[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-ieee8023--covid-chestxray-dataset":3,"tool-ieee8023--covid-chestxray-dataset":64},[4,17,27,35,43,56],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":16},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,3,"2026-04-05T11:01:52",[13,14,15],"开发框架","图像","Agent","ready",{"id":18,"name":19,"github_repo":20,"description_zh":21,"stars":22,"difficulty_score":23,"last_commit_at":24,"category_tags":25,"status":16},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",138956,2,"2026-04-05T11:33:21",[13,15,26],"语言模型",{"id":28,"name":29,"github_repo":30,"description_zh":31,"stars":32,"difficulty_score":23,"last_commit_at":33,"category_tags":34,"status":16},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",107662,"2026-04-03T11:11:01",[13,14,15],{"id":36,"name":37,"github_repo":38,"description_zh":39,"stars":40,"difficulty_score":23,"last_commit_at":41,"category_tags":42,"status":16},3704,"NextChat","ChatGPTNextWeb\u002FNextChat","NextChat 是一款轻量且极速的 AI 助手，旨在为用户提供流畅、跨平台的大模型交互体验。它完美解决了用户在多设备间切换时难以保持对话连续性，以及面对众多 AI 模型不知如何统一管理的痛点。无论是日常办公、学习辅助还是创意激发，NextChat 都能让用户随时随地通过网页、iOS、Android、Windows、MacOS 或 Linux 端无缝接入智能服务。\n\n这款工具非常适合普通用户、学生、职场人士以及需要私有化部署的企业团队使用。对于开发者而言，它也提供了便捷的自托管方案，支持一键部署到 Vercel 或 Zeabur 等平台。\n\nNextChat 的核心亮点在于其广泛的模型兼容性，原生支持 Claude、DeepSeek、GPT-4 及 Gemini Pro 等主流大模型，让用户在一个界面即可自由切换不同 AI 能力。此外，它还率先支持 MCP（Model Context Protocol）协议，增强了上下文处理能力。针对企业用户，NextChat 提供专业版解决方案，具备品牌定制、细粒度权限控制、内部知识库整合及安全审计等功能，满足公司对数据隐私和个性化管理的高标准要求。",87618,"2026-04-05T07:20:52",[13,26],{"id":44,"name":45,"github_repo":46,"description_zh":47,"stars":48,"difficulty_score":23,"last_commit_at":49,"category_tags":50,"status":16},2268,"ML-For-Beginners","microsoft\u002FML-For-Beginners","ML-For-Beginners 是由微软推出的一套系统化机器学习入门课程，旨在帮助零基础用户轻松掌握经典机器学习知识。这套课程将学习路径规划为 12 周，包含 26 节精炼课程和 52 道配套测验，内容涵盖从基础概念到实际应用的完整流程，有效解决了初学者面对庞大知识体系时无从下手、缺乏结构化指导的痛点。\n\n无论是希望转型的开发者、需要补充算法背景的研究人员，还是对人工智能充满好奇的普通爱好者，都能从中受益。课程不仅提供了清晰的理论讲解，还强调动手实践，让用户在循序渐进中建立扎实的技能基础。其独特的亮点在于强大的多语言支持，通过自动化机制提供了包括简体中文在内的 50 多种语言版本，极大地降低了全球不同背景用户的学习门槛。此外，项目采用开源协作模式，社区活跃且内容持续更新，确保学习者能获取前沿且准确的技术资讯。如果你正寻找一条清晰、友好且专业的机器学习入门之路，ML-For-Beginners 将是理想的起点。",84991,"2026-04-05T10:45:23",[14,51,52,53,15,54,26,13,55],"数据工具","视频","插件","其他","音频",{"id":57,"name":58,"github_repo":59,"description_zh":60,"stars":61,"difficulty_score":10,"last_commit_at":62,"category_tags":63,"status":16},3128,"ragflow","infiniflow\u002Fragflow","RAGFlow 是一款领先的开源检索增强生成（RAG）引擎，旨在为大语言模型构建更精准、可靠的上下文层。它巧妙地将前沿的 RAG 技术与智能体（Agent）能力相结合，不仅支持从各类文档中高效提取知识，还能让模型基于这些知识进行逻辑推理和任务执行。\n\n在大模型应用中，幻觉问题和知识滞后是常见痛点。RAGFlow 通过深度解析复杂文档结构（如表格、图表及混合排版），显著提升了信息检索的准确度，从而有效减少模型“胡编乱造”的现象，确保回答既有据可依又具备时效性。其内置的智能体机制更进一步，使系统不仅能回答问题，还能自主规划步骤解决复杂问题。\n\n这款工具特别适合开发者、企业技术团队以及 AI 研究人员使用。无论是希望快速搭建私有知识库问答系统，还是致力于探索大模型在垂直领域落地的创新者，都能从中受益。RAGFlow 提供了可视化的工作流编排界面和灵活的 API 接口，既降低了非算法背景用户的上手门槛，也满足了专业开发者对系统深度定制的需求。作为基于 Apache 2.0 协议开源的项目，它正成为连接通用大模型与行业专有知识之间的重要桥梁。",77062,"2026-04-04T04:44:48",[15,14,13,26,54],{"id":65,"github_repo":66,"name":67,"description_en":68,"description_zh":69,"ai_summary_zh":69,"readme_en":70,"readme_zh":71,"quickstart_zh":72,"use_case_zh":73,"hero_image_url":74,"owner_login":75,"owner_name":76,"owner_avatar_url":77,"owner_bio":78,"owner_company":79,"owner_location":79,"owner_email":80,"owner_twitter":81,"owner_website":82,"owner_url":83,"languages":84,"stars":97,"forks":98,"last_commit_at":99,"license":79,"difficulty_score":100,"env_os":101,"env_gpu":101,"env_ram":101,"env_deps":102,"category_tags":105,"github_topics":106,"view_count":10,"oss_zip_url":79,"oss_zip_packed_at":79,"status":16,"created_at":113,"updated_at":114,"faqs":115,"releases":145},1007,"ieee8023\u002Fcovid-chestxray-dataset","covid-chestxray-dataset","We are building an open database of COVID-19 cases with chest X-ray or CT images.","covid-chestxray-dataset 是一个由蒙特利尔大学支持的COVID-19胸部影像开放数据库项目，专门收集并整理新冠肺炎及其他病毒性、细菌性肺炎（如SARS、MERS、ARDS等）患者的胸部X光和CT图像。项目已获得大学伦理委员会正式批准，数据来源于公开渠道及医疗机构授权，确保合规性。\n\n这个数据集解决了医学AI研究中最关键的\"数据获取难\"问题，为研究人员提供了319例COVID-19阳性PA\u002FAP视图影像和132例仰卧位视图影像，并配有详细的元数据。除了基础图像，还包含肺部边界框标注、肺炎严重程度评分、肺部分割掩码等多层次标注信息，支持从分类到分割等多种研究任务。\n\n需要特别注意的是，项目方明确提醒：这不是Kaggle竞赛数据集，研究者不应在没有临床研究验证的情况下声称模型诊断性能，避免误导性结论。\n\ncovid-chestxray-dataset 主要适合医学影像AI研究者、流行病学家和算法开发者使用，帮助他们快速获取合规数据，加速COVID-19辅助诊断、病情评估等工具的研发。数据集提供PyTorch加载器示例，技术接入门槛低，是开展相关研究的理想起点。","\n#### 🛑 Note: please do not claim diagnostic performance of a model without a clinical study! This is not a kaggle competition dataset. Please read this paper about evaluation issues: [https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.12823](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.12823) and [https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.05405](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.05405)\n\n\n## COVID-19 image data collection ([🎬 video about the project](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=ineWmqfelEQ))\n\nProject Summary: To build a public open dataset of chest X-ray and CT images of patients which are positive or suspected of COVID-19 or other viral and bacterial pneumonias ([MERS](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FMiddle_East_respiratory_syndrome), [SARS](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FSevere_acute_respiratory_syndrome), and [ARDS](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FAcute_respiratory_distress_syndrome).). Data will be collected from public sources as well as through indirect collection from hospitals and physicians. All images and data will be released publicly in this GitHub repo. \n\nThis project is approved by the University of Montreal's Ethics Committee #CERSES-20-058-D\n\n## View current [images](images) and [metadata](metadata.csv) and [a dataloader example](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1A-gIZ6Xp-eh2b4CGS6BHH7-OgZtyjeP2)\n\nThe labels are arranged in a hierarchy:\n\n\u003Cimg width=300 src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fieee8023_covid-chestxray-dataset_readme_eb68025a2a85.jpg\"\u002F>\n\n\nCurrent stats of PA, AP, and AP Supine views. Labels 0=No or 1=Yes. Data loader is [here](https:\u002F\u002Fgithub.com\u002Fmlmed\u002Ftorchxrayvision\u002Fblob\u002Fmaster\u002Ftorchxrayvision\u002Fdatasets.py#L867)\n``` \nCOVID19_Dataset num_samples=481 views=['PA', 'AP']\n{'ARDS': {0.0: 465, 1.0: 16},\n 'Bacterial': {0.0: 445, 1.0: 36},\n 'COVID-19': {0.0: 162, 1.0: 319},\n 'Chlamydophila': {0.0: 480, 1.0: 1},\n 'E.Coli': {0.0: 481},\n 'Fungal': {0.0: 459, 1.0: 22},\n 'Influenza': {0.0: 478, 1.0: 3},\n 'Klebsiella': {0.0: 474, 1.0: 7},\n 'Legionella': {0.0: 474, 1.0: 7},\n 'Lipoid': {0.0: 473, 1.0: 8},\n 'MERS': {0.0: 481},\n 'Mycoplasma': {0.0: 476, 1.0: 5},\n 'No Finding': {0.0: 467, 1.0: 14},\n 'Pneumocystis': {0.0: 459, 1.0: 22},\n 'Pneumonia': {0.0: 36, 1.0: 445},\n 'SARS': {0.0: 465, 1.0: 16},\n 'Streptococcus': {0.0: 467, 1.0: 14},\n 'Varicella': {0.0: 476, 1.0: 5},\n 'Viral': {0.0: 138, 1.0: 343}}\n\nCOVID19_Dataset num_samples=173 views=['AP Supine']\n{'ARDS': {0.0: 170, 1.0: 3},\n 'Bacterial': {0.0: 169, 1.0: 4},\n 'COVID-19': {0.0: 41, 1.0: 132},\n 'Chlamydophila': {0.0: 173},\n 'E.Coli': {0.0: 169, 1.0: 4},\n 'Fungal': {0.0: 171, 1.0: 2},\n 'Influenza': {0.0: 173},\n 'Klebsiella': {0.0: 173},\n 'Legionella': {0.0: 173},\n 'Lipoid': {0.0: 173},\n 'MERS': {0.0: 173},\n 'Mycoplasma': {0.0: 173},\n 'No Finding': {0.0: 170, 1.0: 3},\n 'Pneumocystis': {0.0: 171, 1.0: 2},\n 'Pneumonia': {0.0: 26, 1.0: 147},\n 'SARS': {0.0: 173},\n 'Streptococcus': {0.0: 173},\n 'Varicella': {0.0: 173},\n 'Viral': {0.0: 41, 1.0: 132}}\n\n ```\n \n## Annotations\n\n[Lung Bounding Boxes](https:\u002F\u002Fgithub.com\u002FGeneralBlockchain\u002Fcovid-19-chest-xray-lung-bounding-boxes-dataset) and [Chest X-ray Segmentation](https:\u002F\u002Fgithub.com\u002FGeneralBlockchain\u002Fcovid-19-chest-xray-segmentations-dataset) (license: CC BY 4.0) contributed by [General Blockchain, Inc.](https:\u002F\u002Fgithub.com\u002FGeneralBlockchain)\n\n[Pneumonia severity scores for 94 images](annotations\u002Fcovid-severity-scores.csv) (license: CC BY-SA) from the paper [Predicting COVID-19 Pneumonia Severity on Chest X-ray with Deep Learning](http:\u002F\u002Farxiv.org\u002Fabs\u002F2005.11856)\n\n[Generated Lung Segmentations](annotations\u002FlungVAE-masks) (license: CC BY-SA) from the paper [Lung Segmentation from Chest X-rays using Variational Data Imputation](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.10052)\n\n[Brixia score for 192 images](https:\u002F\u002Fgithub.com\u002FBrixIA\u002FBrixia-score-COVID-19) (license: CC BY-NC-SA) from the paper [End-to-end learning for semiquantitative rating of COVID-19 severity on Chest X-rays](https:\u002F\u002Farxiv.org\u002Fabs\u002F2006.04603)\n\n[Lung and other segmentations for 517 images](https:\u002F\u002Fgithub.com\u002Fv7labs\u002Fcovid-19-xray-dataset\u002Ftree\u002Fmaster\u002Fannotations) (license: CC BY) in COCO and raster formats by [v7labs](https:\u002F\u002Fgithub.com\u002Fv7labs\u002Fcovid-19-xray-dataset)\n\n## Contribute\n\n - Submit data directly to the project. View our [research protocol](https:\u002F\u002Fdocs.google.com\u002Fdocument\u002Fd\u002F14b7cou98YhYcJ2jwOKznChtn5y6-mi9bgjeFv2DxOt0\u002Fedit). Contact us to start the process.\n - We can extract images from publications. Help identify publications which are not already included using a GitHub issue (DOIs we have are listed in the metadata file). There is a searchable database of COVID-19 papers [here](https:\u002F\u002Fwww.who.int\u002Femergencies\u002Fdiseases\u002Fnovel-coronavirus-2019\u002Fglobal-research-on-novel-coronavirus-2019-ncov), and a non-searchable one (requires download) [here](https:\u002F\u002Fpages.semanticscholar.org\u002Fcoronavirus-research).\n \n - Submit data to these sites (we can scrape the data from them):\n    - https:\u002F\u002Fradiopaedia.org\u002F (license CC BY-NC-SA)\n    - https:\u002F\u002Fwww.sirm.org\u002Fcategory\u002Fsenza-categoria\u002Fcovid-19\u002F \n    - https:\u002F\u002Fwww.eurorad.org\u002F (license CC BY-NC-SA)\n    - https:\u002F\u002Fcoronacases.org\u002F (preferred for CT scans, license Apache 2.0)\n \n - Provide bounding box\u002Fmasks for the detection of problematic regions in images already collected.\n\n - See [SCHEMA.md](SCHEMA.md) for more information on the metadata schema.\n\n*Formats:* For chest X-ray dcm, jpg, or png are preferred. For CT nifti (in gzip format) is preferred but also dcms. Please contact with any questions.\n\n## Background \n\nIn the context of a COVID-19 pandemic, we want to improve prognostic predictions to triage and manage patient care. Data is the first step to developing any diagnostic\u002Fprognostic tool. While there exist large public datasets of more typical chest X-rays from the NIH [Wang 2017], Spain [Bustos 2019], Stanford [Irvin 2019], MIT [Johnson 2019] and Indiana University [Demner-Fushman 2016], there is no collection of COVID-19 chest X-rays or CT scans designed to be used for computational analysis.\n\nThe 2019 novel coronavirus (COVID-19) presents several unique features [Fang, 2020](https:\u002F\u002Fpubs.rsna.org\u002Fdoi\u002F10.1148\u002Fradiol.2020200432) and [Ai 2020](https:\u002F\u002Fpubs.rsna.org\u002Fdoi\u002F10.1148\u002Fradiol.2020200642). While the diagnosis is confirmed using polymerase chain reaction (PCR), infected patients with pneumonia may present on chest X-ray and computed tomography (CT) images with a pattern that is only moderately characteristic for the human eye [Ng, 2020](https:\u002F\u002Fpubs.rsna.org\u002Fdoi\u002F10.1148\u002Fryct.2020200034). In late January, a Chinese team published a paper detailing the clinical and paraclinical features of COVID-19. They reported that patients present abnormalities in chest CT images with most having bilateral involvement [Huang 2020](https:\u002F\u002Fwww.thelancet.com\u002Fjournals\u002Flancet\u002Farticle\u002FPIIS0140-6736(20)30183-5\u002Ffulltext). Bilateral multiple lobular and subsegmental areas of consolidation constitute the typical findings in chest CT images of intensive care unit (ICU) patients on admission [Huang 2020](https:\u002F\u002Fwww.thelancet.com\u002Fjournals\u002Flancet\u002Farticle\u002FPIIS0140-6736(20)30183-5\u002Ffulltext). In comparison, non-ICU patients show bilateral ground-glass opacity and subsegmental areas of consolidation in their chest CT images [Huang 2020](https:\u002F\u002Fwww.thelancet.com\u002Fjournals\u002Flancet\u002Farticle\u002FPIIS0140-6736(20)30183-5\u002Ffulltext). In these patients, later chest CT images display bilateral ground-glass opacity with resolved consolidation [Huang 2020](https:\u002F\u002Fwww.thelancet.com\u002Fjournals\u002Flancet\u002Farticle\u002FPIIS0140-6736(20)30183-5\u002Ffulltext). \n\n\n## Goal\n\nOur goal is to use these images to develop AI based approaches to predict and understand the infection. Our group will work to release these models using our open source [Chester AI Radiology Assistant platform](https:\u002F\u002Fmlmed.org\u002Ftools\u002Fxray\u002F).\n\nThe tasks are as follows using chest X-ray or CT (preference for X-ray) as input to predict these tasks:\n\n- Healthy vs Pneumonia (prototype already implemented [Chester](https:\u002F\u002Fmlmed.org\u002Ftools\u002Fxray\u002F) with ~74% AUC, validation study [here](https:\u002F\u002Farxiv.org\u002Fabs\u002F2002.02497))\n\n- ~~Bacterial vs Viral vs COVID-19 Pneumonia~~ (not relevant enough for the clinical workflows)\n\n- Prognostic\u002Fseverity predictions (survival, need for intubation, need for supplemental oxygen)\n\n## Expected outcomes\n\nTool impact: This would give physicians an edge and allow them to act with more confidence while they wait for the analysis of a radiologist by having a digital second opinion confirm their assessment of a patient's condition. Also, these tools can provide quantitative scores to consider and use in studies.\n\nData impact: Image data linked with clinically relevant attributes in a public dataset that is designed for ML will enable parallel development of these tools and rapid local validation of models. Furthermore, this data can be used for completely different tasks.\n\n\n## Contact\nPI: [Joseph Paul Cohen. Postdoctoral Fellow, Mila, University of Montreal](https:\u002F\u002Fjosephpcohen.com\u002F) \n\n## Citations\n\nSecond Paper available [here](http:\u002F\u002Farxiv.org\u002Fabs\u002F2006.11988) and [source code for baselines](https:\u002F\u002Fgithub.com\u002Fmlmed\u002Ftorchxrayvision\u002Ftree\u002Fmaster\u002Fscripts\u002Fcovid-baselines)\n\n```\nCOVID-19 Image Data Collection: Prospective Predictions Are the Future\nJoseph Paul Cohen and Paul Morrison and Lan Dao and Karsten Roth and Tim Q Duong and Marzyeh Ghassemi\narXiv:2006.11988, https:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset, 2020\n```\n\n```\n@article{cohen2020covidProspective,\n  title={COVID-19 Image Data Collection: Prospective Predictions Are the Future},\n  author={Joseph Paul Cohen and Paul Morrison and Lan Dao and Karsten Roth and Tim Q Duong and Marzyeh Ghassemi},\n  journal={arXiv 2006.11988},\n  url={https:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset},\n  year={2020}\n}\n```\n\nPaper available [here](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.11597)\n\n```\nCOVID-19 image data collection, arXiv:2003.11597, 2020\nJoseph Paul Cohen and Paul Morrison and Lan Dao\nhttps:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset\n```\n\n```\n@article{cohen2020covid,\n  title={COVID-19 image data collection},\n  author={Joseph Paul Cohen and Paul Morrison and Lan Dao},\n  journal={arXiv 2003.11597},\n  url={https:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset},\n  year={2020}\n}\n```\n\n\u003Cmeta name=\"citation_title\" content=\"COVID-19 image data collection\" \u002F>\n\u003Cmeta name=\"citation_publication_date\" content=\"2020\" \u002F>\n\u003Cmeta name=\"citation_author\" content=\"Joseph Paul Cohen and Paul Morrison and Lan Dao\" \u002F>\n\n## License\n\nEach image has license specified in the metadata.csv file. Including Apache 2.0, CC BY-NC-SA 4.0, CC BY 4.0.\n\nThe metadata.csv, scripts, and other documents are released under a CC BY-NC-SA 4.0 license. Companies are free to perform research. Beyond that contact us.\n","#### 🛑 注意：请不要在没有临床研究的情况下声称模型的诊断性能！这不是一个 Kaggle 竞赛数据集。请阅读这篇关于评估问题的论文：[https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.12823](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.12823) 和 [https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.05405](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.05405)\n\n\n## COVID-19 影像数据收集 ([🎬 项目视频](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=ineWmqfelEQ))\n\n项目摘要：构建一个公开的胸部 X 光（chest X-ray）和 CT 影像数据集，包含 COVID-19 阳性或疑似患者，以及其他病毒性和细菌性肺炎患者（[MERS](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FMiddle_East_respiratory_syndrome)、[SARS](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FSevere_acute_respiratory_syndrome) 和 [ARDS](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FAcute_respiratory_distress_syndrome)）。数据将从公开来源收集，并通过医院和医生间接收集。所有影像和数据都将在此 GitHub 仓库中公开发布。\n\n本项目已获得蒙特利尔大学伦理委员会批准，批准号：#CERSES-20-058-D\n\n## 查看当前的[影像](images)和[元数据](metadata.csv)以及[数据加载器示例](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1A-gIZ6Xp-eh2b4CGS6BHH7-OgZtyjeP2)\n\n标签按层级结构排列：\n\n\u003Cimg width=300 src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fieee8023_covid-chestxray-dataset_readme_eb68025a2a85.jpg\"\u002F>\n\n\n当前 PA、AP 和 AP Supine 视图的统计信息。标签 0=否 或 1=是。数据加载器（Data loader）位于[此处](https:\u002F\u002Fgithub.com\u002Fmlmed\u002Ftorchxrayvision\u002Fblob\u002Fmaster\u002Ftorchxrayvision\u002Fdatasets.py#L867)\n``` \nCOVID19_Dataset num_samples=481 views=['PA', 'AP']\n{'ARDS': {0.0: 465, 1.0: 16},\n 'Bacterial': {0.0: 445, 1.0: 36},\n 'COVID-19': {0.0: 162, 1.0: 319},\n 'Chlamydophila': {0.0: 480, 1.0: 1},\n 'E.Coli': {0.0: 481},\n 'Fungal': {0.0: 459, 1.0: 22},\n 'Influenza': {0.0: 478, 1.0: 3},\n 'Klebsiella': {0.0: 474, 1.0: 7},\n 'Legionella': {0.0: 474, 1.0: 7},\n 'Lipoid': {0.0: 473, 1.0: 8},\n 'MERS': {0.0: 481},\n 'Mycoplasma': {0.0: 476, 1.0: 5},\n 'No Finding': {0.0: 467, 1.0: 14},\n 'Pneumocystis': {0.0: 459, 1.0: 22},\n 'Pneumonia': {0.0: 36, 1.0: 445},\n 'SARS': {0.0: 465, 1.0: 16},\n 'Streptococcus': {0.0: 467, 1.0: 14},\n 'Varicella': {0.0: 476, 1.0: 5},\n 'Viral': {0.0: 138, 1.0: 343}}\n\nCOVID19_Dataset num_samples=173 views=['AP Supine']\n{'ARDS': {0.0: 170, 1.0: 3},\n 'Bacterial': {0.0: 169, 1.0: 4},\n 'COVID-19': {0.0: 41, 1.0: 132},\n 'Chlamydophila': {0.0: 173},\n 'E.Coli': {0.0: 169, 1.0: 4},\n 'Fungal': {0.0: 171, 1.0: 2},\n 'Influenza': {0.0: 173},\n 'Klebsiella': {0.0: 173},\n 'Legionella': {0.0: 173},\n 'Lipoid': {0.0: 173},\n 'MERS': {0.0: 173},\n 'Mycoplasma': {0.0: 173},\n 'No Finding': {0.0: 170, 1.0: 3},\n 'Pneumocystis': {0.0: 171, 1.0: 2},\n 'Pneumonia': {0.0: 26, 1.0: 147},\n 'SARS': {0.0: 173},\n 'Streptococcus': {0.0: 173},\n 'Varicella': {0.0: 173},\n 'Viral': {0.0: 41, 1.0: 132}}\n\n ```\n \n## 标注\n\n[肺部边界框（Lung Bounding Boxes）](https:\u002F\u002Fgithub.com\u002FGeneralBlockchain\u002Fcovid-19-chest-xray-lung-bounding-boxes-dataset)和[胸部 X 光分割（Chest X-ray Segmentation）](https:\u002F\u002Fgithub.com\u002FGeneralBlockchain\u002Fcovid-19-chest-xray-segmentations-dataset)（许可证：CC BY 4.0）由 [General Blockchain, Inc.](https:\u002F\u002Fgithub.com\u002FGeneralBlockchain) 提供\n\n[94 张影像的肺炎严重程度评分](annotations\u002Fcovid-severity-scores.csv)（许可证：CC BY-SA）来自论文[使用深度学习在胸部 X 光上预测 COVID-19 肺炎严重程度](http:\u002F\u002Farxiv.org\u002Fabs\u002F2005.11856)\n\n[生成的肺部分割](annotations\u002FlungVAE-masks)（许可证：CC BY-SA）来自论文[使用变分数据填补从胸部 X 光进行肺部分割](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.10052)\n\n[192 张影像的 Brixia 评分](https:\u002F\u002Fgithub.com\u002FBrixIA\u002FBrixia-score-COVID-19)（许可证：CC BY-NC-SA）来自论文[在胸部 X 光上对 COVID-19 严重程度进行半定量评分的端到端学习](https:\u002F\u002Farxiv.org\u002Fabs\u002F2006.04603)\n\n[517 张影像的肺部及其他分割](https:\u002F\u002Fgithub.com\u002Fv7labs\u002Fcovid-19-xray-dataset\u002Ftree\u002Fmaster\u002Fannotations)（许可证：CC BY），采用 COCO 和光栅格式，由 [v7labs](https:\u002F\u002Fgithub.com\u002Fv7labs\u002Fcovid-19-xray-dataset) 提供\n\n## 贡献\n\n - 直接向项目提交数据。查看我们的[研究方案](https:\u002F\u002Fdocs.google.com\u002Fdocument\u002Fd\u002F14b7cou98YhYcJ2jwOKznChtn5y6-mi9bgjeFv2DxOt0\u002Fedit)。请联系我们以启动流程。\n - 我们可以从出版物中提取影像。请使用 GitHub issue 帮助识别尚未包含的出版物（我们已有的 DOI 列在元数据文件中）。可搜索的 COVID-19 论文数据库位于[此处](https:\u002F\u002Fwww.who.int\u002Femergencies\u002Fdiseases\u002Fnovel-coronavirus-2019\u002Fglobal-research-on-novel-coronavirus-2019-ncov)，不可搜索的（需要下载）位于[此处](https:\u002F\u002Fpages.semanticscholar.org\u002Fcoronavirus-research)。\n \n - 向这些网站提交数据（我们可以从中抓取数据）：\n    - https:\u002F\u002Fradiopaedia.org\u002F (license CC BY-NC-SA)\n    - https:\u002F\u002Fwww.sirm.org\u002Fcategory\u002Fsenza-categoria\u002Fcovid-19\u002F \n    - https:\u002F\u002Fwww.eurorad.org\u002F (license CC BY-NC-SA)\n    - https:\u002F\u002Fcoronacases.org\u002F (preferred for CT scans, license Apache 2.0)\n \n - 为已收集影像中的问题区域检测提供边界框\u002F掩码。\n\n - 有关元数据架构的更多信息，请参见 [SCHEMA.md](SCHEMA.md)。\n\n*格式：* 胸部 X 光优先使用 dcm、jpg 或 png 格式。CT 优先使用 nifti（gzip 格式）格式，但也接受 dcm 格式。如有任何问题，请联系我们。\n\n## 背景\n\n在COVID-19（新型冠状病毒肺炎）大流行的背景下，我们希望改进预后预测，以对患者进行分诊和管理护理。数据是开发任何诊断\u002F预后工具的首要步骤。虽然目前存在来自NIH [Wang 2017]、西班牙 [Bustos 2019]、斯坦福 [Irvin 2019]、MIT [Johnson 2019] 和印第安纳大学 [Demner-Fushman 2016] 的更多典型胸部X光片（chest X-rays）大型公共数据集，但目前尚无专为计算分析而设计的COVID-19胸部X光片或CT扫描（CT scans）数据集。\n\n2019新型冠状病毒（COVID-19）具有几个独特特征 [Fang, 2020](https:\u002F\u002Fpubs.rsna.org\u002Fdoi\u002F10.1148\u002Fradiol.2020200432) 和 [Ai 2020](https:\u002F\u002Fpubs.rsna.org\u002Fdoi\u002F10.1148\u002Fradiol.2020200642)。虽然诊断通过聚合酶链式反应（PCR）确认，但肺炎感染患者在胸部X光片和CT图像上可能呈现出对人眼而言仅具有中等程度特征性的影像模式 [Ng, 2020](https:\u002F\u002Fpubs.rsna.org\u002Fdoi\u002F10.1148\u002Fryct.2020200034)。一月底，一个中国团队发表了一篇详细描述COVID-19临床及辅助临床特征的论文。他们报告患者在胸部CT图像中呈现异常，大多数具有双侧受累 [Huang 2020](https:\u002F\u002Fwww.thelancet.com\u002Fjournals\u002Flancet\u002Farticle\u002FPIIS0140-6736(20)30183-5\u002Ffulltext)。双侧多叶和亚段实变（consolidation）区域构成了重症监护室（ICU）患者入院时胸部CT图像的典型发现 [Huang 2020](https:\u002F\u002Fwww.thelancet.com\u002Fjournals\u002Flancet\u002Farticle\u002FPIIS0140-6736(20)30183-5\u002Ffulltext)。相比之下，非ICU患者在胸部CT图像中显示双侧磨玻璃影（ground-glass opacity）和亚段实变区域 [Huang 2020](https:\u002F\u002Fwww.thelancet.com\u002Fjournals\u002Flancet\u002Farticle\u002FPIIS0140-6736(20)30183-5\u002Ffulltext)。在这些患者中，后续的胸部CT图像显示双侧磨玻璃影伴实变消退 [Huang 2020](https:\u002F\u002Fwww.thelancet.com\u002Fjournals\u002Flancet\u002Farticle\u002FPIIS0140-6736(20)30183-5\u002Ffulltext)。\n\n## 目标\n\n我们的目标是使用这些图像开发基于人工智能（AI）的方法来预测和理解感染。我们的团队将致力于使用我们的开源Chester AI放射学助手平台发布这些模型。\n\n任务如下，使用胸部X光片或CT（优先X光片）作为输入来预测：\n\n- 健康 vs 肺炎（原型已在[Chester](https:\u002F\u002Fmlmed.org\u002Ftools\u002Fxray\u002F)中实现，AUC（Area Under Curve，曲线下面积）约为74%，验证研究见[此处](https:\u002F\u002Farxiv.org\u002Fabs\u002F2002.02497)）\n\n- ~~细菌性 vs 病毒性 vs COVID-19肺炎~~（与临床工作流程相关性不足）\n\n- 预后\u002F严重程度预测（生存率、插管（intubation）需求、补充氧气（supplemental oxygen）需求）\n\n## 预期成果\n\n工具影响：这将为医生提供支持，使他们在等待放射科医生分析的同时，能够更有信心地采取行动，通过数字化第二意见来确认其对患者病情的评估。此外，这些工具可以提供定量评分以供考虑并在研究中使用。\n\n数据影响：图像数据与临床相关属性相关联，并包含在专为机器学习（ML）设计的公共数据集中，将使这些工具的并行开发和模型的快速本地验证成为可能。此外，这些数据可用于完全不同的任务。\n\n## 联系方式\n\n项目负责人（PI）：[Joseph Paul Cohen，蒙特利尔大学Mila博士后研究员](https:\u002F\u002Fjosephpcohen.com\u002F)\n\n## 引用\n\n第二篇论文可[在此](http:\u002F\u002Farxiv.org\u002Fabs\u002F2006.11988)获取，基线源代码见[此处](https:\u002F\u002Fgithub.com\u002Fmlmed\u002Ftorchxrayvision\u002Ftree\u002Fmaster\u002Fscripts\u002Fcovid-baselines)\n\n```\nCOVID-19 Image Data Collection: Prospective Predictions Are the Future\nJoseph Paul Cohen and Paul Morrison and Lan Dao and Karsten Roth and Tim Q Duong and Marzyeh Ghassemi\narXiv:2006.11988, https:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset, 2020\n```\n\n```\n@article{cohen2020covidProspective,\n  title={COVID-19 Image Data Collection: Prospective Predictions Are the Future},\n  author={Joseph Paul Cohen and Paul Morrison and Lan Dao and Karsten Roth and Tim Q Duong and Marzyeh Ghassemi},\n  journal={arXiv 2006.11988},\n  url={https:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset},\n  year={2020}\n}\n```\n\n论文可[在此](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.11597)获取\n\n```\nCOVID-19 image data collection, arXiv:2003.11597, 2020\nJoseph Paul Cohen and Paul Morrison and Lan Dao\nhttps:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset\n```\n\n```\n@article{cohen2020covid,\n  title={COVID-19 image data collection},\n  author={Joseph Paul Cohen and Paul Morrison and Lan Dao},\n  journal={arXiv 2003.11597},\n  url={https:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset},\n  year={2020}\n}\n```\n\n\u003Cmeta name=\"citation_title\" content=\"COVID-19 image data collection\" \u002F>\n\u003Cmeta name=\"citation_publication_date\" content=\"2020\" \u002F>\n\u003Cmeta name=\"citation_author\" content=\"Joseph Paul Cohen and Paul Morrison and Lan Dao\" \u002F>\n\n## 许可证\n\n每张图像的许可证在metadata.csv（元数据文件）中指定。包括Apache 2.0许可证、CC BY-NC-SA 4.0（知识共享署名-非商业性使用-相同方式共享4.0）、CC BY 4.0（知识共享署名4.0）。\n\nmetadata.csv、脚本和其他文档在CC BY-NC-SA 4.0许可证下发布。企业可自由开展研究。超出此范围请与我们联系。","# covid-chestxray-dataset 快速上手指南\n\n> **⚠️ 重要警告**：请勿在没有临床研究的情况下宣称模型的诊断性能！这不是 Kaggle 竞赛数据集。在使用数据进行评估前，请务必阅读相关评估问题论文：[https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.12823](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.12823) 和 [https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.05405](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.05405)。\n\n## 环境准备\n\n*   **操作系统**：Linux \u002F Windows \u002F macOS\n*   **编程语言**：Python 3.6+\n*   **依赖工具**：\n    *   Git（用于克隆仓库）\n    *   PyTorch（推荐使用，配合 `torchxrayvision` 加载器）\n*   **存储空间**：建议预留足够空间存放 X 光及 CT 影像数据（具体大小视下载子集而定）\n\n## 安装步骤\n\n1.  **克隆数据集仓库**\n    使用 Git 将项目代码及元数据下载到本地：\n    ```bash\n    git clone https:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset.git\n    cd covid-chestxray-dataset\n    ```\n\n2.  **安装数据加载依赖（可选）**\n    若需使用官方推荐的数据加载器，请安装 `torchxrayvision` 库：\n    ```bash\n    pip install torchxrayvision\n    ```\n\n3.  **获取影像数据**\n    影像文件位于 `images` 目录中。部分大文件可能未直接包含在 Git 历史中，请检查仓库 releases 或根据 `metadata.csv` 中的来源链接获取原始影像。\n\n## 基本使用\n\n1.  **查看元数据**\n    所有影像的标签和临床信息均记录在 `metadata.csv` 文件中。标签采用层级结构（如 COVID-19, Pneumonia, Viral 等）。\n    ```python\n    import pandas as pd\n    metadata = pd.read_csv(\"metadata.csv\")\n    print(metadata.head())\n    ```\n\n2.  **使用官方数据加载器**\n    项目提供了基于 `torchxrayvision` 的 DataLoader 示例。您可以直接参考 Google Colab 示例运行：\n    [🔗 DataLoader Example on Colab](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1A-gIZ6Xp-eh2b4CGS6BHH7-OgZtyjeP2)\n\n    或在代码中调用（参考 `torchxrayvision` 实现）：\n    ```python\n    import torchxrayvision as xrv\n    # 示例：加载 COVID19 数据集\n    dataset = xrv.datasets.COVID19_Dataset(root=\".\u002Fimages\", transform=None)\n    ```\n\n3.  **数据标注与分割**\n    如需使用肺部边界框或分割掩码，请参考 `annotations` 目录下的相关数据集链接（如 Lung Bounding Boxes, Chest X-ray Segmentation 等），注意各子集遵循不同的许可证（CC BY 4.0, CC BY-SA 等）。\n\n---\n**许可证说明**：元数据及脚本遵循 CC BY-NC-SA 4.0 许可证。每张影像的具体许可证请在 `metadata.csv` 中查询（包含 Apache 2.0, CC BY-NC-SA 4.0, CC BY 4.0 等）。商业用途请联系项目方。","某高校医学 AI 实验室计划开发新冠肺炎 X 光辅助筛查模型，急需高质量影像数据训练算法。\n\n### 没有 covid-chestxray-dataset 时\n- 数据收集极其困难：单个医院确诊病例有限，跨院协调患者隐私数据往往耗时数月。\n- 标注标准混乱不一：不同医生对“病毒性肺炎”与“新冠”标注定义不同，导致模型难以收敛。\n- 缺乏关键辅助信息：原始数据没有肺部分割掩码或严重程度评分，需人工重新标注，成本极高。\n- 伦理合规风险高：自行收集患者临床数据需通过复杂伦理审查，团队面临巨大法律压力。\n\n### 使用 covid-chestxray-dataset 后\n- 数据获取即时可用：直接下载包含数百例确诊新冠及其他肺炎对比病例的公开影像集。\n- 标签层级清晰明确：利用提供的层级标签（如 COVID-19、Viral、Pneumonia），实现多分类精准训练。\n- 注解资源直接集成：使用现有的肺部分割掩码和 Brixia 严重程度评分，免去重复标注工作。\n- 合规性得到保障：数据集已通过伦理委员会批准，显著降低法律与伦理风险。\n\ncovid-chestxray-dataset 通过提供标准化、合规的公开影像数据，大幅降低了医疗 AI 研发的门槛与周期。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Fieee8023_covid-chestxray-dataset_287924e7.png","ieee8023","Joseph Paul Cohen PhD","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Fieee8023_09c813ea.png","Amazon, Butterfly Network, Stanford AIMI, Mila,\r\nDirector: Institute for Reproducible Research, MLMed.org, AcademicTorrents.com, ShortScience.org",null,"joseph@josephpcohen.com","josephpaulcohen","https:\u002F\u002Fjosephpcohen.com","https:\u002F\u002Fgithub.com\u002Fieee8023",[85,89,93],{"name":86,"color":87,"percentage":88},"Jupyter Notebook","#DA5B0B",82.8,{"name":90,"color":91,"percentage":92},"Python","#3572A5",16.3,{"name":94,"color":95,"percentage":96},"JavaScript","#f1e05a",0.9,3056,1280,"2026-04-04T17:11:56",1,"未说明",{"notes":103,"python":101,"dependencies":104},"本项目主要为数据集而非独立软件工具。图像格式支持 DICOM、JPG、PNG（X 光）及 Nifti、DICOM（CT）。元数据和脚本采用 CC BY-NC-SA 4.0 许可，图像许可各异（含 Apache 2.0 等）。警告未经临床研究不得声称模型诊断性能。数据加载可参考 torchxrayvision 库。",[101],[14,13,51],[107,108,109,110,111,112],"covid-19","deep-learning","computer-vision","dataset","xray","computed-tomography","2026-03-27T02:49:30.150509","2026-04-06T07:23:05.861073",[116,121,126,131,135,140],{"id":117,"question_zh":118,"answer_zh":119,"source_url":120},4483,"如何在这个数据集中找到 CT 图像？","可以在 `metadata.csv` 文件中筛选 `modality = CT` 的行，然后使用 `filename` 列对应的文件路径。不过需要注意，数据集中的 CT 图像数量相对 X 光片较少。","https:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset\u002Fissues\u002F35",{"id":122,"question_zh":123,"answer_zh":124,"source_url":125},4484,"为什么 metadata 中部分病例的 offset 字段值为负数？","这是因为 Hanover 数据进行了更新，导致许多 offset 值发生了变化。维护者正在处理并更新这些图像的 offset 值，通常 offset 应为正数。如果发现此类情况，通常是数据更新过程中的临时问题。","https:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset\u002Fissues\u002F136",{"id":127,"question_zh":128,"answer_zh":129,"source_url":130},4485,"为什么模型训练准确率异常高，但解释性分析（如 Grad-CAM）显示特征不符？","可能是因为使用了儿童胸片（如 Kaggle 肺炎数据集）作为健康对照。儿童胸部形状与成人\u002F老人不同，模型可能学到了年龄或胸部大小特征而非病理特征（如斑片状混浊）。建议增加 COVID 肺炎病例数，并从不同数据集收集成人正常病例。","https:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset\u002Fissues\u002F37",{"id":132,"question_zh":133,"answer_zh":134,"source_url":130},4486,"对于健康对照组的训练样本，有什么数据集推荐？","避免使用主要包含儿童胸片的数据集。建议使用主要是成人的肺炎数据（例如 RSNA Kaggle 数据集），或者从不同数据集收集正常病例，以避免模型学习到年龄偏差。放射学家也指出儿童与成人\u002F老人胸部形状差异会导致模型产生虚假的高准确率。",{"id":136,"question_zh":137,"answer_zh":138,"source_url":139},4487,"为什么某些已发表的论文没有被包含在 metadata.csv 中？","项目目前主要专注于添加 X 光片（X-rays）。如果论文仅包含 CT 图像，可能不会被添加。维护者会通过检查 DOI 是否已存在于文件来确认是否重复。如果您发现缺失的 X 光片论文，可以提交 Issue 提供 DOI。","https:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset\u002Fissues\u002F70",{"id":141,"question_zh":142,"answer_zh":143,"source_url":144},4488,"该项目主要接受哪种类型的医学图像数据？","项目主要使用胸部 X 光图像或 CT 体积（CT volumes）。CT 切片（slices）对于构建工具用处不大，通常不被优先处理，但如果有人提取了切片并提交 PR，维护者也会考虑合并。","https:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset\u002Fissues\u002F59",[146,151,155,159,164],{"id":147,"version":148,"summary_zh":149,"released_at":150},103933,"0.41","New dataset paper available [here](http:\u002F\u002Farxiv.org\u002Fabs\u002F2006.11988) and [source code for baselines](https:\u002F\u002Fgithub.com\u002Fmlmed\u002Fcovid-baselines)\r\n\r\n```\r\nCOVID-19 Image Data Collection: Prospective Predictions Are the Future, arXiv:2006.11988, 2020\r\nJoseph Paul Cohen and Paul Morrison and Lan Dao and Karsten Roth and Tim Q Duong and Marzyeh Ghassemi\r\nhttps:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset\r\n```\r\n\r\n```\r\n@article{cohen2020covidProspective,\r\n  title={COVID-19 Image Data Collection: Prospective Predictions Are the Future},\r\n  author={Joseph Paul Cohen and Paul Morrison and Lan Dao and Karsten Roth and Tim Q Duong and Marzyeh Ghassemi},\r\n  journal={arXiv 2006.11988},\r\n  url={https:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset},\r\n  year={2020}\r\n}\r\n```","2020-10-01T17:28:44",{"id":152,"version":153,"summary_zh":149,"released_at":154},103934,"0.4","2020-10-01T00:48:49",{"id":156,"version":157,"summary_zh":149,"released_at":158},103935,"0.3","2020-09-24T20:59:27",{"id":160,"version":161,"summary_zh":162,"released_at":163},103936,"0.2","New dataset paper available [here](http:\u002F\u002Farxiv.org\u002Fabs\u002F2006.11988) and [source code for baselines](https:\u002F\u002Fgithub.com\u002Fmlmed\u002Ftorchxrayvision\u002Ftree\u002Fmaster\u002Fscripts\u002Fcovid-baselines)\r\n\r\n```\r\nCOVID-19 Image Data Collection: Prospective Predictions Are the Future, arXiv:2006.11988, 2020\r\nJoseph Paul Cohen and Paul Morrison and Lan Dao and Karsten Roth and Tim Q Duong and Marzyeh Ghassemi\r\nhttps:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset\r\n```\r\n\r\n```\r\n@article{cohen2020covidProspective,\r\n  title={COVID-19 Image Data Collection: Prospective Predictions Are the Future},\r\n  author={Joseph Paul Cohen and Paul Morrison and Lan Dao and Karsten Roth and Tim Q Duong and Marzyeh Ghassemi},\r\n  journal={arXiv 2006.11988},\r\n  url={https:\u002F\u002Fgithub.com\u002Fieee8023\u002Fcovid-chestxray-dataset},\r\n  year={2020}\r\n}\r\n```","2020-06-23T01:20:16",{"id":165,"version":166,"summary_zh":79,"released_at":167},103937,"v0.1","2020-03-28T04:48:49"]