[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-369881-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-369881-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","uncertainty-aware-ensemble-of-foundation-models-differentiates-glioblastoma-from-its-mimics","Uncertainty-aware ensemble of foundation models differentiates glioblastoma from its mimics","","Accurate pathological diagnosis is essential for personalized treatment of central nervous system cancers, yet distinguishing glioblastoma from primary central nervous system lymphoma remains difficult because of overlapping pathology. The PICTURE system is proposed using 2141 worldwide pathology slides and combines Bayesian inference, deep ensemble learning, and normalizing flow to manage uncertainty in predictions and label noise. PICTURE achieves strong diagnostic performance, with validated AUROC across five independent cohorts, and further identifies rare CNS cancer types beyond gliomas and lymphomas. The framework enables rapid and generalizable differentiation of pathological mimics at intraoperative and final stages.",{"@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/uncertainty-aware-ensemble-of-foundation-models-differentiates-glioblastoma-from-its-mimics/369881/",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/uncertainty-aware-ensemble-of-foundation-models-differentiates-glioblastoma-from-its-mimics/369881.png","ImageObject",300,407,{"name":42,"@type":43},"Ben ","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-03","2026-09-23",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",5,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"Why is distinguishing glioblastoma from its mimics clinically challenging?","Question",{"text":63,"@type":64},"Their pathology can overlap, making accurate differentiation difficult during both frozen section and FFPE evaluation stages.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What is PICTURE and how does it handle uncertainty?",{"text":68,"@type":64},"PICTURE is a pathology image characterization tool that uses Bayesian inference, deep ensemble learning, and normalizing flow to explicitly model uncertainty in predictions and training labels.",{"name":70,"@type":61,"acceptedAnswer":71},"How is PICTURE’s diagnostic performance validated?",{"text":72,"@type":64},"Performance is validated in five independent cohorts, with AUROC reported across these datasets.","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},369881,1790614443,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,100,105,110,114,119,122,126],{"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":55,"doc_module":4,"doc_module_name":25,"category_name":97,"show_sort_weight":98,"slug":99},"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":55,"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":108,"preview_page_count":33,"preview_urls":145},2336478951081,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","Article [https://doi.org/10.1038/s41467-025-64249-6](https://doi.org/10.1038/s41467-025-64249-6)  \nUncertainty-aware ensemble of foundation models differentiates glioblastoma from its mimics  \nReceived: 23 April 2025  \n\n| Accepted: 9 September 2025 |\n| --- |\n| |\n| Check for updates |\n\nJunhan Zhao 1,2,18, Shih-Yen Lin1,18, Raphaël Attias 1, Liza Mathews1, Christian Engel1, Guillaume Larghero1, Dmytro Vremenko1, Ting-Wan Kao1, Tsung-Hua Lee1, Yu-Hsuan Wang3, Cheng Che Tsai1, Eliana Marostica 1, Ying-Chun Lo 4, David Meredith5, Keith L. Ligon 6, Omar Arnaout7, Thomas Roetzer-Pejrimovsky 8, Shih-Chieh Lin9, Natalie NC Shih10, Nipon Chaisuriya 4,11, David J. Cook 4, Jung-Hsien Chiang 3,  \nChia-Jen Liu 1,12,13, Adelheid Woehrer 8,14, Jeffrey A. Golden15, MacLean P. Nasrallah 10 & Kun-Hsing Yu 1,5,16,17   \nAccurate pathological diagnosis is crucial in guiding personalized treatments for patients with central nervous system cancers. Distinguishing glioblastoma and primary central nervous system lymphoma is particularly challenging due to their overlapping pathology features, despite the distinct treatments required. To address this challenge, we establish the Pathology Image Characterization Tool with Uncertainty-aware Rapid Evaluations (PICTURE) system using 2141 pathology slides collected worldwide. PICTURE employs Bayesian inference, deep ensemble, and normalizing ﬂow to account for the uncertainties in its predictions and training set labels. PICTURE accurately diagnoses glioblastoma and primary central nervous system lymphoma with an area under the receiver operating characteristic curve (AUROC) of 0.989, with the results validated in ﬁve independent cohorts (AUROC = 0.924-0.996). In addition, PICTURE identiﬁes samples belonging to 67 types of rare central nervous system cancers that are neither gliomas nor lymphomas. Our approaches provide a generalizable framework for differentiating pathological mimics and enable rapid diagnoses for central nervous system cancer patients.  \nMore than 86,000 patients in the U.S. are diagnosed with CNS neoplasms annually, leading to over 16,000 deaths each year1. The 2021 WHO Classiﬁcation ofCNS Tumors(WHO CNS5)2 identiﬁes109distinct tumor subtypes based on pathology and molecular proﬁles3. Because treatments and prognoses of different CNS tumors vary considerably4–7, obtaining accurate pathological diagnoses is critical. Glioblastoma, the most common brain cancer in the U.S., has a dismal median survival of 8 months1,5, and surgical resection remains the cornerstone of initial treatment7. Notably, previous studies showed that primary central nervous system lymphoma (PCNSL) is the cancer  \ntype most frequently misdiagnosed as glioblastoma8–12. This misclassiﬁcation has important clinical implications: patients with PCNSL have a median survival of more than three years following diagnosis and often respond well to radiotherapy4,5. Although patients’ age, immune status, and imaging features from magnetic resonance imaging inﬂuence clinicians’ initial diagnostic assessments, pathology evaluation using tumor samples provides the ﬁnal diagnosis6,7. When PCNSL is diagnosed during surgery with the intent for tumor removal, neurosurgeons will usually discontinue further surgical intervention topreserve neurological function and refer patients for radiotherapy  \nA full list of afﬁliations appears at the end of the paper. e-mail: [Kun-Hsing_Yu@hms.harvard.edu](Kun-Hsing_Yu@hms.harvard.edu)  \ncombined with chemotherapy7–9. In addition, ﬁnal diagnosis using formalin-ﬁxed, parafﬁn-embedded (FFPE) tissue conﬁrms tumor types and guides long-term treatment planning2. Thus, accurate distinction between glioblastoma and PCNSL at both intraoperative and ﬁnal diagnostic stages is therefore essential to avoid unnecessary surgery and ensure timely initiation of appropriate therapy.  \nSeveral challenges have hindered the accurate pathological diagnosis of CNS neoplasms1,6. The current issue in diagnosing glioblastoma ","cbCaijleINHhu4fb","https://ap.wps.com/l/cbCaijleINHhu4fb","pdf",4405031,16,"English","# Introduction\n# Methods: PICTURE System\n## Uncertainty Modeling\n# Results and Validation\n# Discussion: Clinical Implications and Challenges","[{\"question\":\"Why is distinguishing glioblastoma from its mimics clinically challenging?\",\"answer\":\"Their pathology can overlap, making accurate differentiation difficult during both frozen section and FFPE evaluation stages.\"},{\"question\":\"What is PICTURE and how does it handle uncertainty?\",\"answer\":\"PICTURE is a pathology image characterization tool that uses Bayesian inference, deep ensemble learning, and normalizing flow to explicitly model uncertainty in predictions and training labels.\"},{\"question\":\"How is PICTURE’s diagnostic performance validated?\",\"answer\":\"Performance is validated in five independent cohorts, with AUROC reported across these datasets.\"}]","Uncertainty-aware ensemble of foundation models differentiates glioblastoma from its mimics | PDF",1790185050,[146,150,154,158],{"page_no":22,"url":147,"width":148,"height":149},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/42f6808d46072074f5b8064360b70b0c.webp",983,1305,{"page_no":26,"url":151,"width":152,"height":153},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/4663399cf5114ee6a896bdb6f05fa71a.webp",982,1304,{"page_no":30,"url":155,"width":156,"height":157},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/6f8b75a8a9c3047c95d18b6f9074509b.webp",981,1303,{"page_no":33,"url":159,"width":160,"height":161},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/fdc9ba570cf530e091706448c47dd930.webp",980,1301]