Deep Generative Classification of Blood Cell Morphology
University of Cambridge（剑桥大学）；UZ Leuven（鲁汶大学医院）；University of Leuven（鲁汶大学）；KU Leuven（鲁汶大学）；Zuyderland Medical Center（宰德兰医学中心）；Barts Health NHS Trust（巴茨健康国民保健服务信托）；University College London（伦敦大学学院）；Queen Mary University of London（伦敦玛丽女王大学）
Simon Deltadahl, Julian Gilbey, Christine Van Laer, Nancy Boeckx, Mathie Leers, Tanya Freeman, Laura Aiken, Timothy Farren, Matthew Smith, Mohamad Zeina, BloodC… 展开作者
Simon Deltadahl, Julian Gilbey, Christine Van Laer, Nancy Boeckx, Mathie Leers, Tanya Freeman, Laura Aiken, Timothy Farren, Matthew Smith, Mohamad Zeina, BloodCounts consortium, James HF Rudd, Concetta Piazzese, Joseph Taylor, Nicholas Gleadall, Carola-Bibiane Schönlieb, Suthesh Sivapalaratnam, Michael Roberts, Parashkev Nachev
Accurate classification of haematological cells is critical for diagnosing blood disorders, but presents significant challenges for machine automation owing to the complexity of cell morphology, heterogeneities of biological, pathological, and imaging characteristics, and the imbalance of cell type frequencies. We introduce CytoDiffusion, a diffusion-based classifier that effectively models blood cell morphology, combining accurate classification with robust anomaly detection, resistance to distributional shifts, interpretability, data efficiency, and superhuman uncertainty quantification. Our approach outperforms state-of-the-art discriminative models in anomaly detection (AUC 0.990 vs. 0.918), resistance to domain shifts (85.85% vs. 74.38% balanced accuracy), and performance in low-data regimes (95.88% vs. 94.95% balanced accuracy). Notably, our model generates synthetic blood cell images that are nearly indistinguishable from real images, as demonstrated by an authenticity test in which expert haematologists achieved only 52.3% accuracy (95% CI: [50.5%, 54.2%]) in distinguishing real from generated images. Furthermore, we enhance model explainability through the generation of directly interpretable counterfactual heatmaps. Our comprehensive evaluation framework, encompassing these multiple performance dimensions, establishes a new benchmark for medical image analysis in haematology, ultimately enabling improved diagnostic accuracy in clinical settings. Our code is available at https://github.com/CambridgeCIA/CytoDiffusion.
Language-Enhanced Generative Modeling for Amyloid PET Synthesis from MRI and Blood Biomarkers
organization= Shanghai Artificial Intelligence Laboratory , city= Shanghai , postcode= 200082 , country= China；organization= School of Medicine, Xiamen University , city= Xiamen , state= Fujian , country= China；organization= Department of Nuclear Medicine \& PET Center, Huashan Hospital, Fudan University , city= Shanghai , country= China；organization= Department of Gerontology, Shanghai Jiao Tong University Affiliated Sixth People’s Hospital , city= Shanghai , country= China；organization= Department of Computer Science, Rutgers University , city= New Brunswick , state= New Jersey , country= United States；organization= Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences , city= Shenzhen , country= China
A Labeled Ophthalmic Ultrasound Dataset with Medical Report Generation Based on Cross-modal Deep Learning
School of Electrical and Control Engineering, North China University of Technology（北方工业大学电气与控制工程学院）；School of Automation and Electrical Engineering, Shenyang Ligong University（沈阳理工大学自动化与电气工程学院）；The Fourth Affiliated Hospital of China Medical University（中国医科大学附属第四医院）
Classification of White Blood Cells Using Machine and Deep Learning Models: A Systematic Review
University of Lahore（拉合尔大学）；University of Limerick（利默里克大学）；Dublin Institute of Technology（都柏林理工学院）；National University of Sciences and Technology（国家科技大学）
Hierarchical Part-based Generative Model for Realistic 3D Blood Vessel
Shenzhen Key Laboratory of Ubiquitous Data Enabling（Ubiquitous Data Enabling 深圳重点实验室）；Tsinghua Shenzhen International Graduate School（清华大学深圳国际研究生院）；Tsinghua University（清华大学）；Pengcheng Laboratory（Pengcheng 实验室）；School of Life Science（生命科学学院）；the Key Laboratory of Convergence Medical Engineering System（融合医学工程系统重点实验室）；Information Technology, Beijing Institute of Technology（信息 technology，北京理工大学）；Institute of Medical Artificial Intelligence（医疗人工智能研究所）；the Second Affiliated Hospital of Xi'an Jiaotong University（西安交通大学第二附属医院）；MoE Key Laboratory of Brain-inspired Intelligent Perception（教育部脑启发智能感知与认知重点实验室）；Cognition, University of Science（认知，科学大学）
Sickle-cell disease diagnosis support selecting the most appropriate machinelearning method: Towards a general and interpretable approach for cellmorphology analysis from microscopy images
University of the Balearic Islands（巴利阿里群岛大学）；Health Research Institute of the Balearic Islands (IdISBa)（巴利阿里群岛健康研究所）
A Morphology Focused Diffusion Probabilistic Model for Synthesis of Histopathology Images
University of British Columbia（不列颠哥伦比亚大学）；Eindhoven University of Technology（埃因霍温理工大学）；Massachusetts General Hospital（马萨诸塞州综合医院）；Harvard Medical School（哈佛医学院）；BC Cancer Agency（不列颠哥伦比亚癌症机构）
Human Treelike Tubular Structure Segmentation: A Comprehensive Review and Future Perspectives
National Heart and Lung Institute, Faculty of Medicine, Imperial College London（帝国理工学院医学院国家心肺研究所）；Department of Bioengineering, Faculty of Engineering, Imperial College London（帝国理工学院工程系生物工程系）；Royal Brompton Hospital（皇家布朗普顿医院）
Novel Radiomic Measurements of Tumor- Associated Vasculature Morphology on Clinical Imaging as a Biomarker of Treatment Response in Multiple Cancers
Case Western Reserve University（凯斯西储大学）；Picture Health（Picture Health公司）；Stony Brook University（石溪大学）；University Hospitals Cleveland Medical Center（克利夫兰大学医院医学中心）；Yale School of Medicine（耶鲁医学院）；The Cleveland Clinic Foundation (CCF)（克利夫兰诊所基金会（CCF））；New York University (NYU) Langone Medical Center（纽约大学（NYU）朗格尼医学中心）；Louis Stokes Cleveland Veterans Medical Center（路易斯·斯托克斯克利夫兰退伍军人医学中心）
目的：肿瘤相关血管系统与健康血管的不同之处在于其混乱的结构和扭曲性，这会促进治疗抵抗。这些属性的可测量差异可能有助于根据全身治疗（例如化疗）的可能获益对患者进行分层。在这项工作中，我们提出了一类新的放射性生物标志物，称为定量肿瘤相关血管系统（QuanTAV）特征，并证明了它们预测多种癌症，成像方式和治疗方案的反应和生存的能力。实验设计：我们对558名患者的常规治疗前放射学（CT或对比增强MRI）进行了肿瘤血管分割并计算了扭曲和组织的数学测量值，这些患者接受了乳腺癌（n=371）或非小细胞肺癌（NSCLC，n=187）的四种一线化疗治疗干预策略之一。结果如下：在4种基于化疗的治疗策略中，QuanTAV测量的分类器显著（p<0.05）预测了单独保持测试队列的反应（AUC=0.63-0.71），并且当添加到单独的重要临床变量模型中时，AUC增加了0.06-0.12。QuanTAV风险评分是乳腺癌化疗（p=0.002，HR=1.25，95% CI 1.08-1.44，C指数= 0.66）和NSCLC放化疗（p=0.039，HR=1.28，95% CI 1.01-1.62，C指数=0.66）治疗队列中无复发生存期的预后指标。在所有治疗组中，包括接受化疗的NSCLC患者（p=0.034，HR=2.29，95% CI 1.07-4.94，C指数=0.62），分类QuanTAV风险组具有独立预后。…
Recursive Variational Autoencoders for 3D Blood Vessel Generative Modeling
Consejo Nacional de Investigaciones Científicas y Técnicas（阿根廷国家科学与技术研究院）；Universidad Torcuato Di Tella（托科托迪塔拉大学）；Depto. Ing. Eléctrica y Computadoras, Universidad Nacional del Sur（国家南部大学电子工程与计算机系）
Synergizing Deep Learning and Biological Heuristics for Extreme Long-Tail White Blood Cell Classification
VinUni-Illinois Smart Health Center, VinUniversity（VinUniversity-VinUni-Illinois智能健康中心）；The Computer Vision and Medical AI Lab, VinUniversity（VinUniversity计算机视觉与医学人工智能实验室）
X-GAN: A Generative AI-Powered Unsupervised Model for Main Vessel Segmentation of Glaucoma Screening
Southern Methodist University（南方 Methodist 大学）；National Aeronautics and Space Administration（美国国家航空航天局）；University of Texas Southwestern Medical Center（德克萨斯西南医学中心）；Northeastern University（东北大学）
Automatic Classification of Circulating Blood Cell Clusters based on Multi-channel Flow Cytometry Imaging
School of Chemical, Materials, and Biomedical Engineering, University of Georgia（化学、材料与生物医学工程学院，佐治亚大学）；Institute for Artificial Intelligence, University of Georgia（人工智能研究所，佐治亚大学）；Pathology Core, Unit for Laboratory Animal Medicine, University of Michigan（病理核心，实验室动物医学单位，密歇根大学）；School of Environmental, Civil, Agricultural and Mechanical Engineering, College of Engineering, University of Georgia（环境、土木、农业和机械工程学院，工程学院，佐治亚大学）；College of Public Health, University of Georgia, Department of Epidemiology & Biostatistics（公共健康学院，佐治亚大学，流行病学与生物统计学系）；Department of Data Science, Ying Wu College of Computing, New Jersey Institute of Technology（数据科学系，英伟学院计算机学院，新泽西理工学院）；McGovern Institute for Brain Research, Massachusetts Institute of Technology（麦戈文脑研究所，麻省理工学院）；Division of Trauma, Emergency Surgery, and Surgical Critical Care at the Massachusetts General Hospital（麻省总医院创伤、急诊手术和外科重症护理部）；University of Michigan（密歇根大学）
Multimodal Analysis of White Blood Cell Differentiation in Acute Myeloid Leukemia Patients using a β-Variational Autoencoder
Institute of AI for Health, Helmholtz Zentrum München – German Research Center for Environmental Health（慕尼黑亥姆霍兹中心人工智能健康研究所（德国环境健康研究中心））；University of Trento（特伦托大学）；Technical University of Munich（慕尼黑工业大学）；Ludwig-Maximilians University Munich（慕尼黑大学）；Johns Hopkins University（约翰斯·霍普金斯大学）
TRIDENT: A Trimodal Cascade Generative Framework for Drug and RNA-Conditioned Cellular Morphology Synthesis
Department of Big Data and Biomedical AI, College of Future Technology, Peking University（大数据与生物医学人工智能系，未来技术学院，北京大学）；Center for BioMed-X Research, Academy for Advanced Interdisciplinary Studies, Peking University（BioMed-X研究中心，先进跨学科研究学院，北京大学）；State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University（多媒体信息处理国家重点实验室，计算机科学学院，北京大学）；National Engineering Research Center of Visual Technology, School of Computer Science, Peking University（视觉技术国家工程研究中心，计算机科学学院，北京大学）；Yuanpei College, Peking University（元培学院，北京大学）；School of Life Sciences, Tsinghua University（生命科学学院，清华大学）；Peking University-Tsinghua University-National Institute of Biological Sciences Joint Graduate Program (PTN), Tsinghua University（北京大学-清华大学-国家生物医学科学研究院联合研究生项目（PTN），清华大学）
Domain-incremental White Blood Cell Classification with Privacy-aware Continual Learning
University of Regensburg（雷根斯堡大学）；University Hospital RWTH Aachen（亚琛工业大学附属医院）；University Hospital Erlangen（埃尔朗根大学附属医院）；Fraunhofer Institute for Digital Medicine MEVIS（弗劳恩霍夫数字医学梅维斯研究所）
Fundus to Fluorescein Angiography Video Generation as a Retinal Generative Foundation Model
The Hong Kong Polytechnic University（香港理工大学）；Swiss Federal Institute of Technology Lausanne (EPFL)（瑞士洛桑联邦理工学院）；Clemson University（克莱姆森大学）；University of Science and Technology of China（中国科学技术大学）；Research Centre for SHARP Vision (RSCV), The Hong Kong Polytechnic University（香港理工大学SHARP视觉研究中心）；Centre for Eye and Vision Research (CEVR)（眼科视觉研究中心）
Enhancing Retinal Vessel Segmentation Generalization via Layout-Aware Generative Modelling
Faculty of Biomedical Engineering, Technion-IIT（以色列理工学院生物医学工程学院）；Faculty of Mathematics, Technion-IIT（以色列理工学院数学学院）；KU Leuven（鲁汶大学）；UZ Leuven（鲁汶大学医院）；São Paulo Federal University（圣保罗联邦大学）；Massachusetts Institute of Technology（麻省理工学院）；School of Electrical and Computer Engineering, Ben-Gurion University of the Negev（本-古里安大学内盖夫分校电气与计算机工程学院）
SPATIA: Multimodal Generation and Prediction of Spatial Cell Phenotypes
Department of Biomedical Informatics, Harvard Medical School（哈佛医学学校生物医学信息学系）；Department of Computer Science, University of North Carolina（北卡罗来纳大学计算机科学系）；Department of Computer Science, Massachusetts Institute of Technology（麻省理工学院计算机科学系）
StrCGAN: A Generative Framework for Stellar Image Restoration
Department of Information and Communication Technology, Marwadi University, India（信息与通信技术系，马瓦迪大学，印度）；Department of Physics and Astronomy, Heidelberg University, Heidelberg, Germany（物理与天文学系，海德堡大学，德国）
Plasma-CycleGAN: Plasma Biomarker-Guided MRI to PET Cross-modality Translation Using Conditional CycleGAN
School of Computing and Augmented Intelligence, Arizona State University（亚利桑那州立大学计算与增强智能学院）；Banner Alzheimer’s Institute（班纳阿尔茨海默病研究所）；Basis Scottsdale（斯科茨代尔贝斯机构）；College of Health Solutions, Arizona State University（亚利桑那州立大学健康解决方案学院）；Mayo Clinic Arizona（梅奥诊所亚利桑那州分院）
Photoacoustic image synthesis with generative adversarial networks
German Cancer Research Center (DKFZ)（德国癌症研究中心）；Heidelberg University（海德堡大学）；Helmholtz Information and Data Science School for Health（亥姆霍兹健康信息与数据科学学院）；Cancer Research UK Cambridge Institute, University of Cambridge（剑桥大学英国癌症研究中心剑桥研究所）
