{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T13:59:26Z","timestamp":1791381566021,"version":"4.3.3"},"reference-count":54,"publisher":"Wiley","issue":"8","license":[{"start":{"date-parts":[[2025,9,3]],"date-time":"2025-09-03T00:00:00Z","timestamp":1756857600000},"content-version":"vor","delay-in-days":33,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,8,1]],"date-time":"2025-08-01T00:00:00Z","timestamp":1754006400000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/501100000038","name":"Natural Sciences and Engineering Research Council of Canada","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100000038","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":[[2025,8]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Federated learning (FL) facilitates collaborative model training across multiple institutions while preserving privacy by avoiding the sharing of raw data, a critical consideration in medical imaging applications. Despite its potential, FL faces challenges such as high\u2010dimensional data, heterogeneity among datasets from different centers, and resource constraints, which limit its efficiency and effectiveness in healthcare\u00a0settings.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>This study aims to present a novel adaptive FL framework to address the challenges of data heterogeneity and resource constraints in medical imaging. The proposed framework is designed to optimize computational efficiency, enhance training processes, improve model performance, and ensure robustness against non\u2010independent and identically distributed (non\u2010IID) data across decentralized data sources.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>The proposed adaptive FL framework addresses the challenges of high\u2010dimensional data and heterogeneity in nonuniform and decentralized data sources through a key innovation. First, Federated incremental principal component analysis (FIPCA) achieves privacy\u2010preserving dimensionality reduction by aggregating local scatter matrices and means from participating centers, enabling the computation of a global PCA model. This process ensures data alignment across centers, mitigates heterogeneity, and significantly reduces computational complexity. We evaluated the framework's ability to generalize across institutions in a cross\u2010site classification task distinguishing clinically significant prostate cancer (csPCa) from non\u2010csPCa. This assessment used 1500 T2\u2010weighted (T2W) prostate MRI images from three institutions, where two centers (800 + 350 cases) were used for training and validation, and one center (350 cases) served as an independent test site.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The proposed method significantly reduced the number of global training rounds from 200 to 38, achieving a 98% reduction in energy consumption compared to the standard FedAvg algorithm. The effective use of FIPCA for dimensionality reduction enhanced generalizability, while adaptive early stopping prevented overfitting, leading to an improvement in model performance, with the area under the curve (AUC) on the unseen test center increasing from 0.68 to 0.73 (95 % CI 0.70 \u2013 0.77) on the test center's data. Additionally, the method demonstrated improved sensitivity and specificity, indicating superior classification performance. The integration of FIPCA accelerated convergence by reducing data dimensionality, while the adaptive early\u2010stopping mechanism further optimized resource utilization and prevented\u00a0overfitting.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Our adaptive FL approach efficiently handles large, heterogeneous medical imaging data, reducing training time and computational overhead, while improving model accuracy. The substantial reduction in energy consumption and accelerated convergence make it suitable for real\u2010world healthcare\u00a0settings.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.18064","type":"journal-article","created":{"date-parts":[[2025,9,4]],"date-time":"2025-09-04T05:39:21Z","timestamp":1756964361000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":6,"title":["A novel federated learning framework for medical imaging: Resource\u2010efficient approach combining PCA with early stopping"],"prefix":"10.1002","volume":"52","author":[{"given":"Negin Piran","family":"Nanekaran","sequence":"first","affiliation":[{"name":"Department of Engineering University of Guelph Guelph Ontario Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Eranga","family":"Ukwatta","sequence":"additional","affiliation":[{"name":"Department of Engineering University of Guelph Guelph Ontario Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2025,9,3]]},"reference":[{"key":"e_1_2_9_2_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-019-0727-5"},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-020-0181-6"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-020-0265-z"},{"key":"e_1_2_9_5_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-021-00431-6"},{"key":"e_1_2_9_6_1","first-page":"1273","volume-title":"Artificial Intelligence and Statistics","author":"McMahan B","year":"2017"},{"key":"e_1_2_9_7_1","doi-asserted-by":"crossref","unstructured":"NilssonA SmithS UlmG GustavssonE JirstrandM.A performance evaluation of federated learning algorithms. In:Proceedings of the Second Workshop on Distributed Infrastructures for Deep Learning New York USA 1\u20138.Association for Computing Machinery;2018.","DOI":"10.1145\/3286490.3286559"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1561\/2200000083"},{"key":"e_1_2_9_9_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0302539"},{"key":"e_1_2_9_10_1","unstructured":"QuL BalachandarN RubinDL.An experimental study of data heterogeneity in federated learning methods for medical imaging.2021.arXiv preprintarXiv:2105.05929."},{"key":"e_1_2_9_11_1","doi-asserted-by":"crossref","unstructured":"ChenW BhardwajK MarculescuR.FedMax: Mitigating activation divergence for accurate and communication\u2010efficient federated learning. In:Joint European Conference on Machine Learning and Knowledge Discovery in Databases Berlin Germany 563\u2013579.Springer;2020.","DOI":"10.1007\/978-3-030-67661-2_21"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-022-00515-2"},{"key":"e_1_2_9_13_1","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2020.2973615"},{"key":"e_1_2_9_14_1","doi-asserted-by":"crossref","unstructured":"JiangM WangZ DouQ.Harmofl: Harmonizing local and global drifts in federated learning on heterogeneous medical images. In:Proceedings of the AAAI Conference on Artificial Intelligence2022;36:1087\u20101095.","DOI":"10.1609\/aaai.v36i1.19993"},{"key":"e_1_2_9_15_1","first-page":"5132","article-title":"SCAFFOLD: stochastic Controlled Averaging for Federated Learning","volume":"119","author":"Karimireddy SP","year":"2020","journal-title":"Proc Mach Learn Res"},{"key":"e_1_2_9_16_1","doi-asserted-by":"crossref","unstructured":"LiQ HeB SongD.Model\u2010contrastive federated learning. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Nashville USA 10713\u201310722.IEEE;2021.","DOI":"10.1109\/CVPR46437.2021.01057"},{"key":"e_1_2_9_17_1","unstructured":"AcarD ZhaoY NavarroR MattinaM WhatmoughP SaligramaV Federated learning based on dynamic regularization (FedDyn). In:Proceedings of the 9th International Conference on Learning Representations (ICLR 2021).2021."},{"key":"e_1_2_9_18_1","unstructured":"MohriM SivekG SureshAT.Agnostic federated learning. In:International Conference on Machine Learning (ICML)2019."},{"key":"e_1_2_9_19_1","unstructured":"ArivazhaganMK AggarwalV SinghAT ChoudharyS.Federated learning with personalization layers. In:NeurIPS Workshop on Federated Learning for Data Privacy and Confidentiality2019."},{"key":"e_1_2_9_20_1","unstructured":"FallahA MokhtariA OzdaglarA.Personalized federated learning with theoretical guarantees: a model\u2010agnostic meta\u2010learning approach. In:Advances in Neural Information Processing Systems (NeurIPS)2020."},{"key":"e_1_2_9_21_1","unstructured":"DinhCT TranNH NguyenT NguyenC DinhL NguyenDN.Personalized federated learning with moreau envelopes. In:Advances in Neural Information Processing Systems (NeurIPS)2020."},{"key":"e_1_2_9_22_1","unstructured":"LiT SahuAK TalwalkarA SmithV.Ditto: Fair and robust federated learning through personalization. In:International Conference on Machine Learning (ICML)2021."},{"key":"e_1_2_9_23_1","unstructured":"CollinsL HassaniHS RangwalaH.Exploiting shared representations for personalized federated learning. In:International Conference on Machine Learning (ICML)2021."},{"key":"e_1_2_9_24_1","unstructured":"GhoshAK ChungJ YinD RamchandranK.Efficient model aggregation in federated learning via probabilistic clustering. In:International Conference on Learning Representations (ICLR)2020."},{"key":"e_1_2_9_25_1","unstructured":"MarfoqO BelletA TommasiT.Federated learning under fairness constraints. In:Advances in Neural Information Processing Systems (NeurIPS)2021."},{"key":"e_1_2_9_26_1","unstructured":"HsuT\u2010MH QiH BrownMatthew.Measuring the effects of non\u2010identical data distribution for federated visual classification.arXiv preprint arXiv:1909.06335.2019."},{"key":"e_1_2_9_27_1","unstructured":"ReddiSJ CharlesZ ZaheerM GarrettZ RushK Kone\u010dn\u00fdJ et\u00a0al.Adaptive federated optimization. In:Proceedings of the 9th International Conference on Learning Representations (ICLR2021).2021. arXiv:2003.00295."},{"key":"e_1_2_9_28_1","first-page":"132","volume-title":"Bildverarbeitung f\u00fcr die Medizin 2022","author":"Kades F","year":"2022"},{"key":"e_1_2_9_29_1","unstructured":"AlekseenkoV SunT BruellJ ZhangY LiH RueckertD Distance\u2010Aware Non\u2010IID Federated Learning for Generalization and Personalization in Medical Imaging Segmentation. In:Proceedings of Machine Learning Research (MIDL2024).2024;250."},{"key":"e_1_2_9_30_1","doi-asserted-by":"crossref","unstructured":"DarziS ChenX ZhangX XiaY XuD.Tackling heterogeneity in medical federated learning via aligning vision transformers.2024.arXiv preprintarXiv:2310.09444.","DOI":"10.1016\/j.artmed.2024.102936"},{"key":"e_1_2_9_31_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.101992"},{"key":"e_1_2_9_32_1","first-page":"4044","article-title":"SplitAVG: a heterogeneity\u2010aware federated deep learning method for medical imaging","volume":"26","author":"Zhang H","year":"2022","journal-title":"IEEE J Biomed Health Inf"},{"key":"e_1_2_9_33_1","unstructured":"LiX HeY SongY WangC SunY LiuT.FedBN: federated learning on non\u2010IID features via local batch normalization. In:Proceedings of ICLR2021."},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-022-05539-7"},{"key":"e_1_2_9_35_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2024.3387639"},{"key":"e_1_2_9_36_1","doi-asserted-by":"publisher","DOI":"10.1109\/LWC.2022.3149783"},{"key":"e_1_2_9_37_1","doi-asserted-by":"crossref","unstructured":"WilkinsG DiS CalhounJC et\u00a0al.FedSZ: leveraging error\u2010bounded lossy compression for federated learning communications. In:2024 IEEE 44th International Conference on Distributed Computing Systems (ICDCS).IEEE;2024:577\u2010588.","DOI":"10.1109\/ICDCS60910.2024.00060"},{"key":"e_1_2_9_38_1","doi-asserted-by":"publisher","DOI":"10.3390\/jsan11040070"},{"key":"e_1_2_9_39_1","doi-asserted-by":"publisher","DOI":"10.1109\/JIOT.2021.3120998"},{"key":"e_1_2_9_40_1","doi-asserted-by":"crossref","unstructured":"AbadiM ChuA GoodfellowI et\u00a0al Deep learning with differential privacy. In:Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security (CCS \u201916).New York NY:ACM;2016.308\u2010318. doi:10.1145\/2976749.2978318","DOI":"10.1145\/2976749.2978318"},{"key":"e_1_2_9_41_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-022-07186-4"},{"key":"e_1_2_9_42_1","unstructured":"TavakoliF WuC LiuJ MaC XuM.A comprehensive view of personalized federated learning on heterogeneous clinical datasets.2023.arXiv preprintarXiv:2309.16825."},{"key":"e_1_2_9_43_1","doi-asserted-by":"publisher","DOI":"10.1016\/S1470\u20102045(24)00123\u20104"},{"key":"e_1_2_9_44_1","doi-asserted-by":"publisher","DOI":"10.1117\/12.2688568"},{"key":"e_1_2_9_45_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2022.3141913"},{"key":"e_1_2_9_46_1","doi-asserted-by":"publisher","DOI":"10.2307\/2531595"},{"key":"e_1_2_9_47_1","unstructured":"SukH\u2010Il LiuM YanP LianC.Privacy\u2010preserving federated brain tumour segmentation. In:International Workshop on Machine Learning in Medical Imaging.Springer;2019:92\u2010100."},{"key":"e_1_2_9_48_1","doi-asserted-by":"crossref","unstructured":"YangQ ZhangJ HaoW SpellG CarinL.FLOP: Federated learning on medical datasets using partial networks. In:Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining.ACM;2021:3845\u20103853.","DOI":"10.1145\/3447548.3467185"},{"key":"e_1_2_9_49_1","doi-asserted-by":"publisher","DOI":"10.1093\/jamia\/ocy017"},{"key":"e_1_2_9_50_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-020-69250-1"},{"key":"e_1_2_9_51_1","doi-asserted-by":"publisher","DOI":"10.1109\/JSAC.2019.2904348"},{"key":"e_1_2_9_52_1","unstructured":"Kone\u010dn\u00fdJ McMahanHB YuFX Richt\u00e1rikP SureshAT BaconD.Federated learning: strategies for improving communication efficiency.2016.arXiv preprintarXiv:1610.05492."},{"key":"e_1_2_9_53_1","doi-asserted-by":"crossref","unstructured":"ShokriR ShmatikovV.Privacy\u2010preserving deep learning. In:Proceedings of the 22nd ACM SIGSAC Conference on Computer and Communications Security (CCS).ACM;2015:1310\u20101321.","DOI":"10.1145\/2810103.2813687"},{"key":"e_1_2_9_54_1","unstructured":"VepakommaP GuptaO SwedishT RaskarR.Split learning for health: distributed deep learning without sharing raw patient data.2018.arXiv preprintarXiv:1812.00564. (Presented at NeurIPS 2018 Workshop)."},{"key":"e_1_2_9_55_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2009.191"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.18064","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.18064","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.18064","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T13:37:42Z","timestamp":1791380262000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.18064"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8]]},"references-count":54,"journal-issue":{"issue":"8","published-print":{"date-parts":[[2025,8]]}},"alternative-id":["10.1002\/mp.18064"],"URL":"https:\/\/doi.org\/10.1002\/mp.18064","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,8]]},"assertion":[{"value":"2024-12-13","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-07-28","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-09-03","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e18064"}}