[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-356853-105":59,"doc-detail-356853-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","comprehensive-evaluation-of-cross-cancer-generalization-in-histopathology-segmentation-models-across-21-tumor-types","Comprehensive evaluation of cross cancer generalization in histopathology segmentation models across 21 tumor types","","Artificial intelligence has advanced computational pathology by enabling high-resolution tumor segmentation, but building models for each tumor type is resource intensive and often limited to single-domain training. This work evaluates cross-tumor generalization of five segmentation models across 21 cancer types using over 7,700 whole slide images. Results show strong performance for several models, especially lung, supporting faster reuse and future pan-cancer model development.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/comprehensive-evaluation-of-cross-cancer-generalization-in-histopathology-segmentation-models-across-21-tumor-types/356853/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/comprehensive-evaluation-of-cross-cancer-generalization-in-histopathology-segmentation-models-across-21-tumor-types/356853.png","ImageObject",300,407,{"name":92,"@type":93},"Cipher","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-14","2026-09-23",true,{"@type":102,"interactionType":103,"userInteractionCount":29},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"Why is cross-cancer generalization important for histopathology segmentation models?","Question",{"text":112,"@type":113},"Deep learning segmentation tools are often restricted to one tumor type, making model development costly. Cross-cancer generalization can enable reuse across cancers and support pan-cancer segmentation.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How was model generalization evaluated in this study?",{"text":117,"@type":113},"Five tissue segmentation models (breast, colon, lung, kidney, prostate) were tested on 21 cancer types from The Cancer Genome Atlas, using over 7,700 whole slide images. Representative tumor and benign regions were manually selected and accuracy was scored on a 0–10 semiquantitative scale.",{"name":119,"@type":110,"acceptedAnswer":120},"Which model showed the strongest cross-cancer performance and what were the key findings?",{"text":121,"@type":113},"The lung model demonstrated excellent cross-cancer performance, achieving an overall mean score of 7.9 ± 2.1 and strong results across many non-lung cancers, including particularly strong performance in ovarian cancer. Breast and colon models also generalized well, while kidney and prostate generalized more limitedly.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},356853,1790445118,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":29,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":144,"read_time":36,"preview_page_count":19,"preview_urls":145},687208528416,"https://ap-avatar.wpscdn.com/davatar_9964176cb1d06d4a9deccf72a44ae3dc","A Nature Portfolio journal  \n[https://doi.org/10.1038/s43856-026-01601-x](https://doi.org/10.1038/s43856-026-01601-x)  \n\n| Comprehensive evaluation of cross cancer generalization in histopathology segmentation models across 21 tumor types\u003Cbr> Check for updates |  |  |\n| --- | --- | --- |\n| Tillmann Bedau 1, Christian Harder1, Abdulkader Al-Shughri1, Yuan Wang1, Alexey Pryalukhin2, Marie-Lisa Eich3, Su Ir Lyu1, Reinhard Büttner 1, Alexander Quaas1 & Yuri Tolkach1  |  |  |\n| Abstract Plain language summary |  |  |\n| Background Artiﬁcial intelligence has signiﬁcantly advanced computational pathology by enabling high-resolution, clinical-grade tumor segmentation models with state-of-the-art diagnostic accuracy. Creating such models is resource-intensive, requiring substantial time and domain expertise. Additionally, deep learning models are typically restricted to single tumor types, making it challenging to develop separate models for each tumor type. Crosscancer generalization of segmentation models could address this bottleneck and pave the way for pan-cancer segmentation models.\u003Cbr>Methods We evaluated the cross-tumor generalization capability of ﬁve tissue segmentation models (breast, colon, lung, kidney, prostate) using 21 cancer types from The Cancer Genome Atlas, totaling over 7,700 whole slide images. Representative large tumor and benign regions were manually selected, and segmentation accuracy was evaluated using a semiquantitative scale (0-10) .\u003Cbr>Results Here we show that the lung model demonstrates excellent cross-cancer performance (overall mean score 7.9 ± 2.1), effectively segmenting tumor regions in many non-lung cancers with segmentation accuracy similar to its native domain in 11 of 19 other epithelial tumors and melanoma, achieving particularly strong results in ovarian cancer (9.2 ± 0.9). The breast and colon models also show strong cross-domain performance, while the kidney and prostate models exhibit more limited generalization. Overall, high-precision segmentation is achievable in most cancer types using existing models.\u003Cbr>Conclusions Existing segmentation models generalize across multiple cancer types, reducing the need to develop new, entity-speciﬁc models from scratch. This cross-domain generalization enables fast-track model development and supports future creation of robust pan-cancer segmentation models. Leveraging these capabilities could accelerate clinical integration of pathology artiﬁcial intelligence tools and enable reproducible biomarker discovery. |  | Computer tools can help doctors who examine tissue samples under the microscope by automatically marking areas of cancer. However, building these tools for each type of cancer takes a lot of time and specialist knowledge. We asked whether a tool built for one type of cancer could also work for other types. We tested ﬁve such tools, each originally built fora speciﬁc cancer (breast, bowel, lung, kidney, or prostate), on tissue images from 21 different cancer types, covering more than 7,700 images in total. We found that several tools, especially the one built for lung cancer, could accurately identify cancer regions in manyother cancer types as well. This means that existing tools can be reused for new cancers without having to build them from scratch, cutting the time needed from months to days and helping bring these computer-assisted tools to hospital laboratories sooner. |\n| Theﬁeld ofcomputational pathology has seen remarkable advancementsin recent years, with artiﬁcial intelligence (AI) models, mostly deep learning (DL)-based, increasingly demonstrating their potential to complement traditional histopathological analysis.1–3 However, certain limitations of DL models constrain their broad application, with only few clinical-grade tools | available on the market. One of the major shortcomings of DL models, particularly in a diagnostic oncopathology domain (e.g., multi-class tissue segmentation, tumor detection, grading, subtyping, etc.), ","cbCaimLNRJMr559p","https://ap.wps.com/l/cbCaimLNRJMr559p","pdf",16689399,16,"English","# Abstract\n## Background\n## Methods\n## Results\n## Conclusions","[{\"question\":\"Why is cross-cancer generalization important for histopathology segmentation models?\",\"answer\":\"Deep learning segmentation tools are often restricted to one tumor type, making model development costly. Cross-cancer generalization can enable reuse across cancers and support pan-cancer segmentation.\"},{\"question\":\"How was model generalization evaluated in this study?\",\"answer\":\"Five tissue segmentation models (breast, colon, lung, kidney, prostate) were tested on 21 cancer types from The Cancer Genome Atlas, using over 7,700 whole slide images. Representative tumor and benign regions were manually selected and accuracy was scored on a 0–10 semiquantitative scale.\"},{\"question\":\"Which model showed the strongest cross-cancer performance and what were the key findings?\",\"answer\":\"The lung model demonstrated excellent cross-cancer performance, achieving an overall mean score of 7.9 ± 2.1 and strong results across many non-lung cancers, including particularly strong performance in ovarian cancer. Breast and colon models also generalized well, while kidney and prostate generalized more limitedly.\"}]","Comprehensive evaluation of cross cancer generalization in histopathology segmentation models across 21 tumor types | PDF",1790127680,[146,150,154,158],{"page_no":8,"url":147,"width":148,"height":149},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/419e95eec8305ed030f0bca6165853b4.webp",983,1305,{"page_no":14,"url":151,"width":152,"height":153},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/050fb9fcc415f8c768e9c3d9f2c626a3.webp",982,1304,{"page_no":81,"url":155,"width":156,"height":157},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/1d9a8b401bab7280a4ab3d7e21119173.webp",981,1303,{"page_no":19,"url":159,"width":160,"height":161},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/013c96424a7c4aeb1657354a0807758a.webp",980,1301]