{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T05:46:46Z","timestamp":1791956806960,"version":"4.3.4"},"reference-count":24,"publisher":"Springer Science and Business Media LLC","license":[{"start":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T00:00:00Z","timestamp":1791849600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T00:00:00Z","timestamp":1791849600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100000057","name":"National Institute of General Medical Sciences","doi-asserted-by":"publisher","award":["T32 GM065841"],"award-info":[{"award-number":["T32 GM065841"]}],"id":[{"id":"10.13039\/100000057","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000057","name":"National Institute of General Medical Sciences","doi-asserted-by":"publisher","award":["T32 GM145408"],"award-info":[{"award-number":["T32 GM145408"]}],"id":[{"id":"10.13039\/100000057","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000069","name":"National Institute of Arthritis and Musculoskeletal and Skin Diseases","doi-asserted-by":"publisher","award":["F31 AR082227"],"award-info":[{"award-number":["F31 AR082227"]}],"id":[{"id":"10.13039\/100000069","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["J Digit Imaging. Inform. med."],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Magnetic resonance-guided radiotherapy (MRgRT) enables real-time tracking of tumor and organ motion for more precise radiation delivery. However, existing clinical tracking software struggles with high-velocity or non-rigid motion. Foundation models, such as Segment Anything Model 2 (SAM2) and MedSAM2, offer robust tracking with fast inference speeds and strong zero-shot accuracy that can improve with fine-tuning. Fine-tuning typically uses expert-labeled datasets, which are expensive to acquire. We explored SAM2\u2019s utility in accelerating the labeling process by developing a workflow where novices label\u2009&lt;\u20091% of frames and SAM2 automatically labels the rest. We then evaluated SAM2 and MedSAM2 for tumor tracking on an open-access MRgRT dataset in three configurations: zero-shot, fine-tuned using manually labeled data, and fine-tuned using the semi-automatically labeled data. We measured accuracy with the Dice similarity coefficient, 95th percentile (P95) Hausdorff distance, and centroid distance. MedSAM2 and SAM2 demonstrated strong zero-shot tracking accuracy with no significant difference among the model variants. Fine-tuning significantly improved Dice coefficients (0.874 to 0.891,\n                    <jats:italic>p<\/jats:italic>\n                    \u2009=\u20090.006), P95 Hausdorff distances (9.069 to 4.378\u00a0mm,\n                    <jats:italic>p<\/jats:italic>\n                    \u2009&lt;\u20090.001), and centroid distances (5.299 to 1.769\u00a0mm,\n                    <jats:italic>p<\/jats:italic>\n                    \u2009&lt;\u20090.001) relative to zero-shot, with no significant difference between using manually labeled and semi-automatically labeled data. Therefore, leveraging SAM2 to semi-automatically generate labeled training data can substantially reduce the cost of data curation and long-term model maintenance without compromising accuracy.\n                  <\/jats:p>","DOI":"10.1007\/s10278-026-02387-z","type":"journal-article","created":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T21:46:45Z","timestamp":1791928005000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Semi-Automated Image Labeling and Object Tracking for MR-Guided Radiotherapy Using a Foundation Vision Model"],"prefix":"10.1007","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-4583-8162","authenticated-orcid":false,"given":"Mike","family":"Zhu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-6467-7910","authenticated-orcid":false,"given":"Jun Hsuang","family":"Jen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7638-9599","authenticated-orcid":false,"given":"Daniel E.","family":"Hyer","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0390-3243","authenticated-orcid":false,"given":"Taylor P.","family":"Trentadue","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Timothy L.","family":"Kline","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bryan J.","family":"Traughber","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8030-8408","authenticated-orcid":false,"given":"Douglas J.","family":"Moseley","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,10,13]]},"reference":[{"key":"2387_CR1","doi-asserted-by":"publisher","unstructured":"Lagendijk JJW, Raaymakers BW, Vulpen Mv. The Magnetic Resonance Imaging\u2013Linac System. Seminars in Radiation Oncology. 2014\/07\/01;24(3). https:\/\/doi.org\/10.1016\/j.semradonc.2014.02.009.","DOI":"10.1016\/j.semradonc.2014.02.009"},{"key":"2387_CR2","doi-asserted-by":"publisher","unstructured":"Mutic S, Dempsey JF. The ViewRay System: Magnetic Resonance\u2013Guided and Controlled Radiotherapy. Seminars in Radiation Oncology. 2014\/07\/01;24(3). https:\/\/doi.org\/10.1016\/j.semradonc.2014.02.008.","DOI":"10.1016\/j.semradonc.2014.02.008"},{"issue":"7","key":"2387_CR3","doi-asserted-by":"publisher","first-page":"458","DOI":"10.1038\/s41571-022-00631-3","volume":"19","author":"PJ Keall","year":"2022","unstructured":"Keall PJ, Brighi C, Glide-Hurst C, Liney G, Liu PZY, Lydiard S, et al. Integrated MRI-guided radiotherapy \u2014 opportunities and challenges. Nature Reviews Clinical Oncology. 2022;19(7):458\u201370. https:\/\/doi.org\/10.1038\/s41571-022-00631-3.","journal-title":"Nature Reviews Clinical Oncology."},{"key":"2387_CR4","doi-asserted-by":"publisher","unstructured":"de Muinck Keizer DM, Kerkmeijer LGW, Willigenburg T, van Lier ALHMW, Hartogh MDD, van der Voort van Zyp JRN, et al. Prostate intrafraction motion during the preparation and delivery of MR-guided radiotherapy sessions on a 1.5T MR-Linac. Radiotherapy and Oncology. 2020\/10\/01;151. https:\/\/doi.org\/10.1016\/j.radonc.2020.06.044.","DOI":"10.1016\/j.radonc.2020.06.044"},{"issue":"5","key":"2387_CR5","doi-asserted-by":"publisher","DOI":"10.1016\/j.adro.2022.100934","volume":"7","author":"JE Leeman","year":"2022","unstructured":"Leeman JE, Cagney DN, Mak RH, Huynh MA, Tanguturi SK, Singer L, et al. Magnetic Resonance\u2013Guided Prostate Stereotactic Body Radiation Therapy With Daily Online Plan Adaptation: Results of a Prospective Phase 1 Trial and Supplemental Cohort. Advances in Radiation Oncology. 2022;7(5):100934. https:\/\/doi.org\/10.1016\/j.adro.2022.100934.","journal-title":"Advances in Radiation Oncology."},{"key":"2387_CR6","doi-asserted-by":"publisher","unstructured":"Jassar H, Tai A, Chen X, Keiper TD, Paulson E, Lathuili\u00e8re F, et al. Real\u2010time motion monitoring using orthogonal cine MRI during MR\u2010guided adaptive radiation therapy for abdominal tumors on 1.5T MR\u2010Linac. Medical Physics. 2023;50(5):3103\u201316. https:\/\/doi.org\/10.1002\/mp.16342.","DOI":"10.1002\/mp.16342"},{"key":"2387_CR7","doi-asserted-by":"publisher","unstructured":"Wittmann D, Paulson ES, Banerjee A, Banla LI, Schultz C, Awan M, et al. Quantification and Dosimetric Impact of Normal Organ Motion During Adaptive Radiation Therapy Planning Using a 1.5 Tesla Magnetic Resonance\u2013Equipped Linear Accelerator (MR-Linac). Advances in Radiation Oncology. 2025\/05\/01;10(5). https:\/\/doi.org\/10.1016\/j.adro.2025.101758.","DOI":"10.1016\/j.adro.2025.101758"},{"key":"2387_CR8","doi-asserted-by":"publisher","unstructured":"Rusu SD, Smith BR, St\u2010Aubin JJ, Shaffer N, Hyer DE. Surrogate gating strategies for the Elekta Unity MR\u2010Linac gating system. Journal of Applied Clinical Medical Physics. 2024 Nov 14;26(2). https:\/\/doi.org\/10.1002\/acm2.14566.","DOI":"10.1002\/acm2.14566"},{"key":"2387_CR9","doi-asserted-by":"publisher","unstructured":"Ravi N, Gabeur V, Hu Y-T, Hu R, Ryali C, Ma T, et al. SAM 2: Segment Anything in Images and Videos. 2024. https:\/\/doi.org\/10.48550\/arXiv.2408.00714.","DOI":"10.48550\/arXiv.2408.00714"},{"key":"2387_CR10","doi-asserted-by":"publisher","unstructured":"Yamagishi Y. Using Segment Anything Model 2 for Zero-Shot 3D Segmentation of Abdominal Organs in Computed Tomography Scans to Adapt Video Tracking Capabilities for 3D Medical Imaging: Algorithm Development and Validation. JMIR AI 2025;4:e72109 https:\/\/aijmirorg\/2025\/1\/e72109. 2025\u201304\u201329;4(1). https:\/\/doi.org\/10.2196\/72109.","DOI":"10.2196\/72109"},{"key":"2387_CR11","doi-asserted-by":"publisher","unstructured":"Bl\u00f6cker T, Lombardo E, Marschner SN, Belka C, Corradini S, Palacios MA, et al. MRgRT real-time target localization using foundation models for contour point tracking and promptable mask refinement. Physics in Medicine & Biology. 2024\u201312\u201324;70(1). https:\/\/doi.org\/10.1088\/1361-6560\/ad9dad.","DOI":"10.1088\/1361-6560\/ad9dad"},{"key":"2387_CR12","doi-asserted-by":"publisher","unstructured":"Ezekiel C, Khunsa F, Alexandra, Stephanie C, Hafsa F. Improving medical image segmentation with SAM2: analyzing the impact of object characteristics and finetuning on multi-planar datasets. European Journal of Radiology Artificial Intelligence. 2025;3:100034. https:\/\/doi.org\/10.1016\/j.ejrai.2025.100034.","DOI":"10.1016\/j.ejrai.2025.100034"},{"key":"2387_CR13","doi-asserted-by":"publisher","unstructured":"Kamtam DN, Shrager JB, Malla SD, Wang X, Lin N, Cardona JJ, et al. A fine-tuned foundational model SurgiSAM2 for surgical video anatomy segmentation and detection. Scientific Reports 2025 15:1. 2025\u201310\u201315;15(1). https:\/\/doi.org\/10.1038\/s41598-025-11759-4.","DOI":"10.1038\/s41598-025-11759-4"},{"key":"2387_CR14","doi-asserted-by":"publisher","unstructured":"Ma J, Yang Z, Kim S, Chen B, Baharoon M, Fallahpour A, et al. MedSAM2: Segment Anything in 3D Medical Images and Videos. 2025. https:\/\/doi.org\/10.48550\/arXiv.2504.03600.","DOI":"10.48550\/arXiv.2504.03600"},{"key":"2387_CR15","doi-asserted-by":"publisher","unstructured":"Ma J, He Y, Li F, Han L, You C, Wang B, et al. Segment anything in medical images. Nature Communications 2024 15:1. 2024\u201301\u201322;15(1). https:\/\/doi.org\/10.1038\/s41467-024-44824-z.","DOI":"10.1038\/s41467-024-44824-z"},{"key":"2387_CR16","doi-asserted-by":"publisher","first-page":"2015","DOI":"10.1155\/2015\/450341","volume":"1","author":"I Despotovi\u0107","year":"2015","unstructured":"Despotovi\u0107 I, Goossens B, Philips W. MRI Segmentation of the Human Brain: Challenges, Methods, and Applications. Computational and Mathematical Methods in Medicine. 2015 Mar 1;2015. https:\/\/doi.org\/10.1155\/2015\/450341.","journal-title":"Computational and Mathematical Methods in Medicine."},{"key":"2387_CR17","doi-asserted-by":"publisher","unstructured":"Sahiner B, Chen W, Samala RK, Petrick N. Data drift in medical machine learning: implications and potential remedies. British Journal of Radiology. 2023\/10\/01;96(1150). https:\/\/doi.org\/10.1259\/bjr.20220878.","DOI":"10.1259\/bjr.20220878"},{"issue":"7","key":"2387_CR18","doi-asserted-by":"publisher","DOI":"10.1002\/mp.17964","volume":"52","author":"Y Wang","year":"2025","unstructured":"Wang Y, Lombardo E, Thummerer A, Bl\u00f6cker T, Fan Y, Zhao Y, et al. TrackRAD2025 challenge dataset: Real-time tumor tracking for MRI-guided radiotherapy. Medical Physics. 2025;52(7):e17964. https:\/\/doi.org\/10.1002\/mp.17964.","journal-title":"Medical Physics."},{"key":"2387_CR19","doi-asserted-by":"publisher","unstructured":"Abid A, Abdalla A, Abid A, Khan D, Alfozan A, Zou J. Gradio: Hassle-Free Sharing and Testing of ML Models in the Wild. 2019. https:\/\/doi.org\/10.48550\/arXiv.1906.02569.","DOI":"10.48550\/arXiv.1906.02569"},{"key":"2387_CR20","unstructured":"HuggingFace: Spaces. https:\/\/huggingface.co\/docs\/hub\/en\/spaces Accessed March 4, 2026."},{"key":"2387_CR21","doi-asserted-by":"publisher","unstructured":"Dunn PK, Smyth GK. Chapter 11: Positive Continuous Data: Gamma and Inverse Gaussian GLMs. Springer Texts in Statistics. 2018. https:\/\/doi.org\/10.1007\/978-1-4419-0118-7_11.","DOI":"10.1007\/978-1-4419-0118-7_11"},{"key":"2387_CR22","doi-asserted-by":"publisher","unstructured":"Fast M, Schoot Avd, Lindt Tvd, Carbaat C, Heide Uvd, Sonke J-J. Tumor Trailing for Liver SBRT on the MR-Linac. International Journal of Radiation Oncology, Biology, Physics. 2019\/02\/01;103(2). https:\/\/doi.org\/10.1016\/j.ijrobp.2018.09.011.","DOI":"10.1016\/j.ijrobp.2018.09.011"},{"key":"2387_CR23","doi-asserted-by":"publisher","unstructured":"Paganelli C, Seregni M, Fattori G, Summers P, Bellomi M, Baroni G, et al. Magnetic Resonance Imaging\u2013Guided versus Surrogate-Based Motion Tracking in Liver Radiation Therapy: A Prospective Comparative Study. International Journal of Radiation Oncology, Biology, Physics. 2015\/03\/15;91(4). https:\/\/doi.org\/10.1016\/j.ijrobp.2014.12.013.","DOI":"10.1016\/j.ijrobp.2014.12.013"},{"key":"2387_CR24","doi-asserted-by":"publisher","unstructured":"Kikinis R, Pieper SD, Vosburgh KG. 3D Slicer: A Platform for Subject-Specific Image Analysis, Visualization, and Clinical Support. Intraoperative Imaging and Image-Guided Therapy. 2014. https:\/\/doi.org\/10.1007\/978-1-4614-7657-3_19.","DOI":"10.1007\/978-1-4614-7657-3_19"}],"container-title":["Journal of Imaging Informatics in Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10278-026-02387-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s10278-026-02387-z","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10278-026-02387-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T05:02:53Z","timestamp":1791954173000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s10278-026-02387-z"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,13]]},"references-count":24,"alternative-id":["2387"],"URL":"https:\/\/doi.org\/10.1007\/s10278-026-02387-z","relation":{},"ISSN":["2948-2933"],"issn-type":[{"value":"2948-2933","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,13]]},"assertion":[{"value":"29 April 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"24 September 2026","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"30 September 2026","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"13 October 2026","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"This study analyzes a publicly available, fully deidentified dataset (TrackRAD2025) which was collected with ethical approval and patient consent. No additional ethics approval was required.","order":1,"name":"Ethics","label":"Ethics Approval","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable.","order":2,"name":"Ethics","label":"Consent to Participate","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable.","order":3,"name":"Ethics","label":"Consent for Publication","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":4,"name":"Ethics","label":"Competing Interests","group":{"name":"EthicsHeading","label":"Declarations"}}]}}