{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T07:11:22Z","timestamp":1780384282287,"version":"3.54.1"},"reference-count":45,"publisher":"IOP Publishing","issue":"10","license":[{"start":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T00:00:00Z","timestamp":1780358400000},"content-version":"vor","delay-in-days":5,"URL":"https:\/\/publishingsupport.iopscience.iop.org\/iop-standard\/v1"},{"start":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T00:00:00Z","timestamp":1780358400000},"content-version":"tdm","delay-in-days":5,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"},{"start":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T00:00:00Z","timestamp":1780358400000},"content-version":"vor","delay-in-days":5,"URL":"https:\/\/publishingsupport.iopscience.iop.org\/questions\/alternative-author-rights-policies\/"},{"start":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T00:00:00Z","timestamp":1780358400000},"content-version":"vor","delay-in-days":5,"URL":"https:\/\/publishingsupport.iopscience.iop.org\/questions\/alternative-author-rights-policies\/"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Phys. Med. Biol."],"published-print":{"date-parts":[[2026,5,28]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    <jats:italic>Objective.<\/jats:italic>\n                    Vision foundation models (FMs) have shown strong potential for adapting to diverse downstream tasks without specialized fine-tuning. However, their performance in head and neck cancer (HNC) tumor segmentation remains insufficiently explored. This work provides a comprehensive evaluation of zero-shot and few-shot performance of five prominent FMs for HNC tumor segmentation.\n                    <jats:italic>Approach.<\/jats:italic>\n                    The FMs tested include SAM, MedSAM, nnInteractive, UniverSeg, and SegGPT. The evaluation of SAM, MedSAM, and nnInteractive utilized a bounding box prompting strategy. The influence of prompt precision is assessed by systematically enlarging the prompt bounding box until it reaches twice the size of the tightest bounding box enclosing the region of interest. For UniverSeg and SegGPT, the prompt consisted of example image-label pairs, and the effect of increasing the size of this support set is tested. Two-factor repeated measures ANOVAs were used for statistical analysis, confirming that both the FM architecture and the prompting strategy have a statistically significant effect on segmentation performance for both primary tumors (GTVp) and metastatic lymph nodes (GTVn).\n                    <jats:italic>Main results.<\/jats:italic>\n                    For SAM, MedSAM, and nnInteractive larger bounding boxes lead to notable performance degradation. Moreover, MedSAM outperformed SAM and nnInteractive for GTVp segmentation, whereas SAM provided slightly better results than the competitors for GTVn segmentation. For UniverSeg, performance improved as more image-label pairs are added to the support set, while SegGPT remained unaffected by this factor. Overall, UniverSeg and SegGPT demonstrate limited few-shot capability, with low mean dice scores across all scenarios. Comparisons against supervised convolutional neural network models trained specifically for HNC segmentation showed that conventional architectures remain more stable and clinically reliable.\n                    <jats:italic>Significance.<\/jats:italic>\n                    These findings highlight the ongoing challenges faced by current FMs in achieving reliable zero- and few-shot medical segmentation. The results emphasize the critical need for improved medical-domain adaptation and enhanced prompt robustness to advance the clinical utility of FMs in oncology.\n                  <\/jats:p>","DOI":"10.1088\/1361-6560\/ae6af8","type":"journal-article","created":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T22:54:31Z","timestamp":1778280871000},"page":"105026","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Is segmentation solved? An evaluation of vision foundation models for head and neck tumor segmentation"],"prefix":"10.1088","volume":"71","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9379-8151","authenticated-orcid":true,"given":"Maria","family":"Baldeon-Calisto","sequence":"first","affiliation":[{"name":"Wingate University"},{"name":"Universidad San Francisco de Quito USFQ"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3166-745X","authenticated-orcid":true,"given":"Noel","family":"Perez-P\u00e9rez","sequence":"additional","affiliation":[{"name":"Universidad San Francisco de Quito USFQ"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1571-5829","authenticated-orcid":true,"given":"Cesar","family":"Guevara","sequence":"additional","affiliation":[{"name":"CUNEF Universidad"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2026,6,2]]},"reference":[{"key":"pmbae6af8bib1","doi-asserted-by":"publisher","first-page":"22","DOI":"10.1117\/12.3045573","type":"journal-article","article-title":"Segment anything model (SAM) for metastatic tumor segmentation: evaluating its impact on chemotherapy response prediction","volume":"13296","author":"Abdoli","year":"2025","journal-title":"Biophotonics and Immune Responses XX"},{"key":"pmbae6af8bib2","doi-asserted-by":"publisher","DOI":"10.1016\/j.compmedimag.2024.102473","type":"journal-article","article-title":"A review of the segment anything model (sam) for medical image analysis: accomplishments and perspectives","volume":"119","author":"Ali","year":"2025","journal-title":"Comput. Med. Imaging Graph."},{"key":"pmbae6af8bib3","first-page":"375","type":"conference-proceedings","article-title":"Can foundation models really segment tumors? A benchmarking odyssey in lung CT imaging","author":"Ayllon","year":"2025"},{"key":"pmbae6af8bib4","doi-asserted-by":"publisher","first-page":"42","DOI":"10.3390\/medsci11020042","type":"journal-article","article-title":"Epidemiology, risk factors, and prevention of head and neck squamous cell carcinoma","volume":"11","author":"Barsouk","year":"2023","journal-title":"Med. Sci."},{"key":"pmbae6af8bib5","first-page":"1","author":"Berrezueta","year":"2025","type":"book"},{"key":"pmbae6af8bib6","first-page":"21438","author":"Butoi","year":"2023","type":"book"},{"key":"pmbae6af8bib7","first-page":"971","author":"Calisto","year":"2022","type":"book"},{"key":"pmbae6af8bib8","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2020.614008","type":"journal-article","article-title":"Trimodality PET\/CT\/MRI and radiotherapy: a mini-review","volume":"10","author":"Decazes","year":"2021","journal-title":"Front. Oncol."},{"key":"pmbae6af8bib9","doi-asserted-by":"publisher","first-page":"232","DOI":"10.1016\/j.carj.2013.07.004","type":"journal-article","article-title":"CRISPS: a pictorial essay of an acronym to interpreting metastatic head and neck lymphadenopathy","volume":"65","author":"Dmytriw","year":"2014","journal-title":"Can. Assoc. Radiol. J."},{"key":"pmbae6af8bib10","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2026.1756011","type":"journal-article","article-title":"Practical application of SAM for breast nodules segmentation","volume":"16","author":"Fan","year":"2026","journal-title":"Front. Oncol."},{"key":"pmbae6af8bib11","first-page":"509","author":"Gao","year":"2024","type":"book"},{"key":"pmbae6af8bib12","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab440d","type":"journal-article","article-title":"Gross tumor volume segmentation for head and neck cancer radiotherapy using deep dense multi-modality network","volume":"64","author":"Guo","year":"2019","journal-title":"Phys. Med. Biol."},{"key":"pmbae6af8bib13","first-page":"4869","type":"conference-proceedings","article-title":"Medverse: A universal model for full-resolution 3d medical image segmentation, transformation and enhancement","volume":"40","author":"Hu","year":"2026"},{"key":"pmbae6af8bib14","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2023.103061","type":"journal-article","article-title":"Segment anything model for medical images?","volume":"92","author":"Huang","year":"2024","journal-title":"Med. Image Anal."},{"key":"pmbae6af8bib15","article-title":"nninteractive: redefining 3d promptable segmentation","author":"Isensee","year":"2025","type":"preprint"},{"key":"pmbae6af8bib16","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":"pmbae6af8bib17","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2024.103202","type":"journal-article","article-title":"USFM: a universal ultrasound foundation model generalized to tasks and organs towards label efficient image analysis","volume":"96","author":"Jiao","year":"2024","journal-title":"Med. Image Anal."},{"key":"pmbae6af8bib18","doi-asserted-by":"publisher","first-page":"283","DOI":"10.1109\/RBME.2025.3531360","type":"journal-article","article-title":"A comprehensive survey of foundation models in medicine","volume":"19","author":"Khan","year":"2026","journal-title":"IEEE Rev. Biomed. Eng."},{"key":"pmbae6af8bib19","first-page":"4015","author":"Kirillov","year":"2023","type":"book"},{"key":"pmbae6af8bib20","doi-asserted-by":"publisher","first-page":"7757","DOI":"10.1002\/mp.15290","type":"journal-article","article-title":"Cascaded deep learning-based auto-segmentation for head and neck cancer patients: organs at risk on T2-weighted magnetic resonance imaging","volume":"48","author":"Korte","year":"2021","journal-title":"Med. Phys."},{"key":"pmbae6af8bib21","author":"Lee","year":"2024","type":"book"},{"key":"pmbae6af8bib22","doi-asserted-by":"publisher","first-page":"2096","DOI":"10.1002\/mp.16703","type":"journal-article","article-title":"SwinCross: cross-modal Swin transformer for head-and-neck tumor segmentation in PET\/CT images","volume":"51","author":"Li","year":"2024a","journal-title":"Med. Phys."},{"key":"pmbae6af8bib23","doi-asserted-by":"publisher","first-page":"759","DOI":"10.1117\/12.3006809","type":"journal-article","volume":"12927","author":"Li","year":"2024b","journal-title":"Polyp-sam: Transfer Sam for Polyp Segmentation Proc. SPIE"},{"key":"pmbae6af8bib24","author":"Liu","year":"2024","type":"book"},{"key":"pmbae6af8bib25","doi-asserted-by":"publisher","first-page":"654","DOI":"10.1038\/s41467-024-44824-z","type":"journal-article","article-title":"Segment anything in medical images","volume":"15","author":"Ma","year":"2024","journal-title":"Nat. Commun."},{"key":"pmbae6af8bib26","doi-asserted-by":"publisher","first-page":"110905","DOI":"10.1109\/ACCESS.2024.3439439","type":"journal-article","article-title":"Weighted fusion transformer for dual PET\/CT head and neck tumor segmentation","volume":"12","author":"Mahdi","year":"2024","journal-title":"IEEE Access"},{"key":"pmbae6af8bib27","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":"pmbae6af8bib28","first-page":"136","author":"Moradi","year":"2024","type":"book"},{"key":"pmbae6af8bib29","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.102336","type":"journal-article","article-title":"Head and neck tumor segmentation in PET\/CT: the HECKTOR challenge","volume":"77","author":"Oreiller","year":"2022","journal-title":"Med. Image Anal."},{"key":"pmbae6af8bib30","first-page":"187","author":"Paranjape","year":"2024","type":"book"},{"key":"pmbae6af8bib31","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s00066-023-02183-6","type":"journal-article","article-title":"The Segment Anything foundation model achieves favorable brain tumor auto-segmentation accuracy in MRI to support radiotherapy treatment planning","volume":"200","author":"Putz","year":"2024","journal-title":"Strahlentherapie Und Onkologie"},{"key":"pmbae6af8bib32","doi-asserted-by":"publisher","first-page":"1947","DOI":"10.3390\/diagnostics13111947","type":"journal-article","article-title":"Generalist vision foundation models for medical imaging: a case study of segment anything model on zero-shot medical segmentation","volume":"13","author":"Shi","year":"2023","journal-title":"Diagnostics"},{"key":"pmbae6af8bib33","first-page":"3104","type":"conference-proceedings","article-title":"The overlooked matters: revisiting background, prototype, and activation in few-shot medical image segmentation","author":"Shu","year":"2025"},{"key":"pmbae6af8bib34","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ac530e","type":"journal-article","article-title":"U-net architecture with embedded Inception-ResNet-v2 image encoding modules for automatic segmentation of organs-at-risk in head and neck cancer radiation therapy based on computed tomography scans","volume":"67","author":"Siciarz","year":"2022","journal-title":"Phys. Med. Biol."},{"key":"pmbae6af8bib35","author":"Su","year":"2024","type":"book"},{"key":"pmbae6af8bib36","doi-asserted-by":"publisher","first-page":"103","DOI":"10.1016\/j.neucom.2020.12.085","type":"journal-article","article-title":"DA-DSUnet: dual attention-based dense SU-net for automatic head-and-neck tumor segmentation in MRI images","volume":"435","author":"Tang","year":"2021","journal-title":"Neurocomputing"},{"key":"pmbae6af8bib37","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.11199558","type":"dataset","article-title":"Training Dataset for HNTSMRG 2024 Challenge","author":"Wahid","year":"2024","unstructured":"Wahid K Dede C Naser M Fuller C 2024 Training Dataset for HNTSMRG 2024 Challenge 10.5281\/zenodo.11199558"},{"key":"pmbae6af8bib38","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/acd29f","type":"journal-article","article-title":"DMCT-Net: dual modules convolution transformer network for head and neck tumor segmentation in PET\/CT","volume":"68","author":"Wang","year":"2023","journal-title":"Phys. Med. Biol."},{"key":"pmbae6af8bib39","first-page":"6830","author":"Wang","year":"2023a","type":"book"},{"key":"pmbae6af8bib40","first-page":"1130","author":"Wang","year":"2023b","type":"book"},{"key":"pmbae6af8bib41","doi-asserted-by":"publisher","DOI":"10.3389\/fgene.2021.624820","type":"journal-article","article-title":"Deep learning in head and neck tumor multiomics diagnosis and analysis: review of the literature","volume":"12","author":"Wang","year":"2021","journal-title":"Front. Genet."},{"key":"pmbae6af8bib42","doi-asserted-by":"publisher","first-page":"144591","DOI":"10.1109\/ACCESS.2019.2944958","type":"journal-article","article-title":"Organ at risk segmentation in head and neck CT images using a two-stage segmentation framework based on 3D U-Net","volume":"7","author":"Wang","year":"2019","journal-title":"IEEE Access"},{"key":"pmbae6af8bib43","first-page":"92","author":"Xie","year":"2021","type":"book"},{"key":"pmbae6af8bib44","doi-asserted-by":"publisher","first-page":"11757","DOI":"10.1038\/s41598-026-40448-z","type":"conference-proceedings","article-title":"Edsamix: a training-free model merging approach for medical image segmentation","author":"Yang","year":"2026"},{"key":"pmbae6af8bib45","author":"Zhang","year":"2023","type":"book"}],"container-title":["Physics in Medicine &amp; Biology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6af8","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6af8\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6af8","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6af8\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6af8\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6af8\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6af8\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6af8\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T06:55:42Z","timestamp":1780383342000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6af8"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,28]]},"references-count":45,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2026,6,2]]},"published-print":{"date-parts":[[2026,5,28]]}},"URL":"https:\/\/doi.org\/10.1088\/1361-6560\/ae6af8","relation":{},"ISSN":["0031-9155","1361-6560"],"issn-type":[{"value":"0031-9155","type":"print"},{"value":"1361-6560","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,28]]},"assertion":[{"value":"Is segmentation solved? An evaluation of vision foundation models for head and neck tumor 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":"2025-12-15","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-05-08","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-06-02","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}