Multi-label Thoracic Disease Image Classification with Cross-Attention Networks
Singapore Management University（新加坡管理大学）；South China University of Technology（华南理工大学）；The University of Adelaide（阿德莱德大学）；Salesforce Research Asia（赛富时亚洲研究院）
Automated disease classification of radiology images has been emerging as a promising technique to support clinical diagnosis and treatment planning. Unlike generic image classification tasks, a real-world radiology image classification task is significantly more challenging as it is far more expensive to collect the training data where the labeled data is in nature multi-label; and more seriously samples from easy classes often dominate; training data is highly class-imbalanced problem exists in practice as well. To overcome these challenges, in this paper, we propose a novel scheme of Cross-Attention Networks (CAN) for automated thoracic disease classification from chest x-ray images, which can effectively excavate more meaningful representation from data to boost the performance through cross-attention by only image-level annotations. We also design a new loss function that beyond cross-entropy loss to help cross-attention process and is able to overcome the imbalance between classes and easy-dominated samples within each class. The proposed method achieves state-of-the-art results.
OXnet: Omni-supervised Thoracic Disease Detection from Chest X-rays
The Chinese University of Hong Kong（香港中文大学）；The Hong Kong University of Science and Technology（香港科技大学）；Imsight AI Research Lab（Imsight人工智能研究实验室）；Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences（中国科学院深圳先进技术研究院）
Anatomy-Structured Hierarchical MIL for Weakly-Supervised Thoracic Disease Detection in Chest X-rays
Ewha Womans University（梨花女子大学）；Ewha Medical Artificial Intelligence Research Institute（梨花女子大学医学人工智能研究所）；REMEDI Inc.（瑞美迪公司）；Ewha Womans University Seoul Hospital（梨花女子大学首尔医院）
From Classification to Localization and Clinical Validation: Large-Scale Development of a Deep Learning System for Thoracic Disease Detection on Chest Radiographs in Thailand
Perceptra Co., Ltd.（Perceptra有限公司）；Radiology Department, Faculty of Medicine Siriraj Hospital, Mahidol University（玛希隆大学诗里拉吉医院医学院放射科）；Institute of Field Robotics, King Mongkut’s University of Technology Thonburi（泰国国王科技大学吞武里分校野外机器人研究所）
Cross- and Intra-image Prototypical Learning for Multi-label Disease Diagnosis and Interpretation
Australian Institute for Machine Learning（澳大利亚机器学习研究所）；University of Adelaide（阿德莱德大学）；Cornell University（康奈尔大学）；St Vincent’s Hospital Melbourne（墨尔本圣文森特医院）；Centre for Vision, Speech and Signal Processing（视觉、语音与信号处理中心）；University of Surrey（萨里大学）
DeepVox and SAVE-CT: a contrast- and dose-independent 3D deep learning approach for thoracic aorta segmentation and aneurysm prediction using computed tomography scans
Albert Einstein Israeli Institute of Education and Research（阿尔伯特·爱因斯坦以色列教育与研究学院）；Health Innovation Techcenter（健康创新技术中心）；Department of Radiology（放射科）；Department of Vascular and Endovascular Surgery（血管与血管外科）
本研究提出 DeepVox 分割模型和 SAVE-CT 分类模型，在独立于对比剂和剂量的 3D CT 扫描上实现全自动胸主动脉瘤筛查，分割和分类测试集表现优异，有望降低死亡率并优化放射科评估队列。
Label uncertainty-guided multi-stream model for disease screening
Zhongshan Ophthalmic Center, Sun Yat-sen University（中山大学中山眼科中心）；Monash eResearch Centre, Monash University（蒙纳士大学电子研究中心）；Centre for Eye Research, Melbourne University（墨尔本大学眼科研究中心）
Prompt-Guided Interactive Segmentation of Interstitial Lung Disease in Thoracic CT
University of Bern（伯尔尼大学）；Graduate School for Cellular and Biomedical Sciences（细胞与生物医学科学研究生院）；Bern University Hospital（伯尔尼大学医院）；Lucern Cantonal Hospital（卢塞恩州立医院）；Lausanne University Hospital (CHUV) and University of Lausanne（洛桑大学医院（CHUV）与洛桑大学）
Towards long-tailed, multi-label disease classification from chest X-ray: Overview of the CXR-LT challenge
The University of Texas at Austin（德克萨斯大学奥斯汀分校）；Weill Cornell Medicine（威尔康奈尔医学院）；Carnegie Mellon University（卡内基梅隆大学）；University of Science, VNU-HCM（越南胡志明市国家大学科学大学）；KT Research & Development Center, KT Corporation（KT公司研发中心）；Ajou University（亚洲大学）
A Structure-Aware Relation Network for Thoracic Diseases Detection and Segmentation
Deepwise AI Lab（深睿医疗人工智能实验室）；School of Electronics Engineering and Computer Science, University of Peking（北京大学电子工程与计算机科学学院）；School of Automation, University of Northwestern Polytechnical（西北工业大学自动化学院）；Department of Computer Science, University of Hong Kong（香港大学计算机科学系）
EXACT: an explainable anomaly-aware vision foundation model for analysis of 3D chest CT
School of Biomedical Engineering, Tsinghua University（清华大学生物医学工程学院）；Department of Radiology, Mianyang Central Hospital（绵阳市中心医院放射科）；Center for Biomedical Imaging Research, Tsinghua University（清华大学生物医学成像研究中心）
An Accurate and Explainable Deep Learning System Improves Interobserver Agreement in the Interpretation of Chest Radiograph
VinUniversity（文大大学）；VinBigData JSC（Vin大数据公司）；Vingroup Big Data Institute（Vingroup大数据研究院）；Hanoi Medical University Hospital（河内医科大学附属医院）；Hospital 108（108医院）
QMix: Quality-aware Learning with Mixed Noise for Robust Retinal Disease Diagnosis
Hong Kong University of Science and Technology（香港科技大学）；Fudan University（复旦大学）；HKUST Shenzhen-Hong Kong Collaborative Innovation Research Institute（香港科技大学深港协同创新研究院）
Delving into Masked Autoencoders for Multi-Label Thorax Disease Classification
Johns Hopkins University（约翰斯·霍普金斯大学）
视觉Transformer（ViT）由于其在许多视觉任务中的巨大可扩展性、计算效率和令人信服的性能而成为最流行的神经架构之一。然而，ViT在医疗任务上表现出不如卷积神经网络（CNN）的性能，这是由于其数据饥渴的性质和缺乏注释的医疗数据。在本文中，我们使用掩码自编码器（MAE）对266，340张胸部X光片进行了预训练ViT，MAE从每张图像的一小部分中重建丢失的像素。为了进行比较，CNN也使用先进的自监督方法（例如，MoCo v2）。结果表明，我们的预训练ViT在多标签胸部疾病分类方面表现优于（有时更好）最先进的CNN（DenseNet-121）。这一表现归功于从我们的经验研究中提取的用于预训练和微调ViT的强大配方。预训练配方表明，与自然成像相比，医学重建需要更小比例的图像（10% vs. 25%）和更适度的随机调整大小的裁剪范围（0.5~1.0 vs. 0.2~1.0）。此外，我们注意到，只要可能，域内迁移学习是优选的。微调配方公开了逐层LR衰减、RandAug幅度和DropPath速率是要考虑的重要因素。我们希望这项研究可以指导未来的研究应用Transformers到更大的各种医学成像任务。
Weighing Features of Lung and Heart Regions for Thoracic Disease Classification
Harbin Institute of Technology（哈尔滨工业大学）；Southern University of Science and Technology（南方科技大学）；CVTE Research（视源研究院）；Cixi Institute of Biomedical Engineering, Chinese Academy of Sciences（中国科学院慈溪生物医学工程研究所）；University of Hong Kong（香港大学）；Guangdong Armed Police Hospital（广东武警医院）
Deep Reinforcement Learning Framework for Thoracic Diseases Classification via Prior Knowledge Guidance
Tianjin University（天津大学）；Tianjin Medical University General Hospital（天津医科大学总医院）；Chest Hospital, Tianjin University（天津大学胸科医院）；Tianjin Medical University（天津医科大学）；Tianjin Institute of Anesthesiology（天津麻醉学研究所）
Explainable Cross-Disease Reasoning for Cardiovascular Risk Assessment from Low-Dose Computed Tomography
Department of Computer Science, Emory University（埃默里大学计算机科学系）；Department of Computer Science, Johns Hopkins University（约翰霍普金斯大学计算机科学系）；Department of Radiation Oncology（放射肿瘤学部）；Winship Cancer Institute, Emory University（埃默里大学Winship癌症研究所）
Weakly Supervised Anomaly Detection for Chest X-Ray Image
School of Artificial Intelligence, Beijing University of Posts and Telecommunications（北京邮电大学人工智能学院）；College of Information and Engineering, Capital Normal University（首都师范大学信息与工程学院）；Peking University Third Hospital（北京大学第三医院）
