{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T12:00:20Z","timestamp":1790251220291,"version":"4.1.0"},"reference-count":30,"publisher":"Wiley","issue":"3","license":[{"start":{"date-parts":[[2024,12,6]],"date-time":"2024-12-06T00:00:00Z","timestamp":1733443200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2024,12,6]],"date-time":"2024-12-06T00:00:00Z","timestamp":1733443200000},"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":[[2025,3]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>The widespread adoption of knowledge\u2010based planning in radiation oncology clinics is hindered by the lack of data and the difficulty associated with sharing medical data.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>This study aims to assess the feasibility of mitigating this challenge through federated learning\u00a0(FL): a centralized model trained with distributed datasets, while keeping data localized and private.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      This concept was tested using 273 prostate 45\u00a0Gy plans. The cases were split into a training set with 220 cases and a validation set with 53 cases. The training set was further separated into 10 subsets to simulate treatment plans from different clinics. A gradient\u2010boosting model was used to predict bladder and rectum V\n                      <jats:sub>30Gy<\/jats:sub>\n                      , V\n                      <jats:sub>35Gy<\/jats:sub>\n                      , and V\n                      <jats:sub>40Gy<\/jats:sub>\n                      . The Federated Averaging algorithm was employed to aggregate the individual model weights from distributed datasets. Grid search with five\u2010fold in\u2010training\u2010set cross\u2010validation was implemented to tune model hyperparameters. Additionally, we evaluated the robustness of the FL approach by varying the distribution of the training set data in several scenarios, including different number of sites and imbalanced data across sites.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      The mean absolute error (MAE) for the FL model (4.7%\u00a0\u00b1\u00a02.9%) is significantly lower than individual models trained separately (6.5%\u00a0\u00b1\u00a04.9%,\n                      <jats:italic>p<\/jats:italic>\n                      \u00a0&lt;\u00a00.001) and similar to a traditional centralized model (4.4%\u00a0\u00b1\u00a02.8%,\n                      <jats:italic>p<\/jats:italic>\n                      \u00a0=\u00a00.14). The federated model is robust to the number of subsets, showing MAE of 4.7%\u00a0\u00b1\u00a03.2%, 4.8%\u00a0\u00b1\u00a03.1%, 4.8%\u00a0\u00b1\u00a02.9%, 4.5%\u00a0\u00b1\u00a02.8%, 4.9%\u00a0\u00b1\u00a03.3%, and 4.8%\u00a0\u00b1\u00a03.1% for 5, 10, 15, 20, 25, and 30 subsets, respectively. For the two imbalanced datasets, the FL model achieves MAEs of 4.5%\u00a0\u00b1\u00a02.9% and 5.6%\u00a0\u00b1\u00a04.0%, non\u2010inferior to the balanced data model. For all bladder and rectum metrics, the FL model significantly outperforms 36.7% of individual models.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>This study demonstrates the potential advantages of implementing a federated model over training individual models: the proposed FL approach achieves similar prediction accuracy as a conventional model without requiring centralized data storage. Even when local models struggle to produce accurate predictions due to data scarcity, the federated model consistently maintains high performance.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.17566","type":"journal-article","created":{"date-parts":[[2024,12,6]],"date-time":"2024-12-06T11:04:23Z","timestamp":1733483063000},"page":"1408-1415","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":7,"title":["Federated learning for enhanced dose\u2013volume parameter prediction with decentralized data"],"prefix":"10.1002","volume":"52","author":[{"given":"Jiahan","family":"Zhang","sequence":"first","affiliation":[{"name":"Department of Radiation Oncology Icahn School of Medicine at Mount Sinai  New York New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yang","family":"Lei","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Icahn School of Medicine at Mount Sinai  New York New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Junyi","family":"Xia","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Icahn School of Medicine at Mount Sinai  New York New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ming","family":"Chao","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Icahn School of Medicine at Mount Sinai  New York New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tian","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Icahn School of Medicine at Mount Sinai  New York New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2024,12,6]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-83047-2_13"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13526"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2021.06.028"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2019.10.036"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2014.11.014"},{"key":"e_1_2_8_7_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2014.11.009"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.phro.2022.01.003"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejmp.2019.07.004"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1007\/s13246-022-01151-1"},{"key":"e_1_2_8_11_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejmp.2018.11.002"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2020.07.2263"},{"key":"e_1_2_8_13_1","unstructured":"Kone\u010dn\u00fdJ McMahanHB YuFX Richt\u00e1rikP SureshAT BaconD Federated Learning: Strategies for Improving Communication Efficiency. arXiv preprint arXiv:1610.05492 (2016)."},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2006.10.042"},{"key":"e_1_2_8_15_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13014-018-1068-0"},{"key":"e_1_2_8_16_1","doi-asserted-by":"crossref","unstructured":"FriedmanJ TibshiraniR HastieT.Additive Logistic Regression: A Statistical View of Boosting.Annals of Statistics\u2010ANN STATIST.2000;28:337\u2010407. doi:10.1214\/aos\/1016120463","DOI":"10.1214\/aos\/1016218223"},{"key":"e_1_2_8_17_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.3539749"},{"key":"e_1_2_8_18_1","doi-asserted-by":"crossref","unstructured":"ChenT andGuestrinC.XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining Association for Computing Machinery San Francisco 13\u201017 August 2016 785\u2010794. doi:10.1145\/2939672.2939785","DOI":"10.1145\/2939672.2939785"},{"key":"e_1_2_8_19_1","unstructured":"SommerJ SarigiannisD ParnellT Learning to tune XGboost with XGboost. arXiv preprint arXiv:1909.07218 (2019)."},{"key":"e_1_2_8_20_1","unstructured":"KapoorS PerroneV A simple and fast baseline for tuning large XGBoost models. arXiv preprint arXiv:2111.06924 (2021)."},{"key":"e_1_2_8_21_1","unstructured":"McMahanB MooreE RamageD HampsonS andy ArcasBA.Communication\u2010efficient learning of deep networks from decentralized data.Artificial intelligence and statistics PMLR 2017."},{"key":"e_1_2_8_22_1","doi-asserted-by":"crossref","unstructured":"MaC QiuX BeutelD andLaneN.Gradient\u2010less federated gradient boosting tree with learnable learning rates.Proceedings of the 3rd Workshop on Machine Learning and Systems 2023.","DOI":"10.1145\/3578356.3592579"},{"key":"e_1_2_8_23_1","unstructured":"BeutelDJ TopalT MathurA QiuX Flower: A friendly federated learning research framework. arXiv preprint arXiv:2007.14390 (2020)."},{"key":"e_1_2_8_24_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.prro.2011.11.012"},{"key":"e_1_2_8_25_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14770"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.3389\/frai.2020.00068"},{"key":"e_1_2_8_27_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/abc12c"},{"key":"e_1_2_8_28_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2020.10.019"},{"key":"e_1_2_8_29_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14114"},{"key":"e_1_2_8_30_1","unstructured":"LvK YeR HuangX YangJ ChenS.Learn What You Need in Personalized Federated Learning. arXiv preprint arXiv:2401.08327 (2024)."},{"key":"e_1_2_8_31_1","unstructured":"FallahA MokhtariA OzdaglarA Personalized federated learning: A meta\u2010learning approach. arXiv preprint arXiv:2002.07948 (2020)."}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.17566","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.17566","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.17566","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,8]],"date-time":"2026-03-08T16:25:41Z","timestamp":1772987141000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.17566"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,12,6]]},"references-count":30,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2025,3]]}},"alternative-id":["10.1002\/mp.17566"],"URL":"https:\/\/doi.org\/10.1002\/mp.17566","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,12,6]]},"assertion":[{"value":"2024-06-04","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-11-28","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-12-06","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}