{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T14:03:55Z","timestamp":1791295435821,"version":"4.1.0"},"reference-count":62,"publisher":"Wiley","issue":"3","license":[{"start":{"date-parts":[[2024,2,6]],"date-time":"2024-02-06T00:00:00Z","timestamp":1707177600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2024,3]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Efficient and accurate delineation of organs at risk (OARs) is a critical procedure for treatment planning and dose evaluation. Deep learning\u2010based auto\u2010segmentation of OARs has shown promising results and is increasingly being used in radiation therapy. However, existing deep learning\u2010based auto\u2010segmentation approaches face two challenges in clinical practice: generalizability and human\u2010AI interaction. A generalizable and promptable auto\u2010segmentation model, which segments OARs of multiple disease sites simultaneously and supports on\u2010the\u2010fly human\u2010AI interaction, can significantly enhance the efficiency of radiation therapy treatment planning.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>Meta's segment anything model (SAM) was proposed as a generalizable and promptable model for next\u2010generation natural image segmentation. We further evaluated the performance of SAM in radiotherapy segmentation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Computed tomography (CT) images of clinical cases from four disease sites at our institute were collected: prostate, lung, gastrointestinal, and head &amp; neck. For each case, we selected the OARs important in radiotherapy treatment planning. We then compared both the Dice coefficients and Jaccard indices derived from three distinct methods: manual delineation (ground truth), automatic segmentation using SAM's \u2019segment anything\u2019 mode, and automatic segmentation using SAM's \u2018box prompt\u2019 mode that implements manual interaction via live prompts during segmentation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Our results indicate that SAM's segment anything mode can achieve clinically acceptable segmentation results in most OARs with Dice scores higher than 0.7. SAM's box prompt mode further improves Dice scores by 0.1\u223c0.5. Similar results were observed for Jaccard indices. The results show that SAM performs better for prostate and lung, but worse for gastrointestinal and head &amp; neck. When considering the size of organs and the distinctiveness of their boundaries, SAM shows better performance for large organs with distinct boundaries, such as lung and liver, and worse for smaller organs with less distinct boundaries, like parotid and cochlea.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Our results demonstrate SAM's robust generalizability with consistent accuracy in automatic segmentation for radiotherapy. Furthermore, the advanced box\u2010prompt method enables the users to augment auto\u2010segmentation interactively and dynamically, leading to patient\u2010specific auto\u2010segmentation in radiation therapy. SAM's generalizability across different disease sites and different modalities makes it feasible to develop a generic auto\u2010segmentation model in radiotherapy.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.16965","type":"journal-article","created":{"date-parts":[[2024,2,7]],"date-time":"2024-02-07T03:48:17Z","timestamp":1707277697000},"page":"2187-2199","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":21,"title":["Technical note: Generalizable and promptable artificial intelligence model to augment clinical delineation in radiation oncology"],"prefix":"10.1002","volume":"51","author":[{"given":"Lian","family":"Zhang","sequence":"first","affiliation":[{"name":"Department of Radiation Oncology Mayo Clinic  Phoenix Arizona USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhengliang","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Computing University of Georgia  Athens Georgia USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lu","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering University of Texas at Arlington  Arlington Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zihao","family":"Wu","sequence":"additional","affiliation":[{"name":"School of Computing University of Georgia  Athens Georgia USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaowei","family":"Yu","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering University of Texas at Arlington  Arlington Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jason","family":"Holmes","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Mayo Clinic  Phoenix Arizona USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongying","family":"Feng","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Mayo Clinic  Phoenix Arizona USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Haixing","family":"Dai","sequence":"additional","affiliation":[{"name":"School of Computing University of Georgia  Athens Georgia USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiang","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Radiology Massachusetts General Hospital and Harvard Medical School  Boston Massachusetts USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Quanzheng","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Radiology Massachusetts General Hospital and Harvard Medical School  Boston Massachusetts USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"William W.","family":"Wong","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Mayo Clinic  Phoenix Arizona USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sujay A.","family":"Vora","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Mayo Clinic  Phoenix Arizona USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dajiang","family":"Zhu","sequence":"additional","affiliation":[{"name":"School of Computing University of Georgia  Athens Georgia USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tianming","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Computing University of Georgia  Athens Georgia USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Mayo Clinic  Phoenix Arizona USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2024,2,6]]},"reference":[{"key":"e_1_2_9_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.metrad.2023.100005"},{"key":"e_1_2_9_3_1","unstructured":"LiuY HanT MaS et\u00a0al.Summary of chatgpt\/gpt\u20104 research and perspective towards the future of large language models. arXiv preprint arXiv:230401852.2023."},{"key":"e_1_2_9_4_1","unstructured":"BubeckS ChandrasekaranV EldanR et\u00a0al.Sparks of artificial general intelligence: early experiments with gpt\u20104. arXiv preprint arXiv:230312712.2023."},{"key":"e_1_2_9_5_1","unstructured":"AnilR DaiAM FiratO et\u00a0al.Palm 2 technical report. arXiv preprint arXiv:230510403.2023."},{"key":"e_1_2_9_6_1","unstructured":"HolmesJ LiuZ ZhangL et\u00a0al.Evaluating large language models on a highly\u2010specialized topic radiation oncology physics. arXiv preprint arXiv:230401938.2023."},{"key":"e_1_2_9_7_1","unstructured":"LiuZ ZhongA LiY et\u00a0al.Radiology\u2010GPT: A Large Language Model for Radiology. arXiv preprint arXiv:230608666.2023."},{"key":"e_1_2_9_8_1","unstructured":"WuZ ZhangL CaoC et\u00a0al.Exploring the trade\u2010offs: Unified large language models vs local fine\u2010tuned models for highly\u2010specific radiology nli task. arXiv preprint arXiv:230409138.2023."},{"key":"e_1_2_9_9_1","unstructured":"LiuZ YuX ZhangL et\u00a0al.Deid\u2010gpt: Zero\u2010shot medical text de\u2010identification by gpt\u20104. arXiv preprint arXiv:230311032.2023."},{"key":"e_1_2_9_10_1","unstructured":"DaiH LiuZ LiaoW et\u00a0al.Chataug: Leveraging chatgpt for text data augmentation. arXiv preprint arXiv:230213007.2023."},{"key":"e_1_2_9_11_1","unstructured":"LiX ZhangL WuZ et\u00a0al.Artificial General Intelligence for Medical Imaging. arXiv preprint arXiv:230605480.2023."},{"key":"e_1_2_9_12_1","unstructured":"HuangY YangX LiuL et\u00a0al.Segment anything model for medical images?arXiv preprint arXiv:230414660.2023."},{"key":"e_1_2_9_13_1","unstructured":"KirillovA MintunE RaviN et\u00a0al.Segment anything. arXiv preprint arXiv:230402643.2023."},{"issue":"7","key":"e_1_2_9_14_1","first-page":"3523","article-title":"Image segmentation using deep learning: a survey","volume":"44","author":"Minaee S","year":"2021","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"issue":"6","key":"e_1_2_9_15_1","doi-asserted-by":"crossref","first-page":"1129","DOI":"10.1002\/cncr.21324","article-title":"The role of radiotherapy in cancer treatment: estimating optimal utilization from a review of evidence\u2010based clinical guidelines","volume":"104","author":"Delaney G","year":"2005","journal-title":"Cancer: Interd Int J Am Cancer Soc"},{"key":"e_1_2_9_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2009.09.062"},{"key":"e_1_2_9_17_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.16548"},{"key":"e_1_2_9_18_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2007.11.016"},{"key":"e_1_2_9_19_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.prro.2014.08.002"},{"key":"e_1_2_9_20_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.prro.2013.12.001"},{"key":"e_1_2_9_21_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2015.11.002"},{"key":"e_1_2_9_22_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.13049"},{"key":"e_1_2_9_23_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12677"},{"key":"e_1_2_9_24_1","doi-asserted-by":"publisher","DOI":"10.5306\/wjco.v5.i4.568"},{"key":"e_1_2_9_25_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.12459"},{"key":"e_1_2_9_26_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4801899"},{"key":"e_1_2_9_27_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2015.01.017"},{"key":"e_1_2_9_28_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13809"},{"key":"e_1_2_9_29_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4774363"},{"key":"e_1_2_9_30_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4711909"},{"key":"e_1_2_9_31_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2019.03.004"},{"key":"e_1_2_9_32_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2019.09.022"},{"key":"e_1_2_9_33_1","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2019.00239"},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2021.04.019"},{"key":"e_1_2_9_35_1","doi-asserted-by":"publisher","DOI":"10.21037\/qims-21-168"},{"key":"e_1_2_9_36_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13014-019-1392-z"},{"key":"e_1_2_9_37_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.phro.2019.11.006"},{"key":"e_1_2_9_38_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13300"},{"key":"e_1_2_9_39_1","doi-asserted-by":"publisher","DOI":"10.1259\/bjr.20190001"},{"key":"e_1_2_9_40_1","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.27908"},{"key":"e_1_2_9_41_1","doi-asserted-by":"crossref","unstructured":"LiW LamS LiT et\u00a0al.Multi\u2010institutional investigation of model generalizability for virtual contrast\u2010enhanced mri synthesis.Paper presented at: International Conference on Medical Image Computing and Computer\u2010Assisted Intervention 2022.","DOI":"10.1007\/978-3-031-16449-1_73"},{"key":"e_1_2_9_42_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2020.2973595"},{"key":"e_1_2_9_43_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-42478-1"},{"key":"e_1_2_9_44_1","unstructured":"PutzF GrigoJ WeissmannT et\u00a0al.The Segment Anything foundation model achieves favorable brain tumor autosegmentation accuracy on MRI to support radiotherapy treatment planning. arXiv preprint arXiv:230407875.2023."},{"key":"e_1_2_9_45_1","article-title":"Segment anything model for medical images?","author":"Huang Y","year":"2023","journal-title":"Med Image Anal"},{"key":"e_1_2_9_46_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2023.102918"},{"key":"e_1_2_9_47_1","unstructured":"MaJ WangB Segment anything in medical images. arXiv preprint arXiv:230412306.2023."},{"key":"e_1_2_9_48_1","unstructured":"HeS BaoR LiJ GrantPE OuY Accuracy of segment\u2010anything model (sam) in medical image segmentation tasks. arXiv preprint arXiv:230409324.2023."},{"key":"e_1_2_9_49_1","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics13111947"},{"key":"e_1_2_9_50_1","unstructured":"ZhangY JiaoR How Segment Anything Model (SAM) Boost Medical Image Segmentation?arXiv preprint arXiv:230503678.2023."},{"key":"e_1_2_9_51_1","unstructured":"WuJ FuR FangH et\u00a0al.Medical sam adapter: Adapting segment anything model for medical image segmentation. arXiv preprint arXiv:230412620.2023."},{"key":"e_1_2_9_52_1","doi-asserted-by":"crossref","unstructured":"ZhangY ZhouT WangS LiangP ZhangY ChenDZ Input augmentation with sam: boosting medical image segmentation with segmentation foundation model.Paper presented at: International Conference on Medical Image Computing and Computer\u2010Assisted Intervention 2023.","DOI":"10.1007\/978-3-031-47401-9_13"},{"key":"e_1_2_9_53_1","unstructured":"LiuY ZhangJ SheZ KheradmandA ArmandM Samm (segment any medical model): A 3d slicer integration to sam. arXiv preprint arXiv:230405622.2023."},{"key":"e_1_2_9_54_1","unstructured":"RoyS WaldT KoehlerG et\u00a0al.Sam. md: Zero\u2010shot medical image segmentation capabilities of the segment anything model. arXiv preprint arXiv:230405396.2023."},{"key":"e_1_2_9_55_1","unstructured":"GaoY XiaW HuD DeSAMGX :Decoupling Segment Anything Model for Generalizable Medical Image Segmentation. arXiv preprint arXiv:230600499.2023."},{"key":"e_1_2_9_56_1","unstructured":"ZhangR JiangZ GuoZ et\u00a0al.Personalize segment anything model with one shot. arXiv preprint arXiv:230503048.2023."},{"key":"e_1_2_9_57_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1754-9485.2010.02192.x"},{"key":"e_1_2_9_58_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2016.09.009"},{"key":"e_1_2_9_59_1","unstructured":"IsenseeF PetersenJ KohlSA J\u00e4gerPF Maier\u2010HeinKH nnu\u2010net: Breaking the spell on successful medical image segmentation. arXiv preprint arXiv:190408128.2019;1(1\u20108):2."},{"key":"e_1_2_9_60_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-020-01008-z"},{"key":"e_1_2_9_61_1","doi-asserted-by":"publisher","DOI":"10.1111\/1754-9485.13286"},{"key":"e_1_2_9_62_1","unstructured":"DaiH MaC LiuZ et\u00a0al.Samaug: Point prompt augmentation for segment anything model. arXiv preprint arXiv:230701187.2023."},{"key":"e_1_2_9_63_1","unstructured":"Raji\u010dF KeL TaiY\u2010W TangC\u2010K DanelljanM YuF Segment Anything Meets Point Tracking. arXiv preprint arXiv:230701197.2023."}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.16965","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,3,12]],"date-time":"2024-03-12T11:01:54Z","timestamp":1710241314000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.16965"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,2,6]]},"references-count":62,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2024,3]]}},"alternative-id":["10.1002\/mp.16965"],"URL":"https:\/\/doi.org\/10.1002\/mp.16965","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,2,6]]},"assertion":[{"value":"2023-08-23","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-01-14","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-02-06","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}