{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,26]],"date-time":"2026-10-26T16:42:41Z","timestamp":1793032961737,"version":"4.5.0"},"reference-count":41,"publisher":"Wiley","issue":"5","license":[{"start":{"date-parts":[[2026,5,23]],"date-time":"2026-05-23T00:00:00Z","timestamp":1779494400000},"content-version":"vor","delay-in-days":22,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/501100003725","name":"National Research Foundation of Korea","doi-asserted-by":"publisher","award":["2021M3I1A1097909"],"award-info":[{"award-number":["2021M3I1A1097909"]}],"id":[{"id":"10.13039\/501100003725","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003725","name":"National Research Foundation of Korea","doi-asserted-by":"publisher","award":["RS\u20102023\u201000285089"],"award-info":[{"award-number":["RS\u20102023\u201000285089"]}],"id":[{"id":"10.13039\/501100003725","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003725","name":"National Research Foundation of Korea","doi-asserted-by":"publisher","award":["2710003307"],"award-info":[{"award-number":["2710003307"]}],"id":[{"id":"10.13039\/501100003725","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":[[2026,5]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Machine learning models for chest X\u2010ray (CXR)\u2010based tuberculosis (TB) detection often suffer from performance drops when deployed across institutions due to domain shifts caused by differences in patient populations, imaging protocols, and equipment. Domain adaptation has been widely studied to address the problem, but privacy regulations restrict data sharing, making conventional approaches\u00a0infeasible.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>We aim to enhance TB detection performance when a detector model is shared across multiple clinical sites without accessing any source data or statistics. This problem is related to source\u2010free domain adaptation (SFDA) but requires special designs for imbalanced binary\u00a0classification.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We propose two methods for TB detection, which are based on Source HypOthesis Transfer (SHOT) and SFDA via source Distribution Estimation (SFDA\u2010DE), respectively. As the main ideas of our methods, we propose ranking\u2010based pseudo\u2010labeling and selective fine\u2010tuning of network layers to address severe class imbalance and prevent\u00a0overfitting.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>On six unseen target domains with high class imbalance, our proposed SFDA methods significantly improved the average F1 score and AUROC compared to non\u2010adapted models, all without accessing source data during\u00a0adaptation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Our methods offer a privacy\u2010preserving and effective strategy to improve the generalization of TB detection models across institutions, addressing domain shifts while complying with data\u2010sharing\u00a0restrictions.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/acm2.70626","type":"journal-article","created":{"date-parts":[[2026,5,23]],"date-time":"2026-05-23T18:09:47Z","timestamp":1779559787000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Source\u2010free domain adaptation for multi\u2010institutional chest X\u2010ray images"],"prefix":"10.1002","volume":"27","author":[{"given":"Hyoyi","family":"Kim","sequence":"first","affiliation":[{"name":"Department of Nuclear and Quantum Engineering KAIST  Yuseong\u2010gu, Daejeon South Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Seoyoung","family":"Lee","sequence":"additional","affiliation":[{"name":"Department of Radiology Mayo Clinic  Rochester Minnesota USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9409-3628","authenticated-orcid":false,"given":"Seungryong","family":"Cho","sequence":"additional","affiliation":[{"name":"Department of Nuclear and Quantum Engineering KAIST  Yuseong\u2010gu, Daejeon South Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,5,23]]},"reference":[{"key":"e_1_2_11_2_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2017162326"},{"key":"e_1_2_11_3_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10278\u2010015\u20109821\u20101"},{"key":"e_1_2_11_4_1","doi-asserted-by":"crossref","unstructured":"PasaF GolkovV PfeifferF CremersD.Efficient deep network architectures for fast chest X\u2010ray tuberculosis screening. In:International Conference on Image Analysis and Recognition (ICIAR).Springer;2019:211\u2010219.","DOI":"10.1038\/s41598-019-42557-4"},{"key":"e_1_2_11_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2019.07.013"},{"key":"e_1_2_11_6_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2020.2999874"},{"key":"e_1_2_11_7_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.102088"},{"key":"e_1_2_11_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2021.3117407"},{"key":"e_1_2_11_9_1","doi-asserted-by":"publisher","DOI":"10.1016\/J.COMPBIOMED.2023.107912"},{"key":"e_1_2_11_10_1","unstructured":"GaninY LempitskyV.Domain\u2010adversarial training of neural networks. In:International Conference on Machine Learning (ICML).2015:1180\u20101189."},{"key":"e_1_2_11_11_1","unstructured":"LongM CaoY WangJ JordanMI.Learning transferable features with deep adaptation networks. In:International Conference on Machine Learning (ICML).2015:97\u2010105."},{"key":"e_1_2_11_12_1","doi-asserted-by":"crossref","unstructured":"SunB SaenkoK.Return of frustratingly easy domain adaptation: CORAL.European Conference on Computer Vision (ECCV) Workshops.2016:426\u2010443.","DOI":"10.1007\/978-3-319-49409-8_35"},{"key":"e_1_2_11_13_1","doi-asserted-by":"crossref","unstructured":"TzengE HoffmanJ SaenkoK DarrellT.Adversarial discriminative domain adaptation. In:IEEE Conference on Computer Vision and Pattern Recognition (CVPR).2017:2962\u20102971.","DOI":"10.1109\/CVPR.2017.316"},{"key":"e_1_2_11_14_1","unstructured":"HoffmanJ TzengE ParkT et\u00a0al.Cycada: cycle\u2010consistent adversarial domain adaptation. In:International Conference on Machine Learning (ICML).2018:1989\u20101998."},{"key":"e_1_2_11_15_1","doi-asserted-by":"crossref","unstructured":"PengX BaiQ XiaX HuangZ SaenkoK WangB.Moment matching for multi\u2010source domain adaptation. In:IEEE International Conference on Computer Vision (ICCV).2019:1406\u20101415.","DOI":"10.1109\/ICCV.2019.00149"},{"key":"e_1_2_11_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2021.106775"},{"key":"e_1_2_11_17_1","doi-asserted-by":"publisher","DOI":"10.1561\/2200000083"},{"key":"e_1_2_11_18_1","unstructured":"LiangJ HuD FengJ.Do we really need to access the source data? source hypothesis transfer for unsupervised domain adaptation. In:International Conference on Machine Learning PMLR;2020:6028\u20106039."},{"key":"e_1_2_11_19_1","doi-asserted-by":"crossref","unstructured":"DingN XuY TangY XuC WangY TaoD.Source\u2010free domain adaptation via distribution estimation. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition.2022:7212\u20107222.","DOI":"10.1109\/CVPR52688.2022.00707"},{"key":"e_1_2_11_20_1","doi-asserted-by":"crossref","unstructured":"KarimN MithunNC RajvanshiA ChiuHp SamarasekeraS RahnavardN.C\u2010sfda: A curriculum learning aided self\u2010training framework for efficient source free domain adaptation. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition.2023:24120\u201024131.","DOI":"10.1109\/CVPR52729.2023.02310"},{"key":"e_1_2_11_21_1","first-page":"475","article-title":"Two public chest X\u2010ray datasets for computer\u2010aided screening of pulmonary diseases","volume":"46","author":"Jaeger S","year":"2014","journal-title":"Quant Imaging Med Surg"},{"key":"e_1_2_11_22_1","doi-asserted-by":"publisher","DOI":"10.2214\/ajr.174.1.1740071"},{"key":"e_1_2_11_23_1","doi-asserted-by":"publisher","DOI":"10.1186\/s12938-018-0609-y"},{"key":"e_1_2_11_24_1","doi-asserted-by":"publisher","DOI":"10.1117\/1.JMI.10.6.061103"},{"key":"e_1_2_11_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/J.NEUCOM.2018.05.083"},{"key":"e_1_2_11_26_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2005.02.002"},{"key":"e_1_2_11_27_1","doi-asserted-by":"crossref","unstructured":"HeK ZhangX RenS SunJ.Deep residual learning for image recognition. In:Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:770\u2010778.","DOI":"10.1109\/CVPR.2016.90"},{"issue":"1","key":"e_1_2_11_28_1","first-page":"1929","article-title":"Dropout: a simple way to prevent neural networks from overfitting","volume":"15","author":"Srivastava N","year":"2014","journal-title":"J Mach Learn Res"},{"key":"e_1_2_11_29_1","doi-asserted-by":"crossref","unstructured":"DengJ DongW SocherR LiLJ LiK Fei\u2010FeiL.Imagenet: a large\u2010scale hierarchical image database. In:2009 IEEE Conference on Computer Vision and Pattern Recognition IEEE;2009:248\u2010255.","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"e_1_2_11_30_1","unstructured":"PaszkeA GrossS MassaF et\u00a0al.Pytorch: an imperative style high\u2010performance deep learning library. In:Proceedings of the 33rd International Conference on Neural Information Processing Systems.2019;8026\u20108037."},{"key":"e_1_2_11_31_1","doi-asserted-by":"crossref","unstructured":"BurtPJ AdelsonEH.The Laplacian pyramid as a compact image code. In:Readings in Computer Vision. Elsevier;1987:671\u2010679.","DOI":"10.1016\/B978-0-08-051581-6.50065-9"},{"key":"e_1_2_11_32_1","unstructured":"KunduJN VenkatN BabuRV.Universal Source\u2010free domain adaptation. In:IEEE Conference on Computer Vision and Pattern Recognition (CVPR).2020:4544\u20104553."},{"key":"e_1_2_11_33_1","unstructured":"YangY ChenS ZhangH WangQ.A2Net: adapt\u2010and\u2010align networks for source\u2010free domain adaptation. In:International Conference on Computer Vision (ICCV).2021:9154\u20109163."},{"key":"e_1_2_11_34_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2021.108036"},{"key":"e_1_2_11_35_1","unstructured":"ChenJ LuoJ ZhangW FengJ.Model adaptation with only batch\u2010norm statistics. In:European Conference on Computer Vision (ECCV).2022:489\u2010506."},{"key":"e_1_2_11_36_1","unstructured":"KurmiV PatraS GurjarR NamboodiriV.Domain raptor: a comprehensive self\u2010training framework for source\u2010free adaptation. In:ACM International Conference on Multimedia (MM).2021:4277\u20104285."},{"key":"e_1_2_11_37_1","unstructured":"GaoC LiR WangP XuC.Model surgery for source\u2010free domain adaptation. In:IEEE Conference on Computer Vision and Pattern Recognition (CVPR).2023:2232\u20102241."},{"key":"e_1_2_11_38_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compmedimag.2023.102232"},{"key":"e_1_2_11_39_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598\u2010024\u201053311\u2010w"},{"key":"e_1_2_11_40_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.bspc.2022.104488"},{"key":"e_1_2_11_41_1","doi-asserted-by":"crossref","unstructured":"QiuS.Causality\u2010inspired source\u2010free domain adaptation for medical image classification. In:International Conference on Image and Graphics Springer;2023:68\u201080.","DOI":"10.1007\/978-3-031-46317-4_7"},{"key":"e_1_2_11_42_1","doi-asserted-by":"crossref","unstructured":"MaJ XuZ XiongX et\u00a0al.Source\u2010free semi\u2010supervised domain adaptation for tuberculosis recognition. In:2024 IEEE International Symposium on Biomedical Imaging (ISBI).IEEE;2024:1\u20105. doi:10.1109\/ISBI56570.2024.10635111","DOI":"10.1109\/ISBI56570.2024.10635111"}],"container-title":["Journal of Applied Clinical Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/acm2.70626","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/acm2.70626","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/acm2.70626","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,26]],"date-time":"2026-10-26T16:32:11Z","timestamp":1793032331000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/acm2.70626"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5]]},"references-count":41,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2026,5]]}},"alternative-id":["10.1002\/acm2.70626"],"URL":"https:\/\/doi.org\/10.1002\/acm2.70626","archive":["Portico"],"relation":{},"ISSN":["1526-9914","1526-9914"],"issn-type":[{"value":"1526-9914","type":"print"},{"value":"1526-9914","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5]]},"assertion":[{"value":"2025-12-06","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-04-20","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-05-23","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70626"}}