{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T04:18:20Z","timestamp":1773893900444,"version":"3.50.1"},"reference-count":45,"publisher":"Wiley","issue":"1","license":[{"start":{"date-parts":[[2025,12,26]],"date-time":"2025-12-26T00:00:00Z","timestamp":1766707200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2025,12,26]],"date-time":"2025-12-26T00:00:00Z","timestamp":1766707200000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2026,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Cone\u2010beam computed tomography (CBCT) has gained considerable traction in medical imaging due to its ability to provide on\u2010board three\u2010dimensional (3D) volumetric imaging with relatively low radiation dose, aiding in tracking anatomical changes. However, artifacts such as scatter, noise, and respiratory motion, especially in thoracic imaging, complicate accurate segmentation of organs\u2010at\u2010risk (OARs). Additionally, patient\u2010specific anatomical and tumor changes during treatment further challenge precise delineation, potentially affecting radiation therapy outcomes.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>This study aims to address the challenges in accurate segmentation of OARs in CBCT imaging, considering the impact of imaging artifacts and patient\u2010specific anatomical and tumor changes, which are critical for improving the efficacy of radiation therapy.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>A 3D patient\u2010 and fraction\u2010specific scalable and transferable U\u2010Net (PFS\u2010STU\u2010Net) was developed using a large\u2010scale pre\u2010trained STU\u2010Net with LoRA\u2010based parameter\u2010efficient fine\u2010tuning. The framework first adapts a general STU\u2010Net model (G\u2010STU\u2010Net) to the CBCT domain using paired planning CT (pCT) and deformed CBCT (dCBCT) images, then incrementally fine\u2010tunes the model with patient\u2010specific CBCT fractions to capture anatomical evolution throughout treatment. Segmentation accuracy was evaluated using dice similarity coefficient (DSC), 95th\u2010percentile Hausdorff distance (HD95), and mean surface distance (MSD) on 33 lung SBRT patients (5 fractions each) and an external 10\u2010patient cohort. Statistical comparisons were performed using paired Wilcoxon signed\u2010rank tests with Benjamini\u2013Hochberg false discovery rate (FDR) correction for multiple comparisons, with statistical significance set at .<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Compared with the pre\u2010trained STU\u2010Net, the proposed PFS\u2010STU\u2010Net achieved statistically significant improvements across all organs (, FDR\u2010corrected), with large effect sizes (Cliff's  for the esophagus and spinal cord). DSC increased from 0.60 to 0.87 for the esophagus, 0.79 to 0.92 for the heart, 0.83 to 0.98 for the left lung, 0.84 to 0.98 for the right lung, and 0.74 to 0.97 for the spinal cord. Average HD95 and MSD were reduced by more than 50%. External validation demonstrated consistent cross\u2010institutional performance (average DSC\u00a0=\u00a00.91\u00a0\u00b1\u00a00.07).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Integrating large\u2010scale pretraining with LoRA\u2010based patient\u2010 and fraction\u2010specific adaptation significantly enhances CBCT multi\u2010organ segmentation accuracy and clinical transferability. The PFS\u2010STU\u2010Net framework demonstrates robust performance across institutions and provides a scalable, efficient solution for real\u2010time CBCT\u2010guided adaptive radiotherapy.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.70251","type":"journal-article","created":{"date-parts":[[2025,12,26]],"date-time":"2025-12-26T15:42:30Z","timestamp":1766763750000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Patient and fraction specific model adaptation for cone beam computed tomography (CBCT) multi\u2010organ segmentation"],"prefix":"10.1002","volume":"53","author":[{"given":"Xiaoqian","family":"Chen","sequence":"first","affiliation":[{"name":"Department of Radiation Oncology Winship Cancer Institute Emory University School of Medicine  Atlanta Georgia USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Richard L. J.","family":"Qiu","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Winship Cancer Institute Emory University School of Medicine  Atlanta Georgia USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Vanessa","family":"Wildman","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Winship Cancer Institute Emory University School of Medicine  Atlanta Georgia USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ryan James","family":"Sanford","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Winship Cancer Institute Emory University School of Medicine  Atlanta Georgia USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Keyur Devendra","family":"Shah","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Winship Cancer Institute Emory University School of Medicine  Atlanta Georgia USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tonghe","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Medical Physics Memorial Sloan Kettering Cancer Center  New York New York USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuheng","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Winship Cancer Institute Emory University School of Medicine  Atlanta Georgia USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chih\u2010Wei","family":"Chang","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Winship Cancer Institute Emory University School of Medicine  Atlanta Georgia USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Joseph W.","family":"Shelton","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Winship Cancer Institute Emory University School of Medicine  Atlanta Georgia USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Aparna H.","family":"Kesarwala","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Winship Cancer Institute Emory University School of Medicine  Atlanta Georgia USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiaofeng","family":"Yang","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Winship Cancer Institute Emory University School of Medicine  Atlanta Georgia USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","published-online":{"date-parts":[[2025,12,26]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.3322\/caac.21660"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/ac9175"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2017.04.023"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13014\u2010023\u201002340\u20102"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.3390\/cancers16061206"},{"key":"e_1_2_8_7_1","doi-asserted-by":"publisher","DOI":"10.1097\/CM9.0000000000003299"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.3390\/ijms22168661"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2021.770382"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.15083"},{"key":"e_1_2_8_11_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00784\u2010023\u201005048\u20105"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-87589-3_62"},{"key":"e_1_2_8_13_1","doi-asserted-by":"publisher","DOI":"10.1049\/ipr2.12419"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.4108\/eai.12\u20104\u20102021.169184"},{"key":"e_1_2_8_15_1","article-title":"Transfusion: understanding transfer learning for medical imaging","volume":"32","author":"Raghu M","year":"2019","journal-title":"Adv Neural Inf Process Syst"},{"key":"e_1_2_8_16_1","unstructured":"KimB OhY YeJC Diffusion adversarial representation learning for self\u2010supervised vessel segmentation(2022) arXiv preprint arXiv:2209.14566."},{"key":"e_1_2_8_17_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.17328"},{"key":"e_1_2_8_18_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/ac5f6e"},{"key":"e_1_2_8_19_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.17463"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2023.04.026"},{"key":"e_1_2_8_21_1","article-title":"Universal language model fine\u2010tuning for text classification","volume":"1801","author":"Howard J","year":"2018","journal-title":"arXiv"},{"key":"e_1_2_8_22_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/aa4f6d"},{"key":"e_1_2_8_23_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/ac92ba"},{"key":"e_1_2_8_24_1","doi-asserted-by":"publisher","DOI":"10.3390\/jpm12050680"},{"key":"e_1_2_8_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2020.101693"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.1111\/1754\u20109485.13261"},{"key":"e_1_2_8_27_1","article-title":"LoRA: Low\u2010rank adaptation of large language models","volume":"2106","author":"Hu EJ","year":"2021","journal-title":"arXiv"},{"key":"e_1_2_8_28_1","doi-asserted-by":"crossref","unstructured":"YuY YangCHH KolehmainenJ et\u00a0al. Low\u2010rank adaptation of large language model rescoring for parameter\u2010efficient speech recognition inIEEE Automatic Speech Recognition and Understanding Workshop (ASRU).2023 pp.1\u20108. IEEE.","DOI":"10.1109\/ASRU57964.2023.10389632"},{"key":"e_1_2_8_29_1","article-title":"The expressive power of low\u2010rank adaptation","volume":"2310","author":"Zeng Y","year":"2023","journal-title":"arXiv"},{"key":"e_1_2_8_30_1","article-title":"LoraHub: Efficient cross\u2010task generalization via dynamic LoRA composition","volume":"2307","author":"Huang C","year":"2023","journal-title":"arXiv"},{"key":"e_1_2_8_31_1","article-title":"LoRA+: efficient low rank adaptation of large models","volume":"2402","author":"Hayou S","year":"2024","journal-title":"arXiv"},{"issue":"11","key":"e_1_2_8_32_1","first-page":"12799","article-title":"On the effectiveness of parameter\u2010efficient fine\u2010tuning","volume":"37","author":"Fu Z","year":"2023","journal-title":"Proc. AAAI Conf. Artif. Intell"},{"key":"e_1_2_8_33_1","article-title":"Prefix\u2010tuning: Optimizing continuous prompts for generation","volume":"2101","author":"Li XL","year":"2021","journal-title":"arXiv"},{"key":"e_1_2_8_34_1","article-title":"DoRA: Weight\u2010decomposed low\u2010rank adaptation","volume":"2402","author":"Liu SY","year":"2024","journal-title":"arXiv"},{"key":"e_1_2_8_35_1","article-title":"STU\u2010Net: Scalable and transferable medical image segmentation models empowered by large\u2010scale supervised pre\u2010training","volume":"2304","author":"Huang Z","year":"2023","journal-title":"arXiv"},{"key":"e_1_2_8_36_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2023.102996"},{"key":"e_1_2_8_37_1","article-title":"Customized Segment Anything Model for medical image segmentation","volume":"2304","author":"Zhang K","year":"2023","journal-title":"arXiv"},{"key":"e_1_2_8_38_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.16799"},{"key":"e_1_2_8_39_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-024-00961-0"},{"key":"e_1_2_8_40_1","first-page":"599","volume-title":"Medical Imaging 2025: Image\u2010Guided Procedures, Robotic Interventions, and Modeling","author":"Chen X","year":"2025"},{"key":"e_1_2_8_41_1","doi-asserted-by":"publisher","DOI":"10.1201\/9781003353751-5"},{"key":"e_1_2_8_42_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13104-022-06096-y"},{"key":"e_1_2_8_43_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00362\u2010012\u20100443\u20104"},{"key":"e_1_2_8_44_1","doi-asserted-by":"publisher","DOI":"10.1111\/1471\u20100528.16706"},{"key":"e_1_2_8_45_1","unstructured":"TomczakM Tomczak&E \u201cThe need to report effect size estimates revisited. An overview of some recommended measures of effect size\u201d. (2014)."},{"key":"e_1_2_8_46_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/add4ba"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70251","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.70251","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70251","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T01:58:10Z","timestamp":1773885490000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.70251"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,26]]},"references-count":45,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,1]]}},"alternative-id":["10.1002\/mp.70251"],"URL":"https:\/\/doi.org\/10.1002\/mp.70251","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,26]]},"assertion":[{"value":"2025-02-21","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-12-11","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-12-26","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70251"}}