[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-355970-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-355970-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","promptable-segmentation-with-region-exploration-enables-minimal-effort-expert-level-prostate-cancer-delineation","Promptable segmentation with region exploration enables minimal-effort expert-level prostate cancer delineation","","Accurate segmentation of prostate cancer on MR images is essential for planning image-guided interventions, yet tumour appearances vary subtly across imaging protocols and expert availability is limited. Automated approaches require large, consistently annotated datasets, while manual delineation remains labor-intensive. This work presents a prompting-driven framework that bridges manual and automated segmentation by using user-provided point prompts. A reinforcement learning agent iteratively selects new points to guide region-growing, improving accuracy and uncertainty-aware exploration, while reducing user effort compared with fully automated methods.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/promptable-segmentation-with-region-exploration-enables-minimal-effort-expert-level-prostate-cancer-delineation/355970/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/promptable-segmentation-with-region-exploration-enables-minimal-effort-expert-level-prostate-cancer-delineation/355970.png","ImageObject",300,407,{"name":42,"@type":43},"Bulrr","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-15","2026-09-23",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",9,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"Why is prostate cancer segmentation on MR images difficult?","Question",{"text":63,"@type":64},"Cancer can appear subtle and variable across imaging protocols, and differences in acquisition and radiologist training add further complexity. Limited expert availability also restricts consistent interpretation and annotation.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How does the proposed framework use user prompts?",{"text":68,"@type":64},"Starting from an initial point prompt, region growing produces a preliminary mask. Reinforcement learning observes the image and current segmentation to propose additional points, and region growing updates the mask iteratively.",{"name":70,"@type":61,"acceptedAnswer":71},"What were the reported results on public datasets?",{"text":72,"@type":64},"The framework was evaluated on PROMIS and PICAI (566 and 1090 cases). It outperformed previous best automated methods by 9.9% and 8.9%, achieving performance comparable to manual radiologist segmentation while reducing annotation time tenfold.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},355970,1791351519,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,119,122,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":97,"slug":129},19,"General","general",{"code":4,"msg":82,"data":131},{"doc_id":79,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":55,"language":139,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":12,"update_tm":143,"read_time":144,"preview_page_count":33,"preview_urls":145},5909892115043,"https://ap-avatar.wpscdn.com/davatar_6f874abed73319feea01a86fa6f0fab8","International Journal of Computer Assisted Radiology and Surgery (2026) 21:997–1005  \n[https://doi.org/10.1007/s1](https://doi.org/10.1007/s1) 1548-026-03628-w  \nPromptable segmentation with region exploration enables minimal-effort expert-level prostate cancer delineation  \nJunqing Yang1 · Natasha Thorley2 · Ahmed Nadeem Abbasi3 · Shonit Punwani2 · Zion Tse4 · Yipeng Hu1 · Shaheer U. Saeed1,4,5  \nReceived: 19 February 2026 / Accepted: 26 March 2026 / Published online: 17 April 2026 © The Author(s) 2026  \nAbstract  \nPurpose Accurate segmentation of prostate cancer on magnetic resonance (MR) images is crucial for planning image-guided interventions such as targeted biopsies, cryoablation, and radiotherapy. However, subtle and variable tumour appearances, differences in imaging protocols, and limited expert availability make consistent interpretation difﬁcult. While automated methods aim to address this, they rely on large expertly annotated datasets that are often inconsistent, whereas manual delineation remains labour-intensive. This work aims to bridge the gap between automated and manual segmentation through a framework driven by user-provided point prompts, enabling accurate segmentation with minimal annotation effort. Methods The framework combines reinforcement learning (RL) with a region-growing segmentation process guided by user prompts. Starting from an initial point prompt, region-growing generates a preliminary segmentation, which is iterativelyreﬁned through RL. At each step, the RL agent observes the image and current segmentation to predict a new point, from which region growing updates the mask. A reward, balancing segmentation accuracy and voxel-wise uncertainty, encourages exploration of ambiguous regions, allowing the agent to escape local optima and perform sample-speciﬁc optimisation. Despite requiring fully supervised training, the framework bridges manual and fully automated segmentation at inference by substantially reducing user effort while outperforming current fully automated methods.  \nResults The framework was evaluated on two public prostate MR datasets (PROMIS and PICAI, with 566 and 1090 cases) . It outperformed the previous best automated methods by 9.9% and 8.9%, respectively, with performance comparable to manual radiologist segmentation, reducing annotation time tenfold.  \nConclusion By combining prompting with RL–driven exploration, the framework achieves radiologist-level prostate cancer segmentation with a fraction of the annotation effort, highlighting the potential of RL to enable adaptive and efﬁcient cancer delineation. Code: URL: [github.com/JQ-Sakura/prostate-rl-segmentation](github.com/JQ-Sakura/prostate-rl-segmentation)  \nKeywords Promptable segmentation · Prostate cancer · Reinforcement learning · Deep learning  \nB Shaheer U. Saeed [shaheer.saeed@qmul.ac.uk](shaheer.saeed@qmul.ac.uk)  \n1 UCL Hawkes Institute; Department of Medical Physics and Biomedical Engineering, University College London, London, UK  \n2 Centre of Medical Imaging, University College London, London, UK  \n3 Department of Oncology, Aga Khan University Hospital, Karachi, Pakistan  \n4 Centre for Bioengineering; School of Engineering and Materials Science, Queen Mary University of London, London, UK  \nIntroduction  \nMagnetic resonance (MR) imaging plays a central role in interventional planning, for both diagnosis and treatment of prostate cancer [1] . However, accurate interpretation of prostate MR images is particularly challenging due to the variable and often subtle imaging appearances of cancerous tissue. This is echoed by the low reported sensitivity, even inexpert readings [1] . The task is further complicated by differences in acquisition protocols, imaging equipment, and radiologist training [2] . Reliable assessment therefore requires  \n5 Digital Environment Research Institute, Queen Mary University of London, London, UK  \nspecialist expertise, which is severely limited, especially in resource-constrained ","cbCaik7dX2o8rjd0","https://ap.wps.com/l/cbCaik7dX2o8rjd0","pdf",1708729,"English","# Abstract\n## Purpose\n## Methods\n## Results\n## Conclusion\n# Introduction\n## Challenges in prostate MR segmentation\n## Limits of fully automated methods\n## Semi-automated and promptable approaches","[{\"question\":\"Why is prostate cancer segmentation on MR images difficult?\",\"answer\":\"Cancer can appear subtle and variable across imaging protocols, and differences in acquisition and radiologist training add further complexity. Limited expert availability also restricts consistent interpretation and annotation.\"},{\"question\":\"How does the proposed framework use user prompts?\",\"answer\":\"Starting from an initial point prompt, region growing produces a preliminary mask. Reinforcement learning observes the image and current segmentation to propose additional points, and region growing updates the mask iteratively.\"},{\"question\":\"What were the reported results on public datasets?\",\"answer\":\"The framework was evaluated on PROMIS and PICAI (566 and 1090 cases). It outperformed previous best automated methods by 9.9% and 8.9%, achieving performance comparable to manual radiologist segmentation while reducing annotation time tenfold.\"}]","Promptable segmentation with region exploration enables minimal-effort expert-level prostate cancer delineation | PDF",1790122025,23,[146,150,154,158],{"page_no":22,"url":147,"width":148,"height":149},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/637e9ca997492d2199aa4640a9a14cdc.webp",983,1305,{"page_no":26,"url":151,"width":152,"height":153},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/16ad6e30859915c0f74827e041200333.webp",982,1304,{"page_no":30,"url":155,"width":156,"height":157},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/30adc2fdcfba60959a31a1db891f9f5a.webp",981,1303,{"page_no":33,"url":159,"width":160,"height":161},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/e06aa4d33e38ce2655a0ef456281bb61.webp",980,1301]