consistency with Segment Anything Model for Semi\u2010supervised Medical Image Segmentation. arXiv preprint arXiv:2407.05416;2024.","DOI":"10.1007\/978-3-031-72120-5_16"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.17712","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,12]],"date-time":"2025-11-12T08:08:21Z","timestamp":1762934901000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.17712"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,6]]},"references-count":48,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2025,6]]}},"alternative-id":["10.1002\/mp.17712"],"URL":"https:\/\/doi.org\/10.1002\/mp.17712","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,3,6]]},"assertion":[{"value":"2024-07-16","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-01-18","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-03-06","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}