Nature Biomedical Engineering最新文献

{"title":"Deep-learning triage of three-dimensional pathology datasets for comprehensive and efficient pathologist assessments.","authors":"Gan Gao, Renao Yan, Andrew H Song, Huai-Ching Hsieh, Lindsey A Erion Barner, Fiona Wang, David Brenes, Sarah S L Chow, Rui Wang, Kevin W Bishop, Yongjun Liu, Xavier Farre, Mukul Divatia, Michelle R Downes, Funda Vakar-Lopez, Priti Lal, Wynn Burke, Anant Madabhushi, Lawrence D True, Deepti M Reddi, William M Grady, Faisal Mahmood, Jonathan T C Liu","doi":"10.1038/s41551-026-01760-1","DOIUrl":"10.1038/s41551-026-01760-1","url":null,"abstract":"<p><p>Standard slide-based two-dimensional (2D) histopathology severely undersamples spatially heterogeneous tissue, with each thin 2D section representing <1% of the entire biopsy volume. Recent advances in non-destructive three-dimensional (3D) pathology, such as open-top light-sheet microscopy, enable comprehensive high-resolution imaging of large clinical specimens. Since manual review of these massive and complex 3D datasets is infeasible in clinical practice, we present TRICARE, a deep-learning triage framework that identifies high-risk 2D cross sections within 3D pathology datasets to enable time-efficient pathologist evaluation, which offers a lower-risk route for accelerated adoption by retaining pathologists for final diagnoses. TRICARE assigns risk scores to all 2D levels within a tissue volume by leveraging context from a subset of neighbouring depth levels, outperforming models in which predictions are based on isolated 2D levels. In two use cases-risk stratification based on prostate cancer biopsies and screening for dysplasia/cancer in endoscopic biopsies of Barrett's esophagus-AI-triaged 3D pathology, enabled by TRICARE, demonstrates the potential to improve the detection of high-risk diseases compared with slide-based 2D histopathology while optimizing pathologist workloads.</p>","PeriodicalId":19063,"journal":{"name":"Nature Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":26.3,"publicationDate":"2026-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148724989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Foundation models in biomedical imaging: turning hype into reality","authors":"Amgad Muneer,&nbsp;Kai Zhang,&nbsp;Ibraheem Hamdi,&nbsp;Rizwan Qureshi,&nbsp;Muhammad Waqas,&nbsp;Shereen Fouad,&nbsp;Hazrat Ali,&nbsp;Syed Muhammad Anwar,&nbsp;Jia Wu","doi":"10.1038/s41551-026-01762-z","DOIUrl":"10.1038/s41551-026-01762-z","url":null,"abstract":"Foundation models (FMs) are driving a prominent shift in biomedical imaging, from task-specific models to unified backbone models for diverse tasks. This opens an avenue to integrate imaging, pathology, clinical records and genomics data into a composite system. However, this vision contrasts sharply with modern medicine’s trajectory towards more granular sub-specialization. This tension, coupled with data scarcity, domain heterogeneity and limited interpretability, creates a gap between benchmark success and real-world clinical value. We argue that the immediate role of FMs lies in augmenting, not replacing, clinical expertise. To separate hype from reality, we introduce real-world evaluation and assessment of FMs (REAL-FM), a multi-dimensional framework assessing data, technical readiness, clinical value, workflow integration and responsible artificial intelligence. Using REAL-FM, we find that although FMs excel in pattern recognition they fall short on causal reasoning, domain robustness and safety. Clinical translation is hindered by scarce representative data for model training, unverified generalization beyond over-simplified benchmark settings and a lack of prospective outcome-based validation. This Perspective provides clinicians with a practical way to interpret FM claims, identify where these systems may safely support imaging workflows and recognize why human oversight remains indispensable. For developers, it defines the validation, workflow, safety and governance requirements that must be met before FMs can become clinically reliable tools. We envision that the path forward lies not in a monolithic medical oracle, but in coordinated subspecialist AI systems that are transparent, safe and clinically grounded. This Perspective assesses the role of foundation models as clinical tools and introduces real-world evaluation and assessment of foundation models (REAL-FM), a framework for assessing data, technical readiness and clinical value towards responsible artificial intelligence.","PeriodicalId":19063,"journal":{"name":"Nature Biomedical Engineering","volume":"10 8","pages":"1557-1575"},"PeriodicalIF":26.3,"publicationDate":"2026-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148754332","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Author Correction: The evolving role of cytokines for CAR-T cell manufacturing and beyond","authors":"Mar Guaza-Lasheras,&nbsp;Johanna Nimmerfroh,&nbsp;Dominic Schwarz,&nbsp;Tanja D. de Gruijl,&nbsp;Jessica Hartmann,&nbsp;Christian Klein,&nbsp;Heinz Läubli,&nbsp;Michal Lotem,&nbsp;Joerg Mittelstaet,&nbsp;Christophe Paget,&nbsp;Sine Reker Hadrup,&nbsp;Sonja Schallenberg,&nbsp;Per Thor Straten,&nbsp;Sebastian Theurich,&nbsp;Dario Venetz,&nbsp;Alfred Zippelius,&nbsp;Laurence Zitvogel,&nbsp;Sebastian Kobold","doi":"10.1038/s41551-026-01783-8","DOIUrl":"10.1038/s41551-026-01783-8","url":null,"abstract":"","PeriodicalId":19063,"journal":{"name":"Nature Biomedical Engineering","volume":"10 9","pages":"1977-1977"},"PeriodicalIF":26.3,"publicationDate":"2026-08-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.comhttps://www.nature.com/articles/s41551-026-01783-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148706684","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Causal graph neural networks for healthcare","authors":"Munib Mesinovic,&nbsp;Max Buhlan,&nbsp;Tingting Zhu","doi":"10.1038/s41551-026-01742-3","DOIUrl":"10.1038/s41551-026-01742-3","url":null,"abstract":"Healthcare artificial intelligence systems often degrade in performance when deployed across institutions, with documented performance drops and perpetuation of discriminatory patterns embedded in data. This brittleness comes, in part, from learning statistical associations rather than causal mechanisms. Causal graph neural networks address this by combining graph-based representations of biomedical data with causal inference to learn invariant mechanisms instead of just spurious correlations. This Perspective reviews the methodology of structural causal models, disentangled causal representation learning, and techniques for interventional prediction and counterfactual reasoning on graphs. We discuss applications across psychiatric diagnosis and brain network analysis, cancer subtyping with multi-omics causal integration, continuous physiological monitoring and drug recommendations. These methods provide building blocks for patient-specific causal digital twins that could support in silico clinical experimentation. Remaining challenges include computational costs that preclude real-time deployment, validation challenges that go beyond standard cross-validation, and the risk of causal-washing where methods adopt causal terminology without rigorous evidentiary support. We propose a tiered framework distinguishing causally inspired architectures from causally validated discoveries and outline future directions, including scalable causal discovery, multimodal data integration and regulatory pathways for these methods. Making practical causal digital twins possible will require an honest assessment of what current methods deliver, sustained collaboration across disciplines and validation standards that match the strength of the causal claims being made. This Perspective explores and outlines how causal graph neural networks can impact the future of artificial intelligence in healthcare, from diagnostics to causal digital twins.","PeriodicalId":19063,"journal":{"name":"Nature Biomedical Engineering","volume":"10 8","pages":"1536-1556"},"PeriodicalIF":26.3,"publicationDate":"2026-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148689629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Engineered lipid nanoparticles for in vivo and durable editing of haematopoietic stem cells within humanized mice.","authors":"Jingjing Du, Zijin Luo, Dan Xie, Yan Chen, Ming Yang, Qiang Li, Lisha Wang, Lu Han, Yiming Zhang, Haiwei Li, Zilin Lan, Haoping Shi, Yinghui Li, Qiang Cheng, Fang Dong, Yingdai Gao, Yao Yao, Tao Cheng, Tuo Wei, Shuquan Rao","doi":"10.1038/s41551-026-01765-w","DOIUrl":"10.1038/s41551-026-01765-w","url":null,"abstract":"<p><p>In vivo genetic engineering of haematopoietic stem and progenitor cells (HSPCs) holds the potential to revolutionize the treatment landscape for numerous diseases. However, despite its transformative potential, it remains hindered by the difficulty in efficiently and specifically targeting quiescent human HSCs while maintaining their long-term functionality. Here, after screening 15 lipid nanoparticles (LNPs), we report an LNP that efficiently delivers reporter mRNA to human HSPCs both in ex vivo and in vivo settings when conjugated with the anti-CD34 antibody (CD34/LNP^DP). Using CRISPR/Cas editing cargos, CD34/LNP^DP achieves high editing efficiency in human HSPCs ex vivo. Intrafemoral administration of CD34/LNP^DP in humanized mice results in efficient editing of the erythroid-specific BCL11A enhancer within human HSPCs, enabling the sustained long-term reactivation of fetal haemoglobin (HbF) expression in erythroid cells. In a humanized neutropaenia model harbouring an ELANE mutation, intrafemoral administration of CD34/LNP^DP achieves robust editing, targeting exon 2 of ELANE in human HSPCs, partially restoring neutrophil development impairment under long-term observation. Collectively, CD34-targeted delivery enables in vivo HSPC modification without perturbing haematopoiesis, underscoring its suitability for clinical translation.</p>","PeriodicalId":19063,"journal":{"name":"Nature Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":26.3,"publicationDate":"2026-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148679444","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"XunZi, an AI biologist, reveals disease-modifying targets.","authors":"Xinhe Huang, Junhong Qin, Fei Tang, Chenyu Yang, Linhua Xu, Junfen Wei, Dan Liu, Kang Chen, Chi Zhang, Miaomiao Chen, Yujie Gou, Jiayi Zhang, Yongze Ma, Lin Zhao, Yingfeng Tu, Peng Lei, Yu Xue, Da Jia","doi":"10.1038/s41551-026-01769-6","DOIUrl":"https://doi.org/10.1038/s41551-026-01769-6","url":null,"abstract":"<p><p>Hypothesis generation in biomedicine is constrained by human cognitive limitations in synthesizing insights from fragmented biomedical knowledge and multimodal data sources. Here we introduce XunZi, an AI biologist that integrates logical reasoning and multimodal data fusion to autonomously generate de novo therapeutic target hypotheses with testable mechanisms. XunZi has been trained on 24.4 million publications and 613.6 TB of multisource data spanning 21,008 human genes and 5,850 diseases, and outperforms existing methods in both accuracy and interpretability across diverse disease contexts. In Parkinson's disease (PD), where complex mechanisms and limited targets hamper therapy development, XunZi identifies aberrant activation of CHK2 and IRAK4 kinases across multiple models. Pharmacological or genetic inhibition of Chk2 rescues dopaminergic neuron loss and motor deficits in PD mice. We further demonstrate XunZi's broad versatility in diseases such as non-small-cell lung cancer. XunZi establishes a paradigm-shifting framework to translate fragmented biomedical knowledge and data into actionable therapeutics.</p>","PeriodicalId":19063,"journal":{"name":"Nature Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":26.3,"publicationDate":"2026-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148674314","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"EU’s fragmented early-stage medical device study requirements need reform","authors":"Muhammad Ibtsaam Qadir,&nbsp;Juliane Wober,&nbsp;Ali McDonnell,&nbsp;Jana Russo,&nbsp;Tom Melvin,&nbsp;Stephen Gilbert,&nbsp;Fiona R. Kolbinger","doi":"10.1038/s41551-026-01750-3","DOIUrl":"10.1038/s41551-026-01750-3","url":null,"abstract":"Legislative change must fix the European Union’s disharmonized map of early-stage medical device clinical study requirements, whereby fragmented interpretations across Member States create barriers to innovation and delay patient access to life-saving technologies.","PeriodicalId":19063,"journal":{"name":"Nature Biomedical Engineering","volume":"10 9","pages":"1798-1801"},"PeriodicalIF":26.3,"publicationDate":"2026-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148648945","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Rebalancing bone remodelling in osteoporosis.","authors":"Shounak Roy, Akhilesh K Gaharwar","doi":"10.1038/s41551-026-01702-x","DOIUrl":"https://doi.org/10.1038/s41551-026-01702-x","url":null,"abstract":"","PeriodicalId":19063,"journal":{"name":"Nature Biomedical Engineering","volume":" ","pages":""},"PeriodicalIF":26.3,"publicationDate":"2026-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148631051","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sulfated polysaccharides target RANK Lys97 to inhibit osteoclast differentiation and reverse osteoporosis","authors":"Xiaogang Wang, Luli Ji, Yuanman Yu, Jiaze Yu, Jing Wang, Changsheng Liu","doi":"10.1038/s41551-026-01744-1","DOIUrl":"https://doi.org/10.1038/s41551-026-01744-1","url":null,"abstract":"Current osteoporosis treatments fail to balance bone resorption and formation. Here we show that engineered 2-N,6-O-sulfated chitosan (26SCS), a synthetic pentasaccharide, restores 115% bone mass (therapeutic) and prevents 66% bone loss, surpassing bisphosphonates by simultaneously suppressing osteoclastogenesis and promoting vascularized osteogenesis. Mechanistically, 26SCS targets K97 in receptor activator of nuclear factor-κB (RANK) through geometrically matched sulfate pairs, blocking RANK ligand (RANKL) signalling to arrest preosteoclast fusion while enhancing platelet-derived growth factor-BB (PDGF-BB) secretion via preserved preosteoclast viability to drive angiogenesis-coupled mineralization. Structural and functional analyses reveal that carboxyl groups in natural glycosaminoglycans such as heparin competitively bind K97, disrupting therapeutic specificity—a limitation overcome by 26SCS’s carboxyl-free design and sequence-controlled sulfation. Unlike monosulfated analogues (2SCS/6SCS) that oversuppress osteoclastogenesis or lack pro-osteogenic effects, 26SCS’s dual sulfation balances inhibition with trophic support. Thus, this work redefines glycosaminoglycan therapeutics via sulfation-patterned, topology-engineered biomaterials harmonizing bone metabolism.","PeriodicalId":19063,"journal":{"name":"Nature Biomedical Engineering","volume":"30 1","pages":""},"PeriodicalIF":28.1,"publicationDate":"2026-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148612619","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"In vivo CRISPR base editing for treatment of Huntington’s disease","authors":"Shraddha Shirguppe, Michael Gapinske, Devyani Swami, Kyrollos Shenouda, Angelo Miskalis, Nicholas Gosstola, Daniela Del Bosque Siller, Isabelle Guerra, Pankaj Acharya, Dana Joulani, Maddie G. Szkwarek, Ananthan Nambiar, Abhishek Bhattacharjee, Anagha S. Dangi, Norah Odle, Gabrielle E. Nathan, Gianna Elias, Michelle Stilger, Jackson Winter, Wendy S. Woods, Daphine Anand, Colin K. W. Lim, Sergei Maslov, Thomas Gaj, Pablo Perez-Pinera","doi":"10.1038/s41551-026-01747-y","DOIUrl":"https://doi.org/10.1038/s41551-026-01747-y","url":null,"abstract":"Huntington’s disease (HD) is a fatal neurodegenerative disorder caused by an expanded CAG repeat within exon 1 of the huntingtin (HTT) gene, resulting in a mutant protein that drives neuronal dysfunction and loss. A key event in the pathogenesis of HD is proteolytic cleavage of mutant HTT, which generates aggregation-prone N-terminal fragments that contribute to toxicity. Strategies that prevent this process thus hold therapeutic potential. Here we develop CRISPR base editors that generate proteolysis-resistant HTT isoforms by disrupting the splice acceptor of HTT exon 13, an exon that encodes critical proteolytic cleavage sites implicated in N-terminal fragment production. When delivered to the striatum of an HD rodent model, these editors reduced HTT fragment formation, decreased aggregation, improved functional deficits and attenuated brain atrophy. Collectively, these results demonstrate the potential of base editing and splice-site modulation to mitigate mutant HTT toxicity in HD.","PeriodicalId":19063,"journal":{"name":"Nature Biomedical Engineering","volume":"24 1","pages":""},"PeriodicalIF":28.1,"publicationDate":"2026-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148612622","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}