[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-369915-105":3,"detail-sidebar-cat-0-en-105":80,"doc-detail-369915-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":73,"head_meta":75,"extra_data":77,"updated_unix":79},105,"en","assessing-the-risk-of-recurrence-in-early-stage-breast-cancer-through-he-stained-whole-slide-images","Assessing the risk of recurrence in early-stage breast cancer through H&E stained whole slide images","","Accurate prediction of recurrence risk is critical for selecting postoperative treatment in early-stage breast cancer. This study evaluates whether deep learning can predict recurrence risk using 125 hematoxylin and eosin (H&E) stained whole slide images from two institutions. Models achieve strong sensitivity across low, intermediate, and high-risk groups and show specificity up to 0.972, with a Pearson correlation of 0.61 to histological grade. Class activation maps highlight tubule formation and mitotic rate. Results suggest H&E WSI–only deep learning can approximate genomic assay risk stratification, while larger, balanced datasets are needed for clinical validation.",{"@graph":14,"@context":72},[15,34,55],{"@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/assessing-the-risk-of-recurrence-in-early-stage-breast-cancer-through-he-stained-whole-slide-images/369915/",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/assessing-the-risk-of-recurrence-in-early-stage-breast-cancer-through-he-stained-whole-slide-images/369915.png","ImageObject",300,407,{"name":42,"@type":43},"Blitz","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-07","2026-09-23",true,{"@type":52,"interactionType":53,"userInteractionCount":26},"InteractionCounter",{"@type":54},"ViewAction",{"@type":56,"mainEntity":57},"FAQPage",[58,64,68],{"name":59,"@type":60,"acceptedAnswer":61},"What is the main goal of the study?","Question",{"text":62,"@type":63},"To determine whether deep learning can predict recurrence risk in early-stage breast cancer by analyzing H&E stained whole slide images.","Answer",{"name":65,"@type":60,"acceptedAnswer":66},"How were the whole slide images obtained and selected?",{"text":67,"@type":63},"The study used 125 H&E stained WSIs, one per patient, collected from surgical specimens obtained between January 2014 and July 2019 at two institutions.",{"name":69,"@type":60,"acceptedAnswer":70},"How well did the models perform in recurrence risk prediction?",{"text":71,"@type":63},"Sensitivity varied by risk category (0.857, 0.746, 0.529) and specificity ranged from 0.816 to 0.972, with a Pearson correlation of 0.61 with histological grade.","https://schema.org",{"og:url":32,"og:type":74,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":76,"canonical":32},"index,follow",{"doc_id":78,"site_id":7},369915,1790185201,{"code":4,"msg":81,"data":82},"success",[83,87,91,95,100,105,110,114,119,122,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":84,"show_sort_weight":85,"slug":86},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":88,"show_sort_weight":89,"slug":90},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":92,"show_sort_weight":93,"slug":94},"Exam",70,"exam",{"id":96,"doc_module":4,"doc_module_name":25,"category_name":97,"show_sort_weight":98,"slug":99},5,"Comic",60,"comic",{"id":101,"doc_module":4,"doc_module_name":25,"category_name":102,"show_sort_weight":103,"slug":104},6,"Technology",50,"technology",{"id":106,"doc_module":4,"doc_module_name":25,"category_name":107,"show_sort_weight":108,"slug":109},7,"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":96,"slug":129},19,"General","general",{"code":4,"msg":81,"data":131},{"doc_id":78,"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":26,"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":79,"read_time":144,"preview_page_count":33,"preview_urls":145},24464137899374,"https://us-avatar.wpscdn.com/davatar_3d24733baf745e90a7e4bdd5f77d97b2","[www. nature.com/scientificreports](www. nature.com/scientificreports)  \nOPEN  \nAssessing the risk of recurrence in early-stage breast cancer through H&E stained whole slide images  \nGeongyu Lee1, Joonho Lee1, Tae-Yeong Kwak1, Sun Woo Kim1, Youngmee Kwon2, Chungyeul Kim3􀀍 & Hyeyoon Chang4􀀍  \nAccurate prediction of the likelihood of recurrence is important in the selection of postoperative treatment for patients with early-stage breast cancer. In this study, we investigated whether deep learning algorithms can predict patients’ risk of recurrence by analyzing the pathology images of their cancer histology. We analyzed 125 hematoxylin and eosin-stained whole slide images (WSIs) from  \n125 patients across two institutions (National Cancer Center and Korea University Medical Center Guro Hospital) to predict breast cancer recurrence risk using deep learning. Sensitivity reached 0.857, 0.746, and 0.529 for low, intermediate, and high-risk categories, respectively, with specificity of 0.816, 0.803, and 0.972, and a Pearson correlation of 0.61 with histological grade. Class activation maps highlighted features like tubule formation and mitotic rate, suggesting a cost-effective approach to risk stratification, pending broader validation. These findings suggest that deep learning models trained exclusively on hematoxylin and eosin stained whole slide images can approximate genomic assay results, offering a cost-effective and scalable tool for breast cancer recurrence risk assessment. However, further validation using larger and more balanced datasets is needed to confirm the clinical applicability of our approach.  \nKeywords Deep learning, OncotypeDX, Recurrence risk, Breast cancer  \nBreast cancer is the most prevalent cancer worldwide, making early detection and rapid selection of treatment methods crucial for patients1,2. A significant number of early-stage breast cancers, particularly those that are estrogen receptor (ER)-positive and human epidermal growth factor receptor 2 (HER2)-negative, are classified as clinically low to intermediate risk based on traditional clinicopathological factors. This classification provides limited guidance for adjuvant chemotherapy decisions. Consequently, prognostic risk profiling has become a fundamental component of contemporary breast cancer diagnostics, offering additional risk information to identify patients who may not require adjuvant chemotherapy1. Among the various prognostic testing methods, genomic assays that assess molecular expression levels are noted for their relatively high predictive accuracy3–5. The Oncotype DX assay, in particular, evaluates the likelihood of recurrence in early breast cancer patients and the potential benefit from chemotherapy by analyzing the expression of 21 genes, including ER, PR, HER2 and Ki-67 . This assay generates a Recurrence Score (RS) ranging from 0 to 100, categorizing risk into low, intermediate, and high. It is endorsed by major oncology guidelines and demonstrates high predictive reliability for breast cancer recurrence6,7. The role of traditional diagnostic parameters, such as tumor size, stage, and grade, remains vital in clinical decision-making. Histologic grade is an especially crucial prognostic factor in breast cancer8–10. However, the classification of approximately 50% of all breast cancers and around 60% of ER-positive/HER2-negative tumors as histologic grade 2 introduces variability in aggressiveness and prognosis, limiting their utility in therapeutic decision-making. The integration of digital pathology and the advancement of artificial intelligence, particularly deep learning, has enhanced the capabilities of computational pathology11–14. These technologies enable the prediction of patient outcomes, response to therapy, and molecular characteristics through the analysis of histopathological images, thus playing a pivotal role in the advancement of precision medicine15–18. This study introduced a novel approach to enhance prognosis pr","cbCaiuTcB2hr4KB2","https://ap.wps.com/l/cbCaiuTcB2hr4KB2","pdf",4204499,11,"English","# Materials and methods\n## Materials","[{\"question\":\"What is the main goal of the study?\",\"answer\":\"To determine whether deep learning can predict recurrence risk in early-stage breast cancer by analyzing H\\u0026E stained whole slide images.\"},{\"question\":\"How were the whole slide images obtained and selected?\",\"answer\":\"The study used 125 H\\u0026E stained WSIs, one per patient, collected from surgical specimens obtained between January 2014 and July 2019 at two institutions.\"},{\"question\":\"How well did the models perform in recurrence risk prediction?\",\"answer\":\"Sensitivity varied by risk category (0.857, 0.746, 0.529) and specificity ranged from 0.816 to 0.972, with a Pearson correlation of 0.61 with histological grade.\"}]","Assessing the risk of recurrence in early-stage breast cancer through H&E stained whole slide images | PDF",28,[146,150,154,158],{"page_no":22,"url":147,"width":148,"height":149},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/c98e2c62d6424f3daf9ab6ff177457f8.webp",983,1291,{"page_no":26,"url":151,"width":152,"height":153},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/519bae0f5e41997e6de7fa56f0ed4b58.webp",982,1290,{"page_no":30,"url":155,"width":156,"height":157},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/0e1ce2367938c4b48989f5bd13cd9424.webp",981,1289,{"page_no":33,"url":159,"width":160,"height":161},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/68427032b4b0f94dc9f5fe751376d240.webp",980,1287]