{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T15:11:19Z","timestamp":1789571479662,"version":"build-2803163510"},"reference-count":31,"publisher":"Wiley","issue":"1","license":[{"start":{"date-parts":[[2025,12,21]],"date-time":"2025-12-21T00:00:00Z","timestamp":1766275200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"},{"start":{"date-parts":[[2025,12,21]],"date-time":"2025-12-21T00:00:00Z","timestamp":1766275200000},"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>Accurate and consistent image segmentation across longitudinal scans is essential in many clinical applications, including surveillance, treatment monitoring, and adaptive interventions. While personalized model adaptation using patient\u2010specific prior scans has shown promise, current approaches typically rely on fixed training durations and lack mechanisms to determine optimal stopping points on a per\u2010patient basis, particularly in the absence of validation labels.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>We propose an uncertainty\u2010guided test\u2010time optimization (TTO) framework that dynamically adjusts the personalization duration for each patient using a validation\u2010free stopping criterion based on predictive uncertainty.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Our framework personalizes a generalized segmentation model using patient\u2010specific prior imaging and selects the optimal checkpoint based on the minimum voxel\u2010wise predictive uncertainty, estimated via Monte Carlo Dropout (TTO\u2010MCD) or Deep Ensembling (TTO\u2010DE). We evaluated the approach on three datasets: 214 pancreas (CT) scans, 243 liver (CT) scans, and 175 head\u2010and\u2010neck tumor (MRI) scans, each containing a subset of patients with paired longitudinal scans to enable patient\u2010specific personalization. Each patient's follow\u2010up scan was held out for testing. As a baseline, we implemented a fixed\u2010epoch personalization strategy (Pre\u2010TTO) using a fivefold cross\u2010test design to emulate deployable model selection without test label leakage.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      TTO methods consistently outperformed the Pre\u2010TTO and unpersonalized baseline across standard metrics, including the Dice Similarity Coefficient (DSC), 95th percentile Hausdorff Distance (HD95), Mean Surface Distance (MSD), and the proposed LogPenalty Score (LPS), which provides a bounded, interpretable scale that jointly reflects volumetric and boundary fidelity. Paired\n                      <jats:italic>t<\/jats:italic>\n                      \u2010tests confirmed statistically significant improvements for pancreas and liver datasets (\n                      <jats:italic>p<\/jats:italic>\n                      \u00a0\u00a0&lt;\u00a0\u00a00.05), while favorable trends were observed in the head\u2010and\u2010neck dataset despite greater anatomical variability. Both TTO\u2010MCD and TTO\u2010DE achieved near\u2010optimal performance without requiring access to labels at test time.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>\n                      Uncertainty\u2010guided TTO provides a robust, validation\u2010free strategy for optimizing patient\u2010specific segmentation models in longitudinal medical imaging. By tailoring personalization based on predictive uncertainty, our method improves segmentation quality across a range of imaging modalities and anatomical targets. This framework supports broad clinical deployment of personalized AI and motivates future extensions to contextual integration and multi\u2010label segmentation. Code is publicly available at\n                      <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" xlink:href=\"https:\/\/github.com\/jchun-ai\/uncertainty-tto\">https:\/\/github.com\/jchun\u2010ai\/uncertainty\u2010tto<\/jats:ext-link>\n                      .\n                    <\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.70206","type":"journal-article","created":{"date-parts":[[2025,12,22]],"date-time":"2025-12-22T04:06:21Z","timestamp":1766376381000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Uncertainty\u2010guided test\u2010time optimization for personalizing segmentation models in longitudinal medical imaging"],"prefix":"10.1002","volume":"53","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9695-6079","authenticated-orcid":false,"given":"Jaehee","family":"Chun","sequence":"first","affiliation":[{"name":"Department of Imaging Physics The University of Texas MD Anderson Cancer Center  Houston Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-3015-3570","authenticated-orcid":false,"given":"Austin","family":"Castelo","sequence":"additional","affiliation":[{"name":"Department of Imaging Physics The University of Texas MD Anderson Cancer Center  Houston Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0995-3695","authenticated-orcid":false,"given":"McKell","family":"Woodland","sequence":"additional","affiliation":[{"name":"Department of Imaging Physics The University of Texas MD Anderson Cancer Center  Houston Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-2871-6692","authenticated-orcid":false,"given":"Caleb","family":"O'Connor","sequence":"additional","affiliation":[{"name":"Department of Imaging Physics The University of Texas MD Anderson Cancer Center  Houston Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-9298-6644","authenticated-orcid":false,"given":"Mais Al","family":"Taie","sequence":"additional","affiliation":[{"name":"Department of Imaging Physics The University of Texas MD Anderson Cancer Center  Houston Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-2078-4718","authenticated-orcid":false,"given":"Mohamed","family":"Eltaher","sequence":"additional","affiliation":[{"name":"Department of Imaging Physics The University of Texas MD Anderson Cancer Center  Houston Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0178-0921","authenticated-orcid":false,"given":"Aashish","family":"Gupta","sequence":"additional","affiliation":[{"name":"Department of Imaging Physics The University of Texas MD Anderson Cancer Center  Houston Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0327-7983","authenticated-orcid":false,"given":"Bilel","family":"Daoud","sequence":"additional","affiliation":[{"name":"Department of Imaging Physics The University of Texas MD Anderson Cancer Center  Houston Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-3215-5555","authenticated-orcid":false,"given":"Shanli","family":"Ding","sequence":"additional","affiliation":[{"name":"Department of Imaging Physics The University of Texas MD Anderson Cancer Center  Houston Texas USA"},{"name":"Graduate School of Biomedical Sciences The University of Texas MD Anderson Cancer Center  Houston Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-2546-1524","authenticated-orcid":false,"given":"Jeddy","family":"Bennett","sequence":"additional","affiliation":[{"name":"Department of Imaging Physics The University of Texas MD Anderson Cancer Center  Houston Texas USA"},{"name":"Department of Mathematics Brigham Young University  Provo Utah USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7923-9978","authenticated-orcid":false,"given":"Anirban","family":"Maitra","sequence":"additional","affiliation":[{"name":"Department of Translational Molecular Pathology The University of Texas MD Anderson Cancer Center  Houston Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7983-3982","authenticated-orcid":false,"given":"Matthew A.","family":"Firpo","sequence":"additional","affiliation":[{"name":"Division of General Surgery University of Utah  Utah USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6876-2752","authenticated-orcid":false,"given":"Kimberly","family":"Kirkwood","sequence":"additional","affiliation":[{"name":"Division of Surgical Oncology University of California San Francisco  San Francisco California USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7675-3461","authenticated-orcid":false,"given":"Eugene J.","family":"Koay","sequence":"additional","affiliation":[{"name":"Department of Gastrointestinal Radiation Oncology The University of Texas MD Anderson Cancer Center  Houston Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9364-5040","authenticated-orcid":false,"given":"Kristy K.","family":"Brock","sequence":"additional","affiliation":[{"name":"Department of Imaging Physics The University of Texas MD Anderson Cancer Center  Houston Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2025,12,21]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.3748\/wjg.v28.i6.624"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.1136\/flgastro\u20102018\u2010101076"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.222778"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1102\/1470\u20107330.2004.0011"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.1102\/1470\u20107330.2012.0044"},{"key":"e_1_2_8_7_1","doi-asserted-by":"publisher","DOI":"10.3390\/cancers16061206"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1148\/rycan.230011"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejca.2008.10.026"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1593\/tlo.11232"},{"key":"e_1_2_8_11_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejrad.2015.12.019"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.clon.2024.07.006"},{"key":"e_1_2_8_13_1","doi-asserted-by":"crossref","unstructured":"RonnebergerO FischerP BroxT U\u2010net: convolutional networks for biomedical image segmentation. In:Medical Image Computing and Computer\u2010Assisted Intervention\u2013MICCAI 2015: 18th International Conference Munich Germany October 5\u20139 2015 Proceedings Part III 18.Springer;2015:234\u2010241.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41592\u2010020\u201001008\u2010z"},{"key":"e_1_2_8_15_1","doi-asserted-by":"crossref","unstructured":"TangY YangD LiW et\u00a0al.Self\u2010supervised pre\u2010training of swin transformers for 3d medical image analysis. In:Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition.2022:20730\u201020740.","DOI":"10.1109\/CVPR52688.2022.02007"},{"key":"e_1_2_8_16_1","doi-asserted-by":"publisher","DOI":"10.59275\/j.melba.2024\u2010g93a"},{"key":"e_1_2_8_17_1","volume-title":"Dataset Shift in Machine Learning","author":"Joaquin Q\u2010C","year":"2009"},{"key":"e_1_2_8_18_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.15352"},{"key":"e_1_2_8_19_1","doi-asserted-by":"crossref","unstructured":"HuM SongT GuY et\u00a0al.Fully test\u2010time adaptation for image segmentation. In:Medical Image Computing and Computer Assisted Intervention\u2013MICCAI 2021: 24th International Conference Strasbourg France September 27\u2013October 1 2021 Proceedings Part III 24.Springer;2021:251\u2010260.","DOI":"10.1007\/978-3-030-87199-4_24"},{"key":"e_1_2_8_20_1","doi-asserted-by":"crossref","unstructured":"KaraniN ChaitanyaK BaumgartnerC KonukogluE A lifelong learning approach to brain MR segmentation across scanners and protocols. In:International Conference on Medical Image Computing and Computer\u2010Assisted Intervention.Springer;2018:476\u2010484.","DOI":"10.1007\/978-3-030-00928-1_54"},{"key":"e_1_2_8_21_1","doi-asserted-by":"crossref","unstructured":"ValindriaVV PawlowskiN RajchlM et\u00a0al.Multi\u2010modal learning from unpaired images: application to multi\u2010organ segmentation in CT and MRI. In:Proceedings \u2010 2018 IEEE Winter Conference on Applications of Computer Vision WACV 2018. Vol2018\u2010January.Institute of Electrical and Electronics Engineers Inc.;2018:547\u2010556. doi:10.1109\/WACV.2018.00066","DOI":"10.1109\/WACV.2018.00066"},{"key":"e_1_2_8_22_1","doi-asserted-by":"publisher","DOI":"10.3390\/cancers13040702"},{"key":"e_1_2_8_23_1","doi-asserted-by":"crossref","unstructured":"ElmahdyMS AhujaT Van Der HeideUA StaringM Patient\u2010Specific finetuning of deep learning models for adaptive radiotherapy in prostate CT.Proceedings\u2014International Symposium on Biomedical Imaging.2020:577\u2010580. doi:10.1109\/ISBI45749.2020.9098702","DOI":"10.1109\/ISBI45749.2020.9098702"},{"key":"e_1_2_8_24_1","doi-asserted-by":"publisher","DOI":"10.1117\/1.jmi.7.6.064003"},{"key":"e_1_2_8_25_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.16056"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.16616"},{"key":"e_1_2_8_27_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.adro.2024.101580"},{"key":"e_1_2_8_28_1","unstructured":"GalY GhahramaniZ.Dropout as a bayesian approximation: Representing model uncertainty in deep learning. In:International Conference on Machine Learning.2016:1050\u20101059."},{"key":"e_1_2_8_29_1","unstructured":"LakshminarayananB PritzelA DeepmindCB Simple and Scalable Predictive Uncertainty Estimation Using Deep Ensembles."},{"key":"e_1_2_8_30_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-83274-1"},{"key":"e_1_2_8_31_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2004.828354"},{"key":"e_1_2_8_32_1","volume-title":"Bayesian Theory","author":"Bernardo JM","year":"2009"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70206","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.70206","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70206","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T01:57:54Z","timestamp":1773885474000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.70206"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,21]]},"references-count":31,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,1]]}},"alternative-id":["10.1002\/mp.70206"],"URL":"https:\/\/doi.org\/10.1002\/mp.70206","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,21]]},"assertion":[{"value":"2025-08-19","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-11-22","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-12-21","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70206"}}