[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-346701-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-346701-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","normal-breast-tissue-nbt-classifiers-advancing-compartment-classification-in-normal-breast-histology","Normal breast tissue (NBT)-classifiers - advancing compartment classification in normal breast histology","","Cancer research highlights the importance of early detection, yet quantitative approaches for analyzing normal breast tissue remain limited. Digitized haematoxylin and eosin (H&E) whole-slide images enable computational histopathology, but AI-driven analysis of normal breast tissue is still scarce. The study curated 70 normal breast tissue WSIs with pathologist-guided annotations and built convolutional neural network patch-level models (NBT-Classifiers). Across external cohorts, models achieved AUCs up to 1.00 and supported interpretability.",{"@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/normal-breast-tissue-nbt-classifiers-advancing-compartment-classification-in-normal-breast-histology/346701/",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/normal-breast-tissue-nbt-classifiers-advancing-compartment-classification-in-normal-breast-histology/346701.png","ImageObject",300,407,{"name":42,"@type":43},"Aurora","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-26","2026-09-22",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",7,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"Why is quantitative AI analysis of normal breast tissue (NBT) important?","Question",{"text":63,"@type":64},"Early detection requires understanding what normal tissue looks like in detail. Existing quantitative methods for normal breast tissue analysis are limited, motivating AI approaches for reliable characterization.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How were the training data for NBT-Classifiers prepared?",{"text":68,"@type":64},"The work curated 70 digitized H&E whole-slide images of NBT from multiple sources and cohorts. Pathologists provided manual annotations for epithelium, stroma, and adipocytes.",{"name":70,"@type":61,"acceptedAnswer":71},"What performance did NBT-Classifiers achieve on external cohorts?",{"text":72,"@type":64},"Across three external cohorts, NBT-Classifiers trained on 128 × 128 µm and 256 × 256 µm patches achieved AUCs in the range of 0.98–1.00.","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},346701,1790195425,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,110,114,119,122,126],{"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":55,"doc_module":4,"doc_module_name":25,"category_name":107,"show_sort_weight":108,"slug":109},"Healthcare",40,"healthcare",{"id":111,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":112,"slug":113},8,30,"research-report",{"id":115,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},9,"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":111,"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":139,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":144,"read_time":112,"preview_page_count":33,"preview_urls":145},4810365810221,"https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d","npj | breast cancer Article  \n\n| Published in partnership with the Breast Cancer Research Foundation |  | |\n| --- | --- | --- |\n| [https://doi.org/10.1038/s41523-026-00896-2](https://doi.org/10.1038/s41523-026-00896-2) |  |  |\n| Normal breast tissue (NBT)-classiﬁers: advancing compartment classiﬁcation in normal breast histology\u003Cbr> Check for updates |  |  |\n| Siyuan Chen1,14, Mario Parreno-Centeno1,14, Graham Booker1, Gregory Verghese1,2,3,\u003Cbr>Fathima Sumayya Mohamed1, Salim Arslan4, Pahini Pandya4, Aasiyah Oozeer2,5, Marcello D’Angelo5, Rachel Barrow2,6, Rachel Nelan2,6, Marcelo Sobral-Leite7, Fabio de Martino8, Cathrin Brisken8,9, Matthew J. Smalley10, Esther H. Lips7, Cheryl Gillett2,5, Louise J. Jones2,6, Christopher R. S. Banerji1,11,12, Sarah E. Pinder13 & Anita Grigoriadis1,2,3  |  |  |\n| Cancer research emphasises early detection, yet quantitative methods for normal tissue analysis remain limited. Digitised haematoxylin and eosin (H&E)-stained slides enable computational histopathology, but artiﬁcial intelligence (AI)-based analysis of normal breast tissue (NBT) in whole slide images (WSIs) remains scarce. We curated 70 WSIs of NBTs from multiple sources and cohorts with pathologist-guided manual annotations of epithelium, stroma, and adipocytes ([https://github](https://github). com/cancerbioinformatics/OASIS). We developed robust convolutional neural network (CNN)-based, patch-level classiﬁcation models, named NBT-Classiﬁers, to tessellate and classify NBTs at different scales. Across three external cohorts, NBT-Classiﬁers trained on 128 × 128 µm and 256 × 256 µm patches achieved AUCs of 0 .98–1.00. The model learned independent normal features different from those of precancerous and cancerous epithelium, which were further visualised using two explainable AI techniques. When integrated into an end-to-end preprocessing pipeline, NBT-Classiﬁers facilitate efﬁcient downstream analysis within peri-lobular regions. NBT-Classiﬁers provide robust compartment-speciﬁc analytical tools and enhance our understanding of NBT appearances, which serve as valuable reference points for identifying premalignant changes and guiding early breast cancer prevention strategies. |  |  |\n| Normal breast tissue (NBT) research is gaining momentum for the early detection of breast cancer1. The epithelium, organised into ducts and lobules, also referred to as terminal duct-lobular units (TDLUs), is of primary importance, as these structures are where most physiological processes and pathological alterations occur2. Apart from the epithelium, recent evidence suggests that the stroma, particularly the peri-lobular stroma, may also harbour precursor alterations and have a crucial role in breast cancer | progression3–6. Therefore, in-depth histopathological research into these structures, especially in women with varying breast cancer risk, holds promise for detecting early signs ofcancer initiation in breast tissues that appear“normal”. This is particularly signiﬁcant for women at higher breast cancer risk, such as germline BRCA1/2 mutation (gBRCA1/2m) carriers7.\u003Cbr>Histopathology remains the gold standard for diagnosing cancer, relying on microscopic examination to detect tissue abnormalities. |  |\n\n1Cancer Bioinformatics, School of Cancer & Pharmaceutical Sciences, Faculty of Life Sciences and Medicine, King’s College London, London, UK. 2PharosAI, London, UK. 3The Breast Cancer Now Research Unit, School of Cancer and Pharmaceutical Sciences, Faculty of Life Sciences and Medicine, King’s College London, London, UK. 4Panakeia Technologies, London, UK. 5King’s Health Partners Cancer Biobank, King’s College London, London, UK. 6Centre for Tumour Biology, Barts Cancer Institute, Queen Mary University of London, London, UK. 7Division of Molecular Pathology, Netherlands Cancer Institute, Amsterdam, The Netherlands. 8Swiss Institute for Experimental Cancer Research, School of Life Sciences, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerl","cbCaiikXejwpRlo4","https://ap.wps.com/l/cbCaiikXejwpRlo4","pdf",4364498,12,"English","# Background and motivation\n## Normal breast tissue structure and clinical relevance\n## Computational pathology and AI for WSI analysis\n# Data and model development\n## Curated WSIs and manual annotations\n## NBT-Classifiers architecture and patch-level classification\n# Results and validation\n## Performance across external cohorts\n## Learned normal features and visualization with explainable AI\n# Application and impact\n## Integration into preprocessing pipelines\n## Downstream peri-lobular analysis and early prevention insights","[{\"question\":\"Why is quantitative AI analysis of normal breast tissue (NBT) important?\",\"answer\":\"Early detection requires understanding what normal tissue looks like in detail. Existing quantitative methods for normal breast tissue analysis are limited, motivating AI approaches for reliable characterization.\"},{\"question\":\"How were the training data for NBT-Classifiers prepared?\",\"answer\":\"The work curated 70 digitized H\\u0026E whole-slide images of NBT from multiple sources and cohorts. Pathologists provided manual annotations for epithelium, stroma, and adipocytes.\"},{\"question\":\"What performance did NBT-Classifiers achieve on external cohorts?\",\"answer\":\"Across three external cohorts, NBT-Classifiers trained on 128 × 128 µm and 256 × 256 µm patches achieved AUCs in the range of 0.98–1.00.\"}]","Normal breast tissue (NBT)-classifiers - advancing compartment classification in normal breast histology | PDF",1790062454,[146,150,154,158],{"page_no":22,"url":147,"width":148,"height":149},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/33d3681e3bbaa5f9dc4b070864c35cda.webp",983,1305,{"page_no":26,"url":151,"width":152,"height":153},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/2a0865c7f46d2b09115321196d67a425.webp",982,1304,{"page_no":30,"url":155,"width":156,"height":157},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/33bffee8892cfa66b792b091550eb0c1.webp",981,1303,{"page_no":33,"url":159,"width":160,"height":161},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/8908b3e557fcb2ac200dd6af654f7e7b.webp",980,1301]