{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T09:24:58Z","timestamp":1787045098885,"version":"build-2736575974"},"reference-count":40,"publisher":"IOP Publishing","issue":"16","license":[{"start":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T00:00:00Z","timestamp":1787011200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/publishingsupport.iopscience.iop.org\/iop-standard\/v1"},{"start":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T00:00:00Z","timestamp":1787011200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Sci-ence Foundation of China","doi-asserted-by":"crossref","award":["12505420"],"award-info":[{"award-number":["12505420"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Joint Founding Project of Innovation Research Institute in Xijing Hospital","award":["LHJJ24YG01"],"award-info":[{"award-number":["LHJJ24YG01"]}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Phys. Med. Biol."],"published-print":{"date-parts":[[2026,8,28]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Accurate segmentation of the gross tumor volume from computed tomography images is a core step in the development of precise radiotherapy planning for esophageal cancer, which directly affects treatment efficacy and normal tissue protection. Recently, foundation models represented by the segment anything model (SAM) have been increasingly applied in medical image segmentation, and several studies have extended them to tumor target segmentation with promising progress. However, existing SAM-based segmentation methods rely on manual prompts, leading to significant limitations in esophageal cancer. Manual localization of tumor prompt points is difficult, and inappropriate prompts easily cause segmentation errors, increasing the risk of damage to organs-at-risk during radiotherapy. To address this, this study proposes a segmentation method based on SAM that streamlines the inference process by using a coarse-to-fine prompt generation strategy. Inspired by the stepwise refinement of clinical target volume delineation, the core design lies in a coarse-to-fine prompt generation strategy. Specifically, coarse segmentation results generated by nnUNet are first converted into initial prompts to provide global anatomical priors for SAM. Furthermore, a dynamic iterative prompt update mechanism is introduced. During inference, the iterative correction prompts are derived solely from the model\u2019s own prediction uncertainty, forming a closed-loop refinement that gradually improves segmentation accuracy. Experimental results based on multi-center datasets show that this method, which iteratively updates prompts based on its own uncertainty, achieves better segmentation performance than existing single-pass methods on both internal and external validation sets. The framework is highly consistent with the logic of clinical target volume delineation and can provide a reliable efficient scheme for the formulation of precise radiotherapy plans for esophageal cancer.<\/jats:p>","DOI":"10.1088\/1361-6560\/ae9688","type":"journal-article","created":{"date-parts":[[2026,8,6]],"date-time":"2026-08-06T22:57:03Z","timestamp":1786057023000},"page":"165005","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Dynamic iterative coarse-to-fine prompt learning based on SAM for precise esophageal cancer gross target volume segmentation"],"prefix":"10.1088","volume":"71","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-5455-5658","authenticated-orcid":true,"given":"Yuxuan","family":"Yao","sequence":"first","affiliation":[{"name":"Air Force Medical University"},{"name":"Co-authors"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongfei","family":"Sun","sequence":"additional","affiliation":[{"name":"Xijing Hospital of Air Force Medical University"},{"name":"Co-authors"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-2665-4104","authenticated-orcid":true,"given":"Chengwei","family":"Chen","sequence":"additional","affiliation":[{"name":"Air Force Medical University"},{"name":"Co-authors"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tianci","family":"Liu","sequence":"additional","affiliation":[{"name":"Air Force Medical University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wencheng","family":"Zhang","sequence":"additional","affiliation":[{"name":"Tianjin Medical University Cancer Institute & Hospital"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qifeng","family":"Wang","sequence":"additional","affiliation":[{"name":"Sichuan Cancer Hospital & Institute"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0984-5128","authenticated-orcid":true,"given":"Yang","family":"Liu","sequence":"additional","affiliation":[{"name":"Air Force Medical University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lina","family":"Zhao","sequence":"additional","affiliation":[{"name":"Xijing Hospital of Air Force Medical University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9746-0032","authenticated-orcid":true,"given":"Dong","family":"Huang","sequence":"additional","affiliation":[{"name":"Air Force Medical University"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2026,8,18]]},"reference":[{"key":"pmbae9688bib1","doi-asserted-by":"publisher","first-page":"10468","DOI":"10.1021\/acsomega.4c10763","type":"journal-article","article-title":"Attention-enhanced multi-task deep learning model for classification and segmentation of esophageal lesions","volume":"10","author":"Aftab","year":"2025","journal-title":"ACS Omega"},{"key":"pmbae9688bib2","article-title":"Multi-target and multi-stage liver lesion segmentation and detection in multi-phase computed tomography scans","author":"Al-Battal","year":"2024","type":"preprint"},{"key":"pmbae9688bib3","doi-asserted-by":"publisher","first-page":"1324","DOI":"10.1016\/j.adro.2020.06.012","type":"journal-article","article-title":"An automated workflow to improve efficiency in radiation therapy treatment planning by prioritizing organs at risk","volume":"5","author":"Aliotta","year":"2020","journal-title":"Adv. Rad. Oncol."},{"key":"pmbae9688bib4","doi-asserted-by":"publisher","first-page":"3691","DOI":"10.3390\/cancers17223691","type":"journal-article","article-title":"Prompt-driven multimodal segmentation with dynamic fusion for adaptive and robust medical imaging with applications to cancer diagnosis","volume":"17","author":"Alsaedi","year":"2025","journal-title":"Cancers"},{"key":"pmbae9688bib5","article-title":"Closed-loop prompt optimization for enhanced accuracy in extractive question answering","author":"Arild Fontes","year":"2025","type":"other"},{"key":"pmbae9688bib6","doi-asserted-by":"publisher","first-page":"7021","DOI":"10.3390\/curroncol29100552","type":"journal-article","article-title":"Dose\u2013volume constraints for organs at risk in radiotherapy (corsair): an \u2018all-in-one\u2019 multicenter\u2013multidisciplinary practical summary","volume":"29","author":"Bisello","year":"2022","journal-title":"Curr. Oncol."},{"key":"pmbae9688bib7","doi-asserted-by":"publisher","DOI":"10.1177\/15330338211034284","type":"journal-article","article-title":"Clinical target volume auto-segmentation of esophageal cancer for radiotherapy after radical surgery based on deep learning","volume":"20","author":"Cao","year":"2021","journal-title":"Technol. Cancer Res. Treatm."},{"key":"pmbae9688bib8","doi-asserted-by":"publisher","first-page":"102","DOI":"10.1186\/s13014-020-01691-4","type":"journal-article","article-title":"Interobserver variability in target volume delineation in definitive radiotherapy for thoracic esophageal cancer: a multi-center study from china","volume":"16","author":"Chang","year":"2021","journal-title":"Radiat. Oncol."},{"key":"pmbae9688bib9","article-title":"Interactive radiotherapy target delineation with 3d-fused context propagation","author":"Chao","year":"2020","type":"preprint"},{"key":"pmbae9688bib10","first-page":"pp 221","type":"book","article-title":"Head and neck tumor segmentation with 3d unet and survival prediction with multiple instance neural network","author":"Chen","year":"2022"},{"key":"pmbae9688bib11","first-page":"pp 3570","type":"conference-proceedings","article-title":"Autoprosam: automated prompting sam for 3d multi-organ segmentation","author":"Chengyin","year":"2025"},{"key":"pmbae9688bib12","first-page":"pp 424","type":"conference-proceedings","article-title":"3d u-net: learning dense volumetric segmentation from sparse annotation","author":"Cicek","year":"2016"},{"key":"pmbae9688bib13","first-page":"pp 105","type":"conference-proceedings","article-title":"Openprompt: An open-source framework for prompt-learning","author":"Ding","year":"2022"},{"key":"pmbae9688bib14","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2022.817413","type":"journal-article","article-title":"Inter-observer and intra-observer variability in gross tumor volume delineation of primary esophageal carcinomas based on different combinations of diagnostic multimodal images","volume":"12","author":"Fengxiang","year":"2022","journal-title":"Front. Oncol."},{"key":"pmbae9688bib15","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2024.103324","type":"journal-article","article-title":"3DSAM-adapter: Holistic adaptation of sam from 2d\u20133d for promptable tumor segmentation","volume":"98","author":"Gong","year":"2024","journal-title":"Med. Image Anal."},{"key":"pmbae9688bib16","first-page":"pp 219","type":"book","article-title":"Practical clinical workflows for online and offline adaptive radiation therapy","volume":"vol 29","author":"Green","year":"2019"},{"key":"pmbae9688bib17","doi-asserted-by":"publisher","DOI":"10.1117\/12.3005879","type":"journal-article","article-title":"Clicksam: fine-tuning segment anything model using click prompts for ultrasound image segmentation","volume":"12932","author":"Guo","year":"2024","journal-title":"Proc. SPIE"},{"key":"pmbae9688bib18","doi-asserted-by":"publisher","first-page":"203","DOI":"10.1038\/s41592-020-01008-z","type":"journal-article","article-title":"nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation","volume":"18","author":"Isensee","year":"2021","journal-title":"Nat. Methods"},{"key":"pmbae9688bib19","first-page":"pp 709","type":"conference-proceedings","article-title":"Visual prompt tuning","author":"Jia","year":"2022"},{"key":"pmbae9688bib20","first-page":"pp 435","type":"conference-proceedings","article-title":"Radsam: segmenting 3d radiological images with a 2d promptable model","author":"Khlaut","year":"2025"},{"key":"pmbae9688bib21","first-page":"pp 3","type":"conference-proceedings","article-title":"3D-SAutoMed: automatic segment anything model for 3d medical image segmentation from local-global perspective","author":"Liang","year":"2024"},{"key":"pmbae9688bib22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00941","type":"conference-proceedings","article-title":"Iteratively-refined interactive 3d medical image segmentation with multi-agent reinforcement learning","author":"Liao","year":"2024"},{"key":"pmbae9688bib23","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3560815","type":"journal-article","article-title":"Pre-train, prompt and predict: a systematic survey of prompting methods in natural language processing","volume":"195","author":"Liu","year":"2023","journal-title":"ACM Comput. Surv."},{"key":"pmbae9688bib24","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2023.102918","type":"journal-article","article-title":"Segment anything model for medical image analysis: an experimental study","volume":"89","author":"Mazurowski","year":"2023","journal-title":"Med. Image Anal."},{"key":"pmbae9688bib25","first-page":"pp 234","type":"conference-proceedings","article-title":"U-net: convolutional networks for biomedical image segmentation","author":"Ronneberger","year":"2015"},{"key":"pmbae9688bib26","doi-asserted-by":"publisher","first-page":"90","DOI":"10.1016\/j.compmedimag.2018.03.001","type":"journal-article","article-title":"An application of cascaded 3d fully convolutional networks for medical image segmentation","volume":"66","author":"Roth","year":"2018","journal-title":"Comput. Med. Imaging Graph."},{"key":"pmbae9688bib27","article-title":"Boundary-aware network for abdominal multi-organ segmentation","author":"Shishuai","year":"2022","type":"preprint"},{"key":"pmbae9688bib28","doi-asserted-by":"publisher","first-page":"2504","DOI":"10.3390\/diagnostics13152504","type":"journal-article","article-title":"A coarse-to-fine fusion network for small liver tumor detection and segmentation: a real-world study","volume":"13","author":"Shu","year":"2023","journal-title":"Diagnostics"},{"key":"pmbae9688bib29","doi-asserted-by":"publisher","DOI":"10.1002\/mp.17818","type":"journal-article","article-title":"A CNN-transformer-based hybrid U-shape model with long-range relay for esophagus 3D CT image gross tumor volume segmentation","volume":"52","author":"Songli","year":"2025","journal-title":"Med. Phys."},{"key":"pmbae9688bib30","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-025-20957-z","type":"journal-article","article-title":"Meta transfer learning for brain tumor segmentation using nnUNet in meningioma and metastasis cases","volume":"15","author":"Tabassum","year":"2025","journal-title":"Sci. Rep."},{"key":"pmbae9688bib31","article-title":"Sam-med3d: towards general-purpose segmentation models for volumetric medical images","author":"Wang","year":"2024","type":"preprint"},{"key":"pmbae9688bib32","doi-asserted-by":"publisher","first-page":"3349","DOI":"10.1109\/TPAMI.2020.2983686","type":"journal-article","article-title":"Deep high-resolution representation learning for visual recognition","volume":"43","author":"Wang","year":"2020","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"pmbae9688bib33","article-title":"MaskSAM: Towards auto-prompt SAM with mask classification for volumetric medical image segmentation","author":"Xie","year":"2024","type":"preprint"},{"key":"pmbae9688bib34","doi-asserted-by":"publisher","first-page":"1991","DOI":"10.1016\/S0140-6736(24)02226-8","type":"journal-article","article-title":"Oesophageal cancer","volume":"404","author":"Yang","year":"2024","journal-title":"Lancet"},{"key":"pmbae9688bib35","doi-asserted-by":"publisher","first-page":"9140","DOI":"10.1088\/1361-6560\/aa94ba","type":"journal-article","article-title":"Atlas ranking and selection for automatic segmentation of the esophagus from ct scans","volume":"62","author":"Yang","year":"2017","journal-title":"Phys. Med. Biol."},{"key":"pmbae9688bib36","doi-asserted-by":"publisher","first-page":"9186","DOI":"10.1038\/s41467-024-53387-y","type":"journal-article","article-title":"Llm-driven multimodal target volume contouring in radiation oncology","volume":"15","author":"Yujin","year":"2024","journal-title":"Nat. Commun."},{"key":"pmbae9688bib37","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2024.108238","type":"journal-article","article-title":"Segment anything model for medical image segmentation: current applications and future directions","volume":"171","author":"Zhang","year":"2024","journal-title":"Comput. Biol. Med."},{"key":"pmbae9688bib38","doi-asserted-by":"publisher","first-page":"3768","DOI":"10.1109\/TCSVT.2025.3620396","type":"journal-article","article-title":"Medsam-u: uncertainty-guided auto multi-prompt adaptation for reliable medsam","volume":"36","author":"Zhou","year":"2025","journal-title":"IEEE Trans. Circuits Syst. Video Technol."},{"key":"pmbae9688bib39","doi-asserted-by":"publisher","first-page":"141","DOI":"10.1120\/jacmp.v15i1.4499","type":"journal-article","article-title":"Interactive contour delineation and refinement in treatment planning of image-guided radiation therapy","volume":"15","author":"Zhou","year":"2014","journal-title":"J. Appl. Clin. Med. Phys."},{"key":"pmbae9688bib40","doi-asserted-by":"publisher","DOI":"10.1109\/3DV.2018.00083","type":"conference-proceedings","article-title":"A 3d coarse-to-fine framework for volumetric medical image segmentation","author":"Zhu","year":"2021"}],"container-title":["Physics in Medicine &amp; Biology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae9688","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae9688\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae9688","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae9688\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae9688\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae9688\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae9688\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae9688\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T08:28:33Z","timestamp":1787041713000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae9688"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,18]]},"references-count":40,"journal-issue":{"issue":"16","published-online":{"date-parts":[[2026,8,18]]},"published-print":{"date-parts":[[2026,8,28]]}},"URL":"https:\/\/doi.org\/10.1088\/1361-6560\/ae9688","relation":{},"ISSN":["0031-9155","1361-6560"],"issn-type":[{"value":"0031-9155","type":"print"},{"value":"1361-6560","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,18]]},"assertion":[{"value":"Dynamic iterative coarse-to-fine prompt learning based on SAM for precise esophageal cancer gross target volume segmentation","name":"article_title","label":"Article Title"},{"value":"Physics in Medicine & Biology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2026 Institute of Physics and Engineering in Medicine. All rights, including for text and data mining, AI training, and similar technologies, are reserved.","name":"copyright_information","label":"Copyright Information"},{"value":"2026-03-08","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-08-06","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-08-18","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}