检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Partin, Alexander Vasanthakumari, Priyanka Narykov, Oleksandr Wilke, Andreas Koussa, Natasha Jones, Sara E. Zhu, Yitan Overbeek, Jamie C. Jain, Rajeev Fernando, Gayara Demini Sanchez-Villalobos, Cesar Garcia-Cardona, Cristina Mohd-Yusof, Jamaludin Chia, Nicholas Wozniak, Justin M. Ghosh, Souparno Pal, Ranadip Brettin, Thomas S. Weil, M. Ryan Stevens, Rick L. Division of Data Science and Learning Argonne National Laboratory LemontIL United States Frederick National Laboratory for Cancer Research Cancer Data Science Initiatives Cancer Research Technology Program FrederickMD United States Department of Statistics University of Nebraska–Lincoln LincolnNE United States Department of Electrical & Computer Engineering Texas Tech University LubbockTX United States Division of Computer Computational and Statistical Sciences Los Alamos National Laboratory Los AlamosNM United States Department of Computer Science The University of Chicago ChicagoIL United States

Deep learning (DL) and machine learning (ML) models have shown promise in drug response prediction (DRP), yet their ability to generalize across datasets remains an open question, raising concerns about their real-world applicability. Due to the lack of standardized benchmarking approaches, model evaluations and comparisons often rely on inconsistent datasets and evaluation criteria, making it difficult to assess true predictive capabilities. In this work, we introduce a benchmarking framework for evaluating cross-dataset prediction generalization in DRP models. Our framework incorporates five publicly available drug screening datasets, six standardized DRP models, and a scalable workflow for systematic evaluation. To assess model generalization, we introduce a set of evaluation metrics that quantify both absolute performance (e.g., predictive accuracy across datasets) and relative performance (e.g., performance drop compared to within-dataset results), enabling a more comprehensive assessment of model transferability. Our results reveal substantial performance drops when models are tested on unseen datasets, underscoring the importance of rigorous generalization assessments. While several models demonstrate relatively strong cross-dataset generalization, no single model consistently outperforms across all datasets. Furthermore, we identify CTRPv2 as the most effective source dataset for training, yielding higher generalization scores across target datasets. By sharing this standardized evaluation framework with the community, our study aims to establish a rigorous foundation for model comparison, and accelerate the development of robust DRP models for real-world applications. © 2025, CC BY.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Low rates of chronic graft-versus-host disease with ruxolitinib maintenance following allogeneic HCT

引用

Blood 2025年第20期145卷 2312-2316页

作者： DeFilipp, Zachariah Kim, Haesook T. Knight, Laura W. Dhaver, Shilton E. White, Meghan Dholaria, Bhagirathbhai Schroeder, Mark A. Vasu, Sumithira Abedin, Sameem Chung, Jooho El-Jawahri, Areej Frigault, Matthew J. McAfee, Steven Newcomb, Richard A. Spitzer, Thomas R. Chen, Yi-Bin Hobbs, Gabriela S. Hematopoietic Cell Transplant and Cellular Therapy Program Massachusetts General Hospital Boston MA United States Department of Data Science Dana-Farber Cancer Institute and Harvard T.H. Chan School of Public Health Boston MA United States Cancer Center Protocol Office Massachusetts General Hospital Boston MA United States Department of Hematology-Oncology and Stem Cell Transplantation Vanderbilt University Medical Center Nashville TN United States Bone Marrow Transplantation and Leukemia Section Division of Oncology Washington University School of Medicine in Saint Louis Saint Louis MO United States Division of Hematology Department of Internal Medicine The Ohio State University Columbus OH United States Division of Hematology/Oncology Department of Medicine Medical College of Wisconsin Milwaukee WI United States Center for Leukemia Massachusetts General Hospital Boston MA United States

Despite recent advances in graft-versus-host disease (GVHD) prophylaxis, novel approaches to effective prevention of chronic GVHD (cGVHD) remain of high importance. In this prospective, multicenter, phase 2 trial, ruxolitinib, an oral inhibitor of Janus kinase (JAK) 1 and 2, was administered as maintenance therapy after reduced-intensity allogeneic hematopoietic cell transplantation (HCT). GVHD prophylaxis consisted of tacrolimus and methotrexate. Ruxolitinib began between day +30 to 100 and was administered continuously in 28-day cycles for up to 24 cycles. Seventy-eight participants were enrolled before HCT;63 participants received the intervention. The median start date of ruxolitinib after HCT was day +45. The most common grade ≥3 adverse events were neutropenia, thrombocytopenia, and anemia. Seven participants experienced grade ≥3 infectious events. GVHD-free, relapse-free survival at 1 year after HCT, the primary end point, was 70%. Grade 3 to 4 acute GVHD at 6 months was 4.8%, and moderate-severe cGVHD at 2 years was 16%. cGVHD requiring systemic therapy was 9.5% at 1 year and 13% at 2 years. Overall survival and progression-free survival at 2 years were 76% and 68%, respectively. Prolonged administration of ruxolitinib following HCT is associated with low rates of clinically significant cGVHD. The incorporation of JAK inhibition into GVHD prevention approaches warrants further investigation. This trial was registered at *** as #NCT03286530. © 2025 American Society of Hematology

关键词：

来源：评论

学校读者我要写书评

暂无评论

The spatial zonation of the murine placental vasculature is specified by epigenetic mechanisms

引用

Developmental Cell 2025年第10期60卷 1467-1482.e8页

作者： Gehrs, Stephanie Jakab, Moritz Gutjahr, Ewgenija Gu, Zuguang Weichenhan, Dieter Mallm, Jan-Philipp Mogler, Carolin Schlesner, Matthias Plass, Christoph Schlereth, Katharina Augustin, Hellmut G. Division of Vascular Oncology and Metastasis German Cancer Research Center (DKFZ) Heidelberg 69120 Germany European Center for Angioscience (ECAS) Medical Faculty Mannheim Heidelberg University Mannheim 68167 Germany Faculty of Biosciences Heidelberg University Heidelberg 69120 Germany Institute of Pathology University Clinic Heidelberg Heidelberg 69120 Germany Computational Oncology Group Molecular Precision Oncology Program National Center for Tumor Diseases (NCT) Heidelberg and German Cancer Research Center (DKFZ) Heidelberg 69120 Germany Division of Cancer Epigenomics German Cancer Research Center (DKFZ) Heidelberg 69120 Germany Division of Chromatin Networks German Cancer Research Center (DKFZ) and Bioquant Heidelberg 69120 Germany Institute of Pathology TUM School of Medicine Technical University of Munich Munich 80333 Germany Biomedical Informatics Data Mining and Data Analytics Faculty of Applied Computer Science and Medical Faculty University of Augsburg Augsburg 86159 Germany

The labyrinthian fetoplacental capillary network is vital for proper nourishment of the developing embryo. Dysfunction of the maternal-fetal circulation is a primary cause of placental insufficiency. Here, we show that the spatial zonation of the murine placental labyrinth vasculature is controlled by flow-regulated epigenetic mechanisms. Spatiotemporal transcriptomic profiling identified a gradual change in the expression of epigenetic enzymes, including the de novo DNA methyltransferase 3a (DNMT3A). Loss of Dnmt3a resulted in DNA hypomethylation and perturbation of zonated placental gene expression. The resulting global DNA hypomethylation impaired the angiogenic capacity of endothelial cells. Global or endothelium-predominant deletion of Dnmt3a resulted in impaired placental vascularization and fetal growth retardation (FGR). Human placental endothelial gene expression profiling associated preeclampsia with reduced DNMT3A expression. Collectively, our study identified DMNT3A as critical methylome-regulator of placental endothelial gene expression and function with clinical implications for placental dysfunction, as it occurs during preeclampsia or FGR. © 2024 The Authors

关键词： angiogenesis DNA methylation DNMT3A epigenetic regulation growth retardation placental endothelium spatial zonation

来源：评论

学校读者我要写书评

暂无评论

Narrowing of the Racial Disparities Gap in Survival of Patients With Mycosis Fungoides: A Longitudinal Analysis of the SEER database

引用

Clinical Lymphoma, Myeloma and Leukemia 2025年

作者： Le, Britney Soror, Noha Ismail, Hamid D. Baker, Mohammed Shetayyah, Salman Abu Chung, Catherine G. William, Basem M. Ohio University Heritage College of Osteopathic Medicine Dublin OH The University of Oklahoma College of Medicine Oklahoma City OK United States Department of Computational Data Science and Engineering North Carolina Agricultural and Technical State University Greensboro NC United States Faculty of Medicine Jordan University of Science and Technology Irbid Jordan The Ohio State University Comprehensive Cancer Center: Arthur G. James Cancer Hospital and Richard J. Solove Research Institute Division of Dermatology Columbus OH Ohio Health Blood and Marrow Transplant Program Columbus OH

Background: Mycosis fungoides (MF) is the commonest subtype of cutaneous T-cell lymphoma. In the United States, prior studies reported that African Americans (AA) with MF had poor outcomes. data characterizing differences in racial disparity outcomes over time are limited. Patients and Methods: We collected data from the United States Surveillance, Epidemiology, and End Results (SEER) database to investigate the survival patterns of patients with MF from 1988 to 2011. Cases were divided into 3 cohorts based on the year of diagnosis. Univariable and multivariable analysis were conducted to assess for factors associated with overall survival (OS). Results: 2896 cases of MF were detected, with a median follow-up duration of 60 months. The disparity in survival between the years 1988-1995 and 2004-2011 was significant (P =.05). The parameter estimates of the Cox proportional hazards model for the 1988-1995 period (using the 2004-2011 period as a reference) was also significant (P =.024). Patients diagnosed between 1988 and 1995 were 1.4 times more likely to die from the disease than those diagnosed between 2004 and 2011. The survival gap between AA and white patients narrowed in 1996-2003 and 2004-2011 in comparison to 1988-1995. This indicates improvements in the survival of AA patients over time. Conversely, the survival rates of white patients remained relatively stable over time. Conclusions: Our study demonstrates that AA with MF have reduced survival. Despite the persistent pattern of lower survival across all periods, the gap in survival between white and AA seems to be narrowing over time. © 2025 Elsevier Inc.

关键词： Cutaneous lymphoma Mycosis fungoides Outcome research Racial disparity SEER database

来源：评论

学校读者我要写书评

暂无评论

Correction: Spatial organization of B lymphocytes and prognosis prediction in patients with gastric cancer

引用

Gastric cancer : official journal of the International Gastric cancer Association and the Japanese Gastric cancer Association 2025年 2025 May 16页

作者： Ryan Yong Kiat Tay Manavi Sachdeva Haoran Ma Young-Woo Kim Myeong-Cherl Kook Hyunki Kim Jae-Ho Cheong Lindsay C Hewitt Katharina Nekolla Günter Schmidt Takaki Yoshikawa Takashi Oshima Tomio Arai Supriya Srivastava Ming Teh Xuewen Ong Su Ting Tay Taotao Sheng Joseph J Zhao Patrick Tan Heike I Grabsch Raghav Sundar Yong Loo Lin School of Medicine National University of Singapore 1E Kent Ridge Road Singapore 119228 Singapore. Department of Haematology-Oncology National University Cancer Institute National University Hospital Singapore Singapore. Cancer and Stem Cell Biology Program Duke-NUS Medical School Singapore Singapore. Department of Cancer Policy and Population Health National Cancer Center Graduate School of Cancer Science and Policy and Center for Gastric Cancer and Department of Surgery National Cancer Center Goyang Republic of Korea. Center for Gastric Cancer Department of Pathology National Cancer Center Goyang Republic of Korea. Department of Pathology Yonsei University College of Medicine Seoul Republic of Korea. Department of Surgery Yonsei University College of Medicine Seoul Republic of Korea. Department of Pathology GROW Research Institute for Oncology and Reproduction Maastricht University Medical Center+ Maastricht The Netherlands. Department of Precision Medicine GROW School for Oncology and Reproduction Maastricht University Center+ Maastricht The Netherlands. Computational Pathology Oncology R&D AstraZeneca Munich Germany. Department of Surgery Kanagawa Cancer Center Yokohama Japan. Department of Surgery Yokohama City University Yokohama Japan. Department of Surgery Tokyo Metropolitan Cancer and Infectious Diseases Center Komagome Hospital Tokyo Japan. Department of Pathology Tokyo Metropolitan Institute for Geriatrics and Gerontology Tokyo Japan. Department of Medicine National University of Singapore Singapore Singapore. Department of Pathology National University Hospital Singapore Singapore. Genome Institute of Singapore Agency for Science Technology and Research Singapore Singapore. Cancer Science Institute of Singapore National University of Singapore Singapore Singapore. Cellular and Molecular Research National Cancer Centre Singapore Singapore. Singhealth/Duke-NUS Institute of Precision Medicine National Heart Centre Singapore

来源：评论

学校读者我要写书评

暂无评论

Optimizing Postneoadjuvant Treatment of Residual Breast cancer With Adjuvant Bevacizumab Alone, With Metronomic or Standard-Dose Chemotherapy: A Combined Analysis of DFCI 05-055 and DFCI 09-134/TBCRC 012/ABCDE Clinical Trials

引用

Clinical Breast cancer 2025年第4期25卷 e419-e430.e5页

作者： Trapani, Dario Jin, Qingchun Miller, Kathy D. Rugo, Hope S. Reeder-Hayes, Katherine E. Traina, Tiffany Abdou, Yara Falkson, Carla Abramson, Vandana Ligibel, Jennifer Chen, Wendy Come, Steven Nohria, Anju Ryabin, Nicole Tayob, Nabihah Tolaney, Sara M Burstein, Harold J. Mayer, Erica L. Medical Oncology Dana-Farber Cancer Institute Boston MA United States Breast Oncology Program Dana-Farber Brigham Cancer Center Boston MA United States Harvard Medical School Boston MA United States Data Science Dana-Farber Cancer Institute Boston MA United States Indiana University Melvin and Bren Simon Comprehensive Cancer Center Indianapolis IN United States University of California at San Francisco San Francisco CA United States University of North Carolina Lineberger Comprehensive Cancer Institute Chapel Hill NC United States Memorial Sloan Kettering Cancer Center New York NY United States University of Alabama Birmingham AL United States Vanderbilt University Nashville TN United States Beth Israel Deaconess Medical Center Boston MA United States Division of Cardiovascular Medicine Brigham and Women's Hospital Boston MA United States

Background: Breast cancer patients with residual disease after neoadjuvant therapy have increased risk of recurrence. Novel therapies to decrease this risk are urgently needed. Methods: Two clinical trials (05-055 and 09-134) offered adjuvant bevacizumab-based therapy to stage I–III breast cancer patients with residual disease after neoadjuvant chemotherapy. Study 05-055 evaluated four treatment regimens: bevacizumab (cohort A);bevacizumab with metronomic cyclophosphamide and methotrexate (CM) (cohort B);and bevacizumab with body surface area-dosed capecitabine (cohorts C);or flat-dosed capecitabine (cohort D). The primary endpoint was feasibility and tolerability. In 09-134, patients were randomized to bevacizumab with or without CM;the primary endpoint was recurrence-free survival (RFS). Study 09-134 closed prematurely for lack of accrual. A pooled survival analysis with participants from 05-055 and 09-134 was conducted. Results: Among 213 total patients (05-055, n = 163;09-134, n = 50), the most common adverse events (AEs) of any grade were headache (49.3%) and fatigue (57.3%). Grade 3-4 AEs were highest in cohorts C (71.4%) and D (72.5%). The 36-month RFS was 58.0% with bevacizumab monotherapy, 62.3% with bevacizumab plus CM, and 72.7%-75.0% with bevacizumab plus capecitabine (depending on schedule). Treatment with capecitabine was independently associated with improved RFS in triple-negative breast cancer (TNBC) (HR: 0.47;95% CI, 0.23-0.96). Conclusion: This pooled analysis demonstrates that postneoadjuvant bevacizumab plus capecitabine may be associated with improved RFS, especially in TNBC. Each regimen carries moderate toxicity, and despite these treatments, patients with residual disease after neoadjuvant therapy still experience high rates of recurrence, indicating that new strategies are warranted. Clinical Trial Registration: ***, NCT00121134 (DFCI Protocol Number: 05-055);NCT00925652 (DFCI Protocol Number: 09-134). © 2025 Elsevier Inc.

关键词： Anti-angiogenesis Capecitabine Preoperative TNBC VEGF

来源：评论

学校读者我要写书评

暂无评论

Macrophages direct location-dependent recall of B cell memory to vaccination

引用

Cell 2025年

作者： Dhenni, Rama Hoppé, Alexandra Carey Reynaldi, Arnold Kyaw, Wunna Handoko, Nathalie Tricia Grootveld, Abigail K. Keith, Yuki Honda Bhattacharyya, Nayan Deger Ahel, Holly I. Telfser, Aiden Josiah McCorkindale, Andrew N. Yazar, Seyhan Bui, Christina H.T. Smith, James T. Khoo, Weng Hua Boyd, Mollie Obeid, Solange Milner, Brad Starr, Mitchell Brilot, Fabienne Milogiannakis, Vanessa Akerman, Anouschka Aggarwal, Anupriya Davenport, Miles P. Deenick, Elissa K. Chaffer, Christine L. Croucher, Peter I. Brink, Robert Goldstein, Leonard D. Cromer, Deborah Turville, Stuart G. Kelleher, Anthony D. Venturi, Vanessa Munier, C. Mee Ling Phan, Tri Giang Precision Immunology Program Garvan Institute of Medical Research Sydney NSW Australia St. Vincent's Healthcare Clinical Campus School of Clinical Medicine Faculty of Medicine and Health UNSW Sydney Kensington Sydney NSW Australia Immunovirology and Pathogenesis Program Kirby Institute UNSW Sydney Sydney NSW Australia Infection Analytics Program Kirby Institute UNSW Sydney Sydney NSW Australia Data Science Platform Garvan Institute of Medical Research Sydney 2010 NSW Australia St. Vincent's Hospital Sydney Sydney NSW Australia Cancer Plasticity and Dormancy Program The Kinghorn Cancer Centre Garvan Institute of Medical Research Sydney 2010 NSW Australia St. Vincent's Centre for Applied Medical Research Sydney NSW Australia Immune Biotherapies Program Garvan Institute of Medical Research Sydney 2010 NSW Australia Brain Autoimmunity Group Kids Neuroscience Centre The Children's Hospital at Westmead Faculty of Medicine and Health School of Medical Sciences Sydney NSW Australia School of Medical Science Faculty of Medicine and Health University of Sydney Sydney NSW Australia

Vaccines generate long-lived plasma cells and memory B cells (Bmems) that may re-enter secondary germinal centers (GCs) to further mutate their B cell receptor upon boosting and re-exposure to antigen. We show in mouse models that lymph nodes draining the site of primary vaccination harbor a subset of Bmems that reside in the subcapsular niche, generate larger recall responses, and are more likely to re-enter GCs compared with circulating Bmems in non-draining lymph nodes. This location-dependent recall of Bmems into the GC in the draining lymph node was dependent on CD169⁺ subcapsular sinus macrophages (SSMs) in the subcapsular niche. In human participants, boosting of the BNT162b2 vaccine in the same arm generated more rapid secretion of broadly neutralizing antibodies, GC participation, and clonal expansion of SARS-CoV-2-specific B cells than boosting of the opposite arm. These data reveal an unappreciated role for primed draining lymph node SSMs in Bmem cell fate determination. © 2025 The Author(s)

关键词： affinity maturation broadly neutralizing antibodies germinal center immune imprinting memory original antigenic sin plasma cells SARS-CoV-2 subcapsular sinus macrophage vaccine

来源：评论

学校读者我要写书评

暂无评论

Biomedical data and AI

引用

science China Life sciences 2025年第5期68卷 1536-1540页

作者： Hao Xu Shibo Zhou Zefeng Zhu Vincenzo Vitelli Liangyi Chen Ziwei Dai Ning Yang Luhua Lai Shengyong Yang Sergey Ovchinnikov Zhuoran Qiao Sirui Liu Chen Song Jianfeng Pei Han Wen Jianfeng Feng Yaoyao Zhang Zhengwei Xie Yang-Yu Liu Zhiyuan Li Fulai Jin Hao Li Mohammad Lotfollahi Xuegong Zhang Ge Yang Shihua Zhang Ge Gao Pulin Li Qi Liu Jing-Dong Jackie Han Peking-Tsinghua Center for Life Sciences (CLS) Academy for Advanced Interdisciplinary StudiesPeking University Center for Quantitative Biology (CQB) Academy for Advanced Interdisciplinary StudiesPeking University Peking University-Tsinghua University-National Institute of Biological Sciences Joint Graduate Program Academy for Advanced Interdisciplinary StudiesPeking University Department of Physics University of Chicago School of Life Sciences Southern University of Science and Technology Peking University Chengdu Academy for Advanced Interdisciplinary Biotechnologies College of Chemistry and Molecular Engineering Peking University Department of Biotherapy Cancer Center and State Key Laboratory of BiotherapyWest China HospitalSichuan University Department of Biology Massachusetts Institute of Technology Lambic Therapeutics Inc. Changping Laboratory Al for Science Institute Key Laboratory of Computational Neuroscience and Brain-Inspired Intelligence Fudan University Department of Obstetrics and Gynecology West China Second University HospitalSichuan University Peking University International Cancer Institute and Peking University-Yunnan Baiyao International Medical Institute and State Key Laboratory of Natural and Biomimetic Drugs Department of Molecular and Cellular PharmacologySchool of Pharmaceutical SciencesPeking University Health Science CenterPeking University Channing Division of Network Medicine Department of MedicineBrigham and Women's Hospital and Harvard Medical School Center for Artificial Intelligence and Modeling the Carl R.Woese Institute for Genomic BiologyUniversity of Illinois Urbana-Champaign Department of Genetics and Genome Sciences School of Medicine and Department of Computer and Data Sciences and Department of Population and Quantitative Health SciencesCase Western Reserve University Department of Biochemistry and Biophysics University of California Sanger Institute Department of Automation Tsinghua University State Key Laboratory of Multimodal Artificial Intelligence Systems I

The development of artificial intelligence(AI) and the mining of biomedical data complement each other. From the direct use of computer vision results to analyze medical images for disease screening, to now integrating biological knowledge into models and even accelerating the development of new AI based on biological discoveries, the boundaries of both are constantly expanding, and their connections are becoming closer.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ASO Visual Abstract: Impact of the American Society of Breast Surgeons' Guidelines on Genetic Testing and Contralateral Prophylactic Mastectomy Rates

引用

Annals of surgical oncology 2025年 2025 May 12页

作者： Anna Weiss Maria Sol Rosito Danielle Braun Brenna Barton Monica McGrath Sam Stokes Alison Laws Laura Warren Stefania Morganti Filipa Lynce Brittany Bychkovsky Huma Q Rana Dillon Davis Jill Stopfer Judy E Garber Tari A King Division of Breast Surgery Department of Surgery Brigham and Women's Hospital Boston MA USA. anna_weiss@urmc.rochester.edu. Breast Oncology Program Dana-Farber Brigham Cancer Center Boston MA USA. anna_weiss@urmc.rochester.edu. Harvard Medical School Boston MA USA. anna_weiss@urmc.rochester.edu. Division of Surgical Oncology Department of Surgery University of Rochester Rochester NY USA. anna_weiss@urmc.rochester.edu. Department of Data Science Dana-Farber Cancer Institute Boston MA USA. Department of Biostatistics Harvard T.H. Chan School of Public Health Boston MA USA. Department of Emergency Medicine Beth Israel Deaconess Medical Center Boston MA USA. Division of Breast Surgery Department of Surgery Brigham and Women's Hospital Boston MA USA. Division of Cancer Genetics and Prevention Department of Medical Oncology Dana-Farber Cancer Institute Boston MA USA. Breast Oncology Program Dana-Farber Brigham Cancer Center Boston MA USA. Harvard Medical School Boston MA USA. Departments of Surgery and Oncology University of Calgary Calgary AB Canada. Department of Radiation Oncology Brigham and Women's Hospital Boston MA USA. Medical Oncology Dana-Farber Cancer Institute Boston MA USA.

来源：评论

学校读者我要写书评

暂无评论

Molecular-driven Foundation Model for Oncologic Pathology

arXiv

引用

arXiv 2025年

作者： Vaidya, Anurag Zhang, Andrew Jaume, Guillaume Song, Andrew H. Ding, Tong Wagner, Sophia J. Lu, Ming Y. Doucet, Paul Robertson, Harry Almagro-Pérez, Cristina Chen, Richard J. ElHarouni, Dina Ayoub, Georges Bossi, Connor Ligon, Keith L. Gerber, Georg Le, Long Phi Mahmood, Faisal Department of Pathology Brigham and Women’s Hospital Harvard Medical School BostonMA United States Department of Pathology Massachusetts General Hospital Harvard Medical School BostonMA United States Cancer Program Broad Institute of Harvard and MIT CambridgeMA United States Health Sciences and Technology Harvard-MIT CambridgeMA United States Harvard John A. Paulson School of Engineering and Applied Sciences Harvard University CambridgeMA United States Helmholtz Munich – German Research Center for Environment and Health Munich Germany School of Computation Information and Technology TUM Munich Germany CambridgeMA United States Sydney Precision Data Science Center The University of Sydney CamperdownNSW Australia Department of Oncologic Pathology Dana-Farber Cancer Institute BostonMA02215 United States Department of Pathology Boston Children’s Hospital BostonMA02115 United States Harvard Data Science Initiative Harvard University CambridgeMA United States

Foundation models are reshaping computational pathology by enabling transfer learning, where models pre-trained on vast datasets can be adapted for downstream diagnostic, prognostic, and therapeutic response tasks. Despite these advances, foundation models are still limited in their ability to encode the entire gigapixel whole-slide images without additional training and often lack complementary multimodal data. Here, we introduce THREADS, a slide-level foundation model capable of generating universal representations of whole-slide images of any size. THREADS was pretrained using a multimodal learning approach on a diverse cohort of 47,171 hematoxylin and eosin (H&E)-stained tissue sections, paired with corresponding genomic and transcriptomic profiles—the largest such paired dataset to be used for foundation model development to date. This unique training paradigm enables THREADS to capture the tissue’s underlying molecular composition, yielding powerful representations applicable to a wide array of downstream tasks. In extensive benchmarking across 54 oncology tasks, including clinical subtyping, grading, mutation prediction, immunohistochemistry status determination, treatment response prediction and survival prediction THREADS outperformed all baselines while demonstrating remarkable generalizability and label efficiency. It is particularly well-suited for predicting rare events, further emphasizing its clinical utility. We intend to make the model publicly available for the broader community. © 2025, CC BY-NC-ND.

关键词： Transfer learning

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：